ML15329A273: Difference between revisions

From kanterella
Jump to navigation Jump to search
(Created page by program invented by StriderTol)
 
(StriderTol Bot change)
 
Line 16: Line 16:


=Text=
=Text=
{{#Wiki_filter:RIV000140 Submitted: November 9, 2012 Proceedings of the 12th International Conference on Environmental Degradation of Materials in Nuclear Power System - Water Reactors -
{{#Wiki_filter:INFLUENCE OF A CYCLIC LOADING ON THE INITIATION AND PROPAGATION OF PWSCC IN WELD METAL 182 François Vaillant1, Jean-Marie Boursier1, Thierry Couvant1, Claude Amzallag2, Jacques Champredonde3 1EDF/R&D/MMC, Les Renardires Research Center, 77818 Moret sur Loing, France 2EDF/SEPTEN, 12-14 avenue Dutrievoz, 69628 Villeurbanne Cedex, France 3EDF/CEIDRE, 2 rue Ampre, 93206 - Saint-Denis Cedex 01 Keywords: Stress Corrosion Cracking, weld metal 182, primary water, cyclic loading Abstract PWSCC (Primary Water Stress Corrosion Cracking) of nickel based weld metal 182 is now a major issue in Pressurised Water Reactors, since significant cracking was observed in several welds of the primary circuit in some countries. Many questions arise about the stress (or strain) threshold for initiation and on the crack growth rates of SCC, with significant differences depending on the orientation of the loading axis with regard to the dendrites.
Edited by T.R. Allen, P.J. King, and L. Nelson TMS (The Minerals, Metals & Materials Society), 2005 INFLUENCE OF A CYCLIC LOADING ON THE INITIATION AND PROPAGATION OF PWSCC IN WELD METAL 182 François Vaillant1, Jean-Marie Boursier1, Thierry Couvant1, Claude Amzallag2, Jacques Champredonde3 1
The stress threshold for initiation was determined at the level of 350 MPa with constant load tests on as-welded material. This threshold was not significantly modified by a ripple loading at 360°C. The strain threshold for initiation, determined by interrupted SSRTs for various elongations, was found to range between 1.25 and 1.5%.
EDF/R&D/MMC, Les Renardires Research Center, 77818 Moret sur Loing, France 2
Preliminary data of crack growth rates (CGRs) at 325°C provided a dependence on (K-9)0.4 for the average values of the CGRmax.
EDF/SEPTEN, 12-14 avenue Dutrievoz, 69628 Villeurbanne Cedex, France 3
The strongly oriented microstructure led to faster (x 2 to 3) CGRs in the direction of the dendrites. The influence of a trapezoidal loading (R = 0.7, 2.8 10-4 Hz) on CGRs depended on the orientation : no significant modification was found with respect to a pure static loading in the direction TL perpendicular to the dendrites, but some significant increase (x 5 to 7) could be observed in the direction of the dendrites TS. A stress relief treatment (610°C) could lower the CGRs by a factor 2 to 3.5 with respect to the as-welded condition. The results are discussed with regard to available data from literature and compared to the background on alloy 600 : the chemical composition of the weld metal could lead to some variability (factor 3) in the CGRs (4 welds with significantly different chemical compositions).
EDF/CEIDRE, 2 rue Ampre, 93206 - Saint-Denis Cedex 01 Keywords: Stress Corrosion Cracking, weld metal 182, primary water, cyclic loading Abstract                                                                   Introduction PWSCC (Primary Water Stress Corrosion Cracking) of nickel                    PWSCC (Primary Water Stress Corrosion Cracking) of nickel based weld metal 182 is now a major issue in Pressurised Water               based weld metal 182 is now a major issue in Pressurised Water Reactors, since significant cracking was observed in several welds          Reactors, since significant cracking was observed in several welds of the primary circuit in some countries. Many questions arise               of the primary circuit in several countries. Despite important R&D about the stress (or strain) threshold for initiation and on the crack       efforts to determine the influence of the main parameters which growth rates of SCC, with significant differences depending on               could affect PWSCC of alloy 182, many questions arose about the the orientation of the loading axis with regard to the dendrites.           stress- (and strain-) threshold for initiation and on the crack The stress threshold for initiation was determined at the level of           growth rates of SCC, with significant differences depending on 350 MPa with constant load tests on as-welded material. This                 the orientation of the loading axis with regard to the dendrites.
Tests are scheduled to account for the influences of temperature and composition of the weld metal on the stress threshold and on the crack velocities.
threshold was not significantly modified by a ripple loading at             Moreover, because crack growth rates are sometimes extensively 360°C. The strain threshold for initiation, determined by                   determined using periodic unloading reloading procedures, a interrupted SSRTs for various elongations, was found to range               comparison was proposed for CGRs obtained with this kind of between 1.25 and 1.5%.                                                       loading with respect CGRs values deduced from a pure static Preliminary data of crack growth rates (CGRs) at 325°C provided             loading, depending on the orientation of specimens.
Introduction PWSCC (Primary Water Stress Corrosion Cracking) of nickel based weld metal 182 is now a major issue in Pressurised Water Reactors, since significant cracking was observed in several welds of the primary circuit in several countries. Despite important R&D efforts to determine the influence of the main parameters which could affect PWSCC of alloy 182, many questions arose about the stress- (and strain-) threshold for initiation and on the crack growth rates of SCC, with significant differences depending on the orientation of the loading axis with regard to the dendrites.
a dependence on (K-9)0.4 for the average values of the CGRmax.               This contributing paper presents the status of EDF R&D program The strongly oriented microstructure led to faster (x 2 to 3) CGRs           on the initiation and propagation of SCC in weld metal 182, with in the direction of the dendrites. The influence of a trapezoidal           a particular emphasize on the influence of the loading mode.
Moreover, because crack growth rates are sometimes extensively determined using periodic unloading reloading procedures, a comparison was proposed for CGRs obtained with this kind of loading with respect CGRs values deduced from a pure static loading, depending on the orientation of specimens.
loading (R = 0.7, 2.8 10-4 Hz) on CGRs depended on the orientation : no significant modification was found with respect to                                         Materials a pure static loading in the direction TL perpendicular to the dendrites, but some significant increase (x 5 to 7) could be                 Two welds (D545 and D1054) in alloy 182 were investigated in observed in the direction of the dendrites TS. A stress relief               the as-welded conditions : they were realized by two different treatment (610°C) could lower the CGRs by a factor 2 to 3.5 with             manufacturers with SMAW process, using the same batch of respect to the as-welded condition. The results are discussed with           electrodes (Soudonel CQ5, diameter 4 mm) to fill a V-shape (10°)
This contributing paper presents the status of EDF R&D program on the initiation and propagation of SCC in weld metal 182, with a particular emphasize on the influence of the loading mode.
regard to available data from literature and compared to the                mould (width 50 mm at the bottom, thickness 35 mm, length 350 background on alloy 600 : the chemical composition of the weld              mm) made of alloy 600. The welding conditions were 110 A, 19 metal could lead to some variability (factor 3) in the CGRs (4              volts, temperature between weld deposition higher than 128°C.
Materials Two welds (D545 and D1054) in alloy 182 were investigated in the as-welded conditions : they were realized by two different manufacturers with SMAW process, using the same batch of electrodes (Soudonel CQ5, diameter 4 mm) to fill a V-shape (10°)
welds with significantly different chemical compositions).                  The resulting weld metal consisted in a pile-up of 15 layers with 7 Tests are scheduled to account for the influences of temperature            beads.
mould (width 50 mm at the bottom, thickness 35 mm, length 350 mm) made of alloy 600. The welding conditions were 110 A, 19 volts, temperature between weld deposition higher than 128°C.
and composition of the weld metal on the stress threshold and on            The chemical compositions and mechanical properties met the the crack velocities.                                                        RCC-M requirements.
The resulting weld metal consisted in a pile-up of 15 layers with 7 beads.
Table I. Chemical composition and mechanical properties of weld metals 182 C           Si       Mn         Ni         Cr       Ti     Nb + Ta       Fe         YS20     UTS20      YS350 (MPa)     (MPa)     (MPa)
The chemical compositions and mechanical properties met the RCC-M requirements.
Requir     d 0.10       d1      5.0 to     ! 59     13 to 17   d1        > 1.8     6 to 10     t 250       t 550     > 190 AWS                    0,6      9.5 5.11-76                  max D545       0.029       0.42       7.55     bal.       14.8     0.08       2.13       7.35       395         657       353 D1054       0.026       0.35       6.20       bal       15.0     0.05       2.00       6.90       386         627       347 United States Nuclear Regulatory Commission Official Hearing Exhibit In the Matter of:                   Entergy Nuclear Operations, Inc.
Table I. Chemical composition and mechanical properties of weld metals 182 C
Si Mn Ni Cr Ti Nb + Ta Fe YS20 (MPa)
UTS20 (MPa)
YS350 (MPa)
Requir AWS 5.11-76 d 0.10 d 1 0,6 max 5.0 to 9.5
! 59 13 to 17 d 1
> 1.8 6 to 10 t 250 t 550  
> 190 D545 0.029 0.42 7.55 bal.
14.8 0.08 2.13 7.35 395 657 353 D1054 0.026 0.35 6.20 bal 15.0 0.05 2.00 6.90 386 627 347 Proceedings of the 12th International Conference on Environmental Degradation of Materials in Nuclear Power System - Water Reactors -
Edited by T.R. Allen, P.J. King, and L. Nelson TMS (The Minerals, Metals & Materials Society), 2005 557 RIV000140 Submitted: November 9, 2012 United States Nuclear Regulatory Commission Official Hearing Exhibit In the Matter of:
Entergy Nuclear Operations, Inc.
(Indian Point Nuclear Generating Units 2 and 3)
(Indian Point Nuclear Generating Units 2 and 3)
ASLBP #: 07-858-03-LR-BD01 Docket #: 05000247 l 05000286 Exhibit #: RIV000140-00-BD01                Identified: 11/5/2015 557                    Admitted: 11/5/2015                        Withdrawn:
ASLBP #: 07-858-03-LR-BD01 Docket #: 05000247 l 05000286 Exhibit #:
Rejected:                                     Stricken:
Identified:
Admitted:
Withdrawn:
Rejected:
Stricken:
Other:
Other:
RIV000140-00-BD01 11/5/2015 11/5/2015


Experimental procedure Tests and specimens Initiation tests. The tests were mainly performed to determine the stress and strain thresholds for SCC, they included :
Experimental procedure Tests and specimens Initiation tests. The tests were mainly performed to determine the stress and strain thresholds for SCC, they included :
Line 38: Line 53:
x Constant extension rate tests (SSRTs, 5 10-8 s-1) on tensile specimens with a diameter of 4 mm and a gauge length of 85 mm, cut in the weld D1054. They were interrupted after various elongations in order to provide a threshold strain, most suitable for application to components in plants.
x Constant extension rate tests (SSRTs, 5 10-8 s-1) on tensile specimens with a diameter of 4 mm and a gauge length of 85 mm, cut in the weld D1054. They were interrupted after various elongations in order to provide a threshold strain, most suitable for application to components in plants.
x Periodic unloading-reloading at R = 0.9 every hour (frequency 2.8 10-4 Hz), on tensile specimens with a diameter of 4 mm and a gauge length of 25 mm, cut in the weld D545. They have allowed to assess the influence of a ripple load near the stress threshold defined by the previous constant load tests.
x Periodic unloading-reloading at R = 0.9 every hour (frequency 2.8 10-4 Hz), on tensile specimens with a diameter of 4 mm and a gauge length of 25 mm, cut in the weld D545. They have allowed to assess the influence of a ripple load near the stress threshold defined by the previous constant load tests.
Propagation tests. The tests were performed firstly to determine                       Figure 2. Autoclave for SSRTs at 360°C the influence of the main mechanical and metallurgical parameters         Constant load and ripple load tests were performed in a static on the CGRs, secondly to assess any influence of a periodic               autoclave placed in the load line of a creep machine. The load was unloading-reloading procedure (R = 0.7, frequency 2.8 10-4 Hz,           applied by dead weights to reach the maximum stress level (350 hold time 57 min) on the CGRs, with regard to the CGRs                   or 380 MPa), a part (10%) of the load was applied by a volume of measured with a pure static loading. CT specimens from both               water admitted in a tank installed on the dead weights, at regular welds were used, thickness 15 mm, pre-cracked by fatigue in air           intervals with a constant flow rate in order to reach the desired at Kfmax less than 15 MPam. They were cut in the welds                   trapezoidal loading (rising time 90 s, hold time 57 mn).
Propagation tests. The tests were performed firstly to determine the influence of the main mechanical and metallurgical parameters on the CGRs, secondly to assess any influence of a periodic unloading-reloading procedure (R = 0.7, frequency 2.8 10-4 Hz, hold time 57 min) on the CGRs, with regard to the CGRs measured with a pure static loading. CT specimens from both welds were used, thickness 15 mm, pre-cracked by fatigue in air at Kfmax less than 15 MPam. They were cut in the welds according to the TL orientation (fatigue pre-crack perpendicular to dendrites in a plane containing the axis of the dendrites) and the TS orientation (fatigue pre-crack in the direction of the axis of the dendrites, see Figure 1). They have allowed one to assess the influence of the orientation on the CGRs. The results were compared to previous results obtained both on CT, DCB and WOL specimens [2,5].
according to the TL orientation (fatigue pre-crack perpendicular to dendrites in a plane containing the axis of the dendrites) and the TS orientation (fatigue pre-crack in the direction of the axis of the dendrites, see Figure 1). They have allowed one to assess the influence of the orientation on the CGRs. The results were compared to previous results obtained both on CT, DCB and WOL specimens [2,5].
Figure 1. Orientation of the CT specimens in the welds Environment and facilities The basic environment contained 1000 ppm B as boric acid and 2 ppm Li as LiOH,H2O, with 30 cc/kg of hydrogen. The control of hydrogen depended on the facility and the temperature.
Figure 1. Orientation of the CT specimens in the welds Environment and facilities Figure 3. Facility for SCC tests with constant load or periodic The basic environment contained 1000 ppm B as boric acid and 2                                     unloading-reloading ppm Li as LiOH,H2O, with 30 cc/kg of hydrogen. The control of             Propagation tests. They were performed at 325°C in an autoclave hydrogen depended on the facility and the temperature.                   equipped with a load line containing 4 CT specimens, mounted on Initiation tests. They were conducted in static autoclaves at            a creep machine in which the load was applied by using dead 360°C, as shown in Figure 2 and Figure 3. The SSRT facility was          weights (static loading) or an air-jack (trapezoidal loading, rising described with details in [3] (Figure 2), deaeration was performed        time 90 s, hold time 57 mn), Figure 4. The autoclave was by evaporation of 20% of the water volume at 125°C, then                  constantly fed in primary water with a Cormet loop to ensure the hydrogen was introduced and controlled by a AgPd thimble at              required water chemistry (conductivity and oxygen control) and 360°C.                                                                    hydrogen was monitored by an Orbisphere sensor to meet a content of 30 cc/kg.
Initiation tests. They were conducted in static autoclaves at 360°C, as shown in Figure 2 and Figure 3. The SSRT facility was described with details in [3] (Figure 2), deaeration was performed by evaporation of 20% of the water volume at 125°C, then hydrogen was introduced and controlled by a AgPd thimble at 360°C.
Figure 2. Autoclave for SSRTs at 360°C Constant load and ripple load tests were performed in a static autoclave placed in the load line of a creep machine. The load was applied by dead weights to reach the maximum stress level (350 or 380 MPa), a part (10%) of the load was applied by a volume of water admitted in a tank installed on the dead weights, at regular intervals with a constant flow rate in order to reach the desired trapezoidal loading (rising time 90 s, hold time 57 mn).
Figure 3. Facility for SCC tests with constant load or periodic unloading-reloading Propagation tests. They were performed at 325°C in an autoclave equipped with a load line containing 4 CT specimens, mounted on a creep machine in which the load was applied by using dead weights (static loading) or an air-jack (trapezoidal loading, rising time 90 s, hold time 57 mn), Figure 4. The autoclave was constantly fed in primary water with a Cormet loop to ensure the required water chemistry (conductivity and oxygen control) and hydrogen was monitored by an Orbisphere sensor to meet a content of 30 cc/kg.
558
558


was noted for 1.5% elongation (Figure 5). It meant that the threshold strain ranged between 1.25 and 1.5%, which was roughly consistent with the threshold stress (350 MPa), due to the extremely severe SSRTs. For these reasons, this threshold appeared slightly lower than the value of 2% determined from capsules tests in [1].
Figure 4. Facility for SCC CGRs measurements with constant load or periodic unloading-reloading Evaluation of the crack depth Initiation tests. During the test, a LVDT sensor allowed a record of the elongation of the load line. The crack extension was assessed by examination of fracture surface (optical and/or scanning electron microscope). For unbroken specimen, the examination of the surface was realized optically.
Figure 4. Facility for SCC CGRs measurements with constant load or periodic unloading-reloading                     Figure 5. SC crack after an interrupted SSRT at 1.5% elongation at 360°C (weld D 1054)
Propagation tests. The true extension of the crack was performed on the fracture surface (optical and/or scanning electron microscope), along the longest crack branch in the case of macrobranching) : it has provided maximum value ('amax), and mean value ('amean) of the crack depth, in the latter case after measurement on 15 equally-distributed points of the crack front.
Evaluation of the crack depth                                           Influence of a ripple loading on Vth. Finally, the influence of a ripple loading (R = 0.9, 2.8 10-4 Hz) on the stress threshold was Initiation tests. During the test, a LVDT sensor allowed a record       investigated at 360°C on weld metal D545. At the maximum of the elongation of the load line. The crack extension was             stress of 380 MPa, a significant crack occurred after 7372 h under assessed by examination of fracture surface (optical and/or             a ripple loading but the result on a pure static loading at 380 MPa scanning electron microscope). For unbroken specimen, the               was not available at that time. At the maximum stress of 350 MPa, examination of the surface was realized optically.                       a rupture occurred by SCC after 14034 h under a ripple loading Propagation tests. The true extension of the crack was performed         (Figure 6). This result was very similar to the reference result on the fracture surface (optical and/or scanning electron               obtained with a pure static loading at 350 MPa (13700 h). Despite microscope), along the longest crack branch in the case of               the limited number of available results, a ripple loading did not macrobranching) : it has provided maximum value ('amax), and             seem to induce a significant influence on the stress threshold of mean value ('amean) of the crack depth, in the latter case after         alloy 182. A definitive conclusion will be drawn with regard to measurement on 15 equally-distributed points of the crack front.        the influence of a ripple loading at the end of the test at 380 MPa.
The crack growth rate (CGRmax or CGRmean) was deduced from the crack depth and from the duration of the test (generally 1300 h to 2000 h).
The crack growth rate (CGRmax or CGRmean) was deduced from the crack depth and from the duration of the test (generally 1300 h to 2000 h).
Results Initiation of SCC Threshold stress Vth. From previous constant load tests performed at 350°C on tensile specimens (EDF and CEA) and capsules (Framatome ANP) [1], it was deduced that for reasonable durations of tests, the stress threshold Vth for PWSCC of alloy 182 was near 350 MPa. Since that time, this assumption was reinforced by a rupture by SCC observed after 23771 h at 350 MPa on weld metal D 545 at 350°C [4]. Assuming an activation energy of 185 kJ/mol, it corresponded to a time for failure next to     Figure 6. Fracture surface of specimen in weld metal D 1054 after 13700 h at 360°C, which will be the reference result in the                       a ripple loading at R = 0.9, 2.8 10-4 Hz at 360°C followings.
Results Initiation of SCC Threshold stress Vth. From previous constant load tests performed at 350°C on tensile specimens (EDF and CEA) and capsules (Framatome ANP) [1], it was deduced that for reasonable durations of tests, the stress threshold Vth for PWSCC of alloy 182 was near 350 MPa. Since that time, this assumption was reinforced by a rupture by SCC observed after 23771 h at 350 MPa on weld metal D 545 at 350°C [4]. Assuming an activation energy of 185 kJ/mol, it corresponded to a time for failure next to 13700 h at 360°C, which will be the reference result in the followings.
Strain threshold. In order to improve the use of such data for plant     Propagation of SCC components, the threshold strain was investigated on both welds using interrupted SSRTs (5 10-8 s-1). Preliminary tests on weld         Fracture surface examination. Fracture surface of PWSCC on metal D545 have demonstrated that no cracking occurred after a           weld metal 182 appeared intergranular / interdendritic. The crack (plastic) elongation of 0.9%and a rupture by SCC was noted for           front was uneven and a strong macro-branching at 30-45° was an elongation ranging between 3.4% and 4.9%. Complementary              observed for the TL-orientation (Figure 7), while the crack front data were obtained on weld metal D1054, without any cracking            was most regular and the crack propagated roughly in the plane of for 1.0 and 1.25% elongation, while a very significant cracking 559
Strain threshold. In order to improve the use of such data for plant components, the threshold strain was investigated on both welds using interrupted SSRTs (5 10-8 s-1). Preliminary tests on weld metal D545 have demonstrated that no cracking occurred after a (plastic) elongation of 0.9%and a rupture by SCC was noted for an elongation ranging between 3.4% and 4.9%. Complementary data were obtained on weld metal D1054, without any cracking for 1.0 and 1.25% elongation, while a very significant cracking was noted for 1.5% elongation (Figure 5). It meant that the threshold strain ranged between 1.25 and 1.5%, which was roughly consistent with the threshold stress (350 MPa), due to the extremely severe SSRTs. For these reasons, this threshold appeared slightly lower than the value of 2% determined from capsules tests in [1].
Figure 5. SC crack after an interrupted SSRT at 1.5% elongation at 360°C (weld D 1054)
Influence of a ripple loading on Vth. Finally, the influence of a ripple loading (R = 0.9, 2.8 10-4 Hz) on the stress threshold was investigated at 360°C on weld metal D545. At the maximum stress of 380 MPa, a significant crack occurred after 7372 h under a ripple loading but the result on a pure static loading at 380 MPa was not available at that time. At the maximum stress of 350 MPa, a rupture occurred by SCC after 14034 h under a ripple loading (Figure 6). This result was very similar to the reference result obtained with a pure static loading at 350 MPa (13700 h). Despite the limited number of available results, a ripple loading did not seem to induce a significant influence on the stress threshold of alloy 182. A definitive conclusion will be drawn with regard to the influence of a ripple loading at the end of the test at 380 MPa.
Figure 6. Fracture surface of specimen in weld metal D 1054 after a ripple loading at R = 0.9, 2.8 10-4 Hz at 360°C Propagation of SCC Fracture surface examination. Fracture surface of PWSCC on weld metal 182 appeared intergranular / interdendritic. The crack front was uneven and a strong macro-branching at 30-45° was observed for the TL-orientation (Figure 7), while the crack front was most regular and the crack propagated roughly in the plane of 559


the fatigue pre-crack for the TS-orientation (the direction of the axis of the dendrites), Figure 8.
the fatigue pre-crack for the TS-orientation (the direction of the axis of the dendrites), Figure 8.
Figure 8. SCC at 325°C : fracture surface in the TS orientation (specimen 1471-TS1, D1054 AW, 22.9 MPam, R = 0.7)
Figure 7. SCC at 325°C : fracture surface in the TL orientation (specimen TL9, D1054 AW, 41.4 MPam, R = 0.7)
Figure 7. SCC at 325°C : fracture surface in the TL orientation (specimen TL9, D1054 AW, 41.4 MPam, R = 0.7)                                           x CGRs measured between 310°C and 330°C were considered and re-calculated at 325°C using an activation Re-assessment of previous results : da/dt versus K. Previous CGR                                  energy of 130 kJ/mol. This could be a reason for the data obtained by EDF, CEA and ETH-Zürich [5] with a static                                        typical behaviour of the 3 CEA results within a narrow loading were reassessed on weld metal D545, in the TL                                              range of K values (2 values at 330°C and one value at orientation :                                                                                      310°C).
Re-assessment of previous results : da/dt versus K. Previous CGR data obtained by EDF, CEA and ETH-Zürich [5] with a static loading were reassessed on weld metal D545, in the TL orientation :
x different loading modes have been applied (constant load                            The mean curve for the maximum values of CGR of weld D545 as (CT) or constant deflection (WOL or DCB)).                                        a function of the initial value of K became :
x
Consequently, the initial K values were re-calculated                                                  (da/d)max = 0.5114.(K-9)0.42 taking into account the anisothermal stress relaxation                            using K values ranging between 18 and 41 MPam (Figure 9).
different loading modes have been applied (constant load (CT) or constant deflection (WOL or DCB)).
between 20°C and 325°C for constant deflection tests,                              This K-dependence was not so far from the law reported in the roughly 18%.                                                                      case of alloy 600, with the exponent between 0.3 and 0.5, see [6].
Consequently, the initial K values were re-calculated taking into account the anisothermal stress relaxation between 20°C and 325°C for constant deflection tests, roughly 18%.
Figure 8. SCC at 325°C : fracture surface in the TS orientation (specimen 1471-TS1, D1054 AW, 22.9 MPam, R = 0.7) x
CGRs measured between 310°C and 330°C were considered and re-calculated at 325°C using an activation energy of 130 kJ/mol. This could be a reason for the typical behaviour of the 3 CEA results within a narrow range of K values (2 values at 330°C and one value at 310°C).
The mean curve for the maximum values of CGR of weld D545 as a function of the initial value of K became :
(da/d)max = 0.5114.(K-9)0.42 using K values ranging between 18 and 41 MPam (Figure 9).
This K-dependence was not so far from the law reported in the case of alloy 600, with the exponent between 0.3 and 0.5, see [6].
Significant data are required to better define the curve at low K values.
Significant data are required to better define the curve at low K values.
D54 5 & D1 054 - AW - TL - s tatic 10,00 D545 : CGRmax325 = 0,5114(K-9)^0,42 1,00 CGRmax325 (E-10 m/s ETH D545 AW TL EDF D545 AW TL CEA D545 AW TL CGRmax325 D545 AW TL EDF D1054 AW TL 0,10 0,01 0  10  20        30              40              50                60 K (MPa.m0.5)
D545 & D1054 - AW - TL - static 0,01 0,10 1,00 10,00 0
Figure 9. Mean curve of the max. CGRs as a function of K on alloy 182 (weld D545) in the TL orientation at 325°C The new results obtained in the TL orientation with weld D1054                             present results demonstrated that the orientation induced were in close agreement with these data (Figure 9) and                                     significant changes in CGRs. The CGRs measured in the TS demonstrated a good reproducibility for tests performed at the                             orientation were generally higher than in the TL orientation : 2 same K-value. With the available data, the proposed law for                               times for D545 with a static loading, 2.5 to 3 times for D1054 PWSCC of 182 was : (da/d)max = D(material).(K-9)0.4.                                       with a trapezoidal loading at R = 0.7.
10 20 30 40 50 60 K (MPa.m0.5)
Influence of the orientation. Though true crack lengths have been                          Influence of a trapezoidal loading (R = 0.7). The influence of a considered in the case of macro-branching (TL) or straight                                trapezoidal loading on the CGRs was examined on both welds in propagation in the direction of the axis of dendrites (TS), the                            the TL and TS orientations.
CGRmax325 (E-10 m/s ETH D545 AW TL EDF D545 AW TL CEA D545 AW TL CGRmax325 D545 AW TL EDF D1054 AW TL D545 : CGRmax325 = 0,5114(K-9)^0,42 Figure 9. Mean curve of the max. CGRs as a function of K on alloy 182 (weld D545) in the TL orientation at 325°C The new results obtained in the TL orientation with weld D1054 were in close agreement with these data (Figure 9) and demonstrated a good reproducibility for tests performed at the same K-value. With the available data, the proposed law for PWSCC of 182 was : (da/d)max = D(material).(K-9)0.4.
Influence of the orientation. Though true crack lengths have been considered in the case of macro-branching (TL) or straight propagation in the direction of the axis of dendrites (TS), the present results demonstrated that the orientation induced significant changes in CGRs. The CGRs measured in the TS orientation were generally higher than in the TL orientation : 2 times for D545 with a static loading, 2.5 to 3 times for D1054 with a trapezoidal loading at R = 0.7.
Influence of a trapezoidal loading (R = 0.7). The influence of a trapezoidal loading on the CGRs was examined on both welds in the TL and TS orientations.
560
560


On weld D545, the data from Westinghouse [7] obtained with a                                                                                        the CGRs obtained with a trapezoidal loading in the TS trapezoidal loading (R = 0.7) were compared to the above-                                                                                           orientation were 5 to 7 times higher than the reference (TL) curve mentioned reference curve for a static loading, in the TL                                                                                           in static loading (Figure 10).
On weld D545, the data from Westinghouse [7] obtained with a trapezoidal loading (R = 0.7) were compared to the above-mentioned reference curve for a static loading, in the TL orientation. It could be concluded that no significant influence of the loading procedure was observed in the TL orientation, while the CGRs obtained with a trapezoidal loading in the TS orientation were 5 to 7 times higher than the reference (TL) curve in static loading (Figure 10).
orientation. It could be concluded that no significant influence of                                                                                 .
D545, Static / dynamic 0,01 0,10 1,00 10,00 0
the loading procedure was observed in the TL orientation, while D545, Static / dynamic 1 0,0 0 C GRm ax 32 5 = 0 ,5 11 4(K -9) ^0 ,4 2 1,0 0 CGR ma x3 25 (E-10 m/s)
10 20 30 40 K (M Pa.m0.5)
We st A W TS R = 0.7 We st A W TL R = 0 .7 C GRma x3 25 AW TL s ta tic 0,1 0 0,0 1 0                          10                      20                      30              40 K (M Pa.m 0.5)
CGRmax325 (E-10 m/s)
Figure 10. CGRs on weld D545 : comparison between static and trapezoidal loadings On weld D1054, the CGRs were also very similar in static and in                                                                                     trapezoidal loading from [7] and from the present paper in the trapezoidal loading in the TL orientation, slightly higher than for                                                                                 TS orientation, would be in good agreement (those from [7]
West A W TS R = 0.7 West A W TL R = 0.7 CGRmax325 AW TL s tatic CGRmax325 = 0,5114(K-9) ^0,42 Figure 10. CGRs on weld D545 : comparison between static and trapezoidal loadings On weld D1054, the CGRs were also very similar in static and in trapezoidal loading in the TL orientation, slightly higher than for weld D545 (Figure 11). By contrast, CGRs measured under a trapezoidal loading with the TS orientation were 3 to 5 times higher than those deduced from the reference (TL) curve.
weld D545 (Figure 11). By contrast, CGRs measured under a                                                                                           were the highest). Finally, it seemed that the influence of trapezoidal loading with the TS orientation were 3 to 5 times                                                                                       trapezoidal versus static loading on the SCC CGRs of alloy 182 higher than those deduced from the reference (TL) curve.                                                                                             depended on the orientation : no significant influence in the TL Unfortunately, the data for the TS orientation with a pure static                                                                                   orientation, but a detrimental factor 2 to 7 could be observed loading was not yet available, but it could be expected from the                                                                                     with the TS orientation.
Unfortunately, the data for the TS orientation with a pure static loading was not yet available, but it could be expected from the available CGRs on the two (similar) welds that data with a trapezoidal loading from [7] and from the present paper in the TS orientation, would be in good agreement (those from [7]
available CGRs on the two (similar) welds that data with a 10 D 1054-TS D1054-TL 1
were the highest). Finally, it seemed that the influence of trapezoidal versus static loading on the SCC CGRs of alloy 182 depended on the orientation : no significant influence in the TL orientation, but a detrimental factor 2 to 7 could be observed with the TS orientation.
CGR ma x325 (E-10 m/s)
0,01 0,1 1
D545 AW TL static D545-TL                                                                            D1054 AW TS R 0,7 D1054 AW TL R 0 ,7 D1054 AW TL static 0 ,1 0,01 0                10                    20              30            40              50        60 K (MPa.m0.5)
10 0
Figure 11. CGRs on weld D1054 : comparison between static and trapezoidal loadings Influence of the stress relief treatment. The CGRs obtained at 325°C with welds D545 under a static loading and D1054 under                                                                                                                           Discussion a trapezoidal loading in the stress-relief condition (610°C-20 h) were 2 times lower than the CGRs measured in the as-welded                                                                                           Some comparisons could be realized with available data in condition (at the limit of the scatterband for reproducibility but                                                                                   relation to the influence of some major parameters on PWSCC always in the lowest part).                                                                                                                          of alloy 182.
10 20 30 40 50 60 K (MPa.m0.5)
CGRmax325 (E-10 m/s)
D545 AW TL static D1054 AW TS R 0,7 D1054 AW TL R 0,7 D1054 AW TL static D545-TL D1054-TL D 1054-TS Figure 11. CGRs on weld D1054 : comparison between static and trapezoidal loadings Influence of the stress relief treatment. The CGRs obtained at 325°C with welds D545 under a static loading and D1054 under a trapezoidal loading in the stress-relief condition (610°C-20 h) were 2 times lower than the CGRs measured in the as-welded condition (at the limit of the scatterband for reproducibility but always in the lowest part).
Discussion Some comparisons could be realized with available data in relation to the influence of some major parameters on PWSCC of alloy 182.
561
561


set of international average values from different welds, at 75%
Comparison of da/dt versus K curves between laboratories The K-dependence of CGRs was investigated by several laboratories. The Figure 12 compares the da/dt versus K curves proposed by EPRI and the preliminary reference curve proposed in this paper. It should be remind that EPRI-MRP 115 curve, with assumed KISCC = 0, was obtained as the best fit for a set of international average values from different welds, at 75%
Comparison of da/dt versus K curves between laboratories                                                                  of the cumulative distribution ((da/dt)ave325 = 1.5 10-12.K1.6), with static or trapezoidal loadings. By contrast, the proposed The K-dependence of CGRs was investigated by several                                                                        reference curve in this paper has assumed a KISCC value of 9 laboratories. The Figure 12 compares the da/dt versus K                                                                  MPam, and the proposed law ((da/d)max = 0.5114.(K-9)0.42 for curves proposed by EPRI and the preliminary reference curve                                                                welds D545 or D1054) for a pure static loading was rather proposed in this paper. It should be remind that EPRI-MRP 115                                                              similar to the law proposed for alloy 600 [6].
of the cumulative distribution ((da/dt)ave325 = 1.5 10-12.K1.6), with static or trapezoidal loadings. By contrast, the proposed reference curve in this paper has assumed a KISCC value of 9 MPam, and the proposed law ((da/d)max = 0.5114.(K-9)0.42 for welds D545 or D1054) for a pure static loading was rather similar to the law proposed for alloy 600 [6].
curve, with assumed KISCC = 0, was obtained as the best fit for a 10 D 1054-TS R 0,7 Rhingals curve :
0,01 0,1 1
                                                                                  -23  9,3 K < 25 : CGRmax325 = 7,22 10         K
10 0
                                                                                  -10                                        D1054-TL K > 25 : CGRmax325 = 6 10 1                                                                                                                D545 EDF AW TL static EDF on D545-TL CGRmax325 (E-10 m/s)
10 20 30 40 50 60 K (MPa.m0.5)
D1054 AW TS R 0,7 CGRmax325 = 0,5114(K-9)0,42                              D1054 AW TL R 0,7 D1054 AW TL static EPRI MRP 115 132 JNES AW TS static Studsvik lab data [8]
CGRmax325 (E-10 m/s)
D545 EDF AW TL static D1054 AW TS R 0,7 D1054 AW TL R 0,7 D1054 AW TL static EPRI MRP 115 132 JNES AW TS static Studsvik lab data [8]
Ringhals field data [8]
Ringhals field data [8]
Ringhals curve TS 0,1 EPRI MRP 115 CGRave325 = 1,5 10-12 K1,6 0,01 0              10                   20              30              40                50        60 K (MPa.m0.5)
Ringhals curve TS EDF on D545-TL CGRmax325 = 0,5114(K-9)0,42 D1054-TL D 1054-TS R 0,7 EPRI MRP 115 CGRave325 = 1,5 10-12 K1,6 Rhingals curve :
Figure 12. Reference curves for weld metal 182 by EPRI, Ringhals and EDF from laboratory. Comparison with JNES results The Ringhals curve in the TS orientation presented a part with a                                                            was observed for weld metal 182 in the TL orientation while a dependence on K9.3 at K lower than 25 MPam, and a plateau                                                                 significant increase could be noted with a trapezoidal loading in regime at 6 10-10 m/s at K values higher than 25.1 MPam,                                                                   the most sensitive (TS) orientation.
K < 25 : CGRmax325 = 7,22 10-23 K9,3 K > 25 : CGRmax325 = 6 10-10 Figure 12. Reference curves for weld metal 182 by EPRI, Ringhals and EDF from laboratory. Comparison with JNES results The Ringhals curve in the TS orientation presented a part with a dependence on K9.3 at K lower than 25 MPam, and a plateau regime at 6 10-10 m/s at K values higher than 25.1 MPam, which was consistent with EDF values on weld metal D1054 in the same (TS) orientation and with a trapezoidal loading. The results they obtained in laboratory under a static loading and from the field [8] were in close agreement with the curve suggested by EDF.
which was consistent with EDF values on weld metal D1054 in                                                                 JNES data obtained in the TS orientation at 325&deg;C (calculated the same (TS) orientation and with a trapezoidal loading. The                                                              from 340&deg;C) under a pure static loading ranged between EPRI results they obtained in laboratory under a static loading and                                                              and EDF curves, they referred to a law similar to (K-9)1.16, from the field [8] were in close agreement with the curve                                                                  Figure 12, [11].
Trapezoidal loading versus static loading It should be noted that the data generated in this paper on weld D1054 in the TS orientation with a trapezoidal loading were slightly higher than the EPRI-curve which should account for this kind of loading. The differences in the shape and the level of EPRI and EDF curves could be the consequence of the differences in the loading procedures. It should be noted that Tsutsumi et al [9] did not recommend the use of a trapezoidal loading to measure CGRs on weld metal 132, unless hold times higher than 105 s would be used or unless pre-cracking of the specimen at the beginning of the test.
suggested by EDF.
This kind of conclusion was still brought on alloy 600 in primary water : no change in CGRs was observed on very susceptible heats, while a significant increase of CGRs under a trapezoidal loading with respect to a static loading could be noted for heats with a low susceptibility to SCC [10]. Finally, it was not possible to explain the reason why no similar influence was observed for weld metal 182 in the TL orientation while a significant increase could be noted with a trapezoidal loading in the most sensitive (TS) orientation.
Influence of the chemical composition of the weld Trapezoidal loading versus static loading In this paper, two weld metals from the same batch of electrodes It should be noted that the data generated in this paper on weld                                                            and deposited in similar processes by two different D1054 in the TS orientation with a trapezoidal loading were                                                                manufacturers have provided very similar data, as expected.
JNES data obtained in the TS orientation at 325&deg;C (calculated from 340&deg;C) under a pure static loading ranged between EPRI and EDF curves, they referred to a law similar to (K-9)1.16, Figure 12, [11].
slightly higher than the EPRI-curve which should account for                                                                However, the scatter on CGR results appeared to be somewhat this kind of loading. The differences in the shape and the level of                                                        larger on weld D545 than on weld D1054. This difference could EPRI and EDF curves could be the consequence of the                                                                        come from laboratory-to-laboratory test procedures, since weld differences in the loading procedures. It should be noted that                                                              D545 has been tested by 3 different laboratories and weld Tsutsumi et al [9] did not recommend the use of a trapezoidal                                                              D1054 by EDF only. On weld D545, the scatter was mainly loading to measure CGRs on weld metal 132, unless hold times                                                                observed within the K-range of 15-25 MPam, in which the higher than 105 s would be used or unless pre-cracking of the                                                              influence of K on CGRs was significant, while EDF has specimen at the beginning of the test.                                                                                      investigated a broader range of K (20-42 MPam) on weld This kind of conclusion was still brought on alloy 600 in                                                                  D1054, for which the influence of K was the lowest.
Influence of the chemical composition of the weld In this paper, two weld metals from the same batch of electrodes and deposited in similar processes by two different manufacturers have provided very similar data, as expected.
primary water : no change in CGRs was observed on very                                                                      Besides, some CGRs measured at 330&deg;C by CEA on 3 other susceptible heats, while a significant increase of CGRs under a                                                            welds in alloy 182 with C contents ranging from 0.022% to trapezoidal loading with respect to a static loading could be                                                              0.089%, and Si contents from 0.27% to 0.79% were found to be noted for heats with a low susceptibility to SCC [10]. Finally, it                                                          somewhat similar : calculated values of the CGRmax at 325&deg;C was not possible to explain the reason why no similar influence                                                            were 1.4 to 1.5 10-10 m/s for high or low C low Si weld, 1.8 to 562
However, the scatter on CGR results appeared to be somewhat larger on weld D545 than on weld D1054. This difference could come from laboratory-to-laboratory test procedures, since weld D545 has been tested by 3 different laboratories and weld D1054 by EDF only. On weld D545, the scatter was mainly observed within the K-range of 15-25 MPam, in which the influence of K on CGRs was significant, while EDF has investigated a broader range of K (20-42 MPam) on weld D1054, for which the influence of K was the lowest.
Besides, some CGRs measured at 330&deg;C by CEA on 3 other welds in alloy 182 with C contents ranging from 0.022% to 0.089%, and Si contents from 0.27% to 0.79% were found to be somewhat similar : calculated values of the CGRmax at 325&deg;C were 1.4 to 1.5 10-10 m/s for high or low C low Si weld, 1.8 to 562


2.35 10-10 m/s for the high C high Si weld, and 0.75 to 4.5 10-10 m/s on weld D545 [5].                                                                             References Finally, weld to weld variability led to a factor 3 on the CGRs for the same TL orientation.                                           1 C. Amzallag et al, Stress Corrosion Life Assessment of 182 and 82 Welds used in PWR Components (10th International Influence of the stress relief-treatment                               Conference on Environmental Degradation in Nuclear Power Systems - Water Reactors, South Lake Tahoe (NV), august CGRs measured on weld metal D545 or D1054 in the SR                     2001) conditions were generally 2 to 3.5 times lower than in the AW conditions in most of the laboratories (EDF, CEA, ETH-Z&#xfc;rich)           2 C. Amzallag et al, Stress Corrosion Life Experience of 182 at 320-330&deg;C, just above the limit for significance according to       and 82 welds in French PWRs (Contribution of Materials the reproducibility factor of 2 for this kind of tests. This slight     Investigation to the Resolution of Problems Encountered on beneficial effect of SR treatment at 610&deg;C could not be                 Pressurized Water Reactors, Fontevraud V, France, September explained by any significant change in the microstructure or           2002) density of dislocations [12] but could rely on some contribution of residual stresses to the true K level in the case of the AW         3 Francois Vaillant, Chedly Braham, Jean-Marie Gras, condition. It should be noted that a significant beneficial effect     << Etude de la corrosion sous contrainte en milieu primaire des of SR was noted by [13] at 290&deg;C (factor 3 to 10 in the TL             alliages X-750 et X-750 modifi&#xe9;  19% de chrome  laide de orientation).                                                           lessai de traction lente. Influence du traitement thermique >>,
2.35 10-10 m/s for the high C high Si weld, and 0.75 to 4.5 10-10 m/s on weld D545 [5].
For comparison, some significant beneficial influence of the SR         (EDF report HT-45/PV D 736A), january 1990) treatment was observed on the resistance to the initiation of SCC in the case of highly susceptible materials (high C, high Si), the     [4] << La corrosion sous contrainte des alliages de nickel dans effect was limited in the case of materials with the lowest             leau  haute temp&#xe9;rature >>, F. Vaillant et al Editors, Reference susceptibility to SCC [1].                                             book, (EDF report HT-29/04/077/A, 2005)
Finally, weld to weld variability led to a factor 3 on the CGRs for the same TL orientation.
Finally, the root causes of the major beneficial influence of SR treatment in plants rely partly on the reduction in the residual stresses which could reach 50% of the initial stresses and, in the
Influence of the stress relief-treatment CGRs measured on weld metal D545 or D1054 in the SR conditions were generally 2 to 3.5 times lower than in the AW conditions in most of the laboratories (EDF, CEA, ETH-Z&#xfc;rich) at 320-330&deg;C, just above the limit for significance according to the reproducibility factor of 2 for this kind of tests. This slight beneficial effect of SR treatment at 610&deg;C could not be explained by any significant change in the microstructure or density of dislocations [12] but could rely on some contribution of residual stresses to the true K level in the case of the AW condition. It should be noted that a significant beneficial effect of SR was noted by [13] at 290&deg;C (factor 3 to 10 in the TL orientation).
: 5. S. Le Hong et al, Measurements of Stress Corrosion Cracking Growth Rates in Weld Alloys 182 in Primary Water of case of initiation, on a possible recristallization of the surface PWR (10th International Conference of Environmental cold worked layer. The intrinsic influence of SR treatment on Degradation of Materials in Nuclear Power Systems - Water the initiation of SCC is low on specimens with low C and Si.
For comparison, some significant beneficial influence of the SR treatment was observed on the resistance to the initiation of SCC in the case of highly susceptible materials (high C, high Si), the effect was limited in the case of materials with the lowest susceptibility to SCC [1].
Reactors, South Lake Tahoe (NV), august 2001)
Finally, the root causes of the major beneficial influence of SR treatment in plants rely partly on the reduction in the residual stresses which could reach 50% of the initial stresses and, in the case of initiation, on a possible recristallization of the surface cold worked layer. The intrinsic influence of SR treatment on the initiation of SCC is low on specimens with low C and Si.
Conclusion 6    F. VAILLANT et al, Crack Growth Rates in Thick The initiation and growth of PWSCC of alloy 182 were                   Materials of Alloy 600 and Weld Metals of Alloy 182 in investigated with regard to the influence of the loading               Laboratory Primary Water. Comparison with Field Experience parameters.                                                             (Contribution of Materials Investigation to the Resolution of The stress threshold for initiation of SCC at 360&deg;C was found           Problems Encountered on Pressurized Water Reactors, close to 350 MPa in constant load tests. With the limited               Fontevraud V, France, September 2002, pp 107-116) available data, it seemed that a ripple loading did not modify this value. The strain threshold was determined by interrupted               7. John Hickling, << Materials Reliability Program. Crack SSRTs : SCC occurred for an elongation between 1.25 and                 Growth Rates for Evaluating Primary Water Stress Corrosion 1.5%.                                                                   Cracking (PWSCC) of Alloy 82, 182 and 132 welds (MRP-Preliminary data of crack growth rates (CGR) at 325&deg;C provided         115) (EPRI report 1006696, November 2004) a dependence on (K-9)0.4 for the average values of the CGRmax.
Conclusion The initiation and growth of PWSCC of alloy 182 were investigated with regard to the influence of the loading parameters.
The strongly oriented microstructure led to faster (x 2 to 3)           8 A. Jenssen et al, A Swedish Perspective on PWSCC of Alloy CGRs in the direction of the dendrites (TS) than in the                 182 (2005 International PWSCC of Alloy 600 Conference and perpendicular direction (TL). The influence on the CGRs of a           Exhibit Show, Hyatt Regency Tamaya Resort, Santa Ana Pueblo trapezoidal loading (R = 0.7, 2.8 10-4 Hz) with regard to a pure       (NM), 7-10 March 2005) static loading depended on the orientation : no significant modification in the direction perpendicular to the dendrites (TL),     9 K. Tsutsumi et al, SCC Growth Rate of Nickel based Alloy but some significant increase (x 2 to 7) could be observed in the       132 Weld Metal in PWR Primary Water (11th International direction of the dendrites (TS). A stress relief treatment (610&deg;C)     Conference on Environmental Degradation in Nuclear Power could lower the CGRs by a factor 2 to 3.5 with respect to the as-       Systems - Water Reactors, Stevenson (WA), august 2003) welded condition. The chemical composition of the weld metal could lead to some variability in the CGRs (factor 3).                 10. F. Vaillant et al, Influence of a Cyclic Loading on Crack Work is still in progress to assess the influence of temperature       Growth Rates of Alloy 600 in Primary Environment : An and chemical composition of the weld on the stress threshold for       Overview (11th International Conference of Environmental the initiation of SCC. For propagation, the influence of low           Degradation of Materials in Nuclear Power Systems - Water values of K (less than 20 MPam) on the da/dt versus K               Reactors, Stevenson (WA), august 2003) curve, temperature, stress relief treatment and chemical composition of the weld metal will be investigated.                     11 Y. Yamamoto et al (JNES), Outline of Evaluation Technology for SCC Growth of Ni Base Alloys (NiSCC)
The stress threshold for initiation of SCC at 360&deg;C was found close to 350 MPa in constant load tests. With the limited available data, it seemed that a ripple loading did not modify this value. The strain threshold was determined by interrupted SSRTs : SCC occurred for an elongation between 1.25 and 1.5%.
Preliminary data of crack growth rates (CGR) at 325&deg;C provided a dependence on (K-9)0.4 for the average values of the CGRmax.
The strongly oriented microstructure led to faster (x 2 to 3)
CGRs in the direction of the dendrites (TS) than in the perpendicular direction (TL). The influence on the CGRs of a trapezoidal loading (R = 0.7, 2.8 10-4 Hz) with regard to a pure static loading depended on the orientation : no significant modification in the direction perpendicular to the dendrites (TL),
but some significant increase (x 2 to 7) could be observed in the direction of the dendrites (TS). A stress relief treatment (610&deg;C) could lower the CGRs by a factor 2 to 3.5 with respect to the as-welded condition. The chemical composition of the weld metal could lead to some variability in the CGRs (factor 3).
Work is still in progress to assess the influence of temperature and chemical composition of the weld on the stress threshold for the initiation of SCC. For propagation, the influence of low values of K (less than 20 MPam) on the da/dt versus K curve, temperature, stress relief treatment and chemical composition of the weld metal will be investigated.
References 1
C. Amzallag et al, Stress Corrosion Life Assessment of 182 and 82 Welds used in PWR Components (10th International Conference on Environmental Degradation in Nuclear Power Systems - Water Reactors, South Lake Tahoe (NV), august 2001) 2 C. Amzallag et al, Stress Corrosion Life Experience of 182 and 82 welds in French PWRs (Contribution of Materials Investigation to the Resolution of Problems Encountered on Pressurized Water Reactors, Fontevraud V, France, September 2002) 3 Francois Vaillant, Chedly Braham, Jean-Marie Gras,
<< Etude de la corrosion sous contrainte en milieu primaire des alliages X-750 et X-750 modifi&#xe9; 19% de chrome laide de lessai de traction lente. Influence du traitement thermique >>,
(EDF report HT-45/PV D 736A), january 1990)
[4] << La corrosion sous contrainte des alliages de nickel dans leau haute temp&#xe9;rature >>, F. Vaillant et al Editors, Reference book, (EDF report HT-29/04/077/A, 2005)
: 5.
S. Le Hong et al, Measurements of Stress Corrosion Cracking Growth Rates in Weld Alloys 182 in Primary Water of PWR (10th International Conference of Environmental Degradation of Materials in Nuclear Power Systems - Water Reactors, South Lake Tahoe (NV), august 2001) 6 F. VAILLANT et al, Crack Growth Rates in Thick Materials of Alloy 600 and Weld Metals of Alloy 182 in Laboratory Primary Water. Comparison with Field Experience (Contribution of Materials Investigation to the Resolution of Problems Encountered on Pressurized Water
: Reactors, Fontevraud V, France, September 2002, pp 107-116) 7.
John Hickling, << Materials Reliability Program. Crack Growth Rates for Evaluating Primary Water Stress Corrosion Cracking (PWSCC) of Alloy 82, 182 and 132 welds (MRP-115) (EPRI report 1006696, November 2004) 8 A. Jenssen et al, A Swedish Perspective on PWSCC of Alloy 182 (2005 International PWSCC of Alloy 600 Conference and Exhibit Show, Hyatt Regency Tamaya Resort, Santa Ana Pueblo (NM), 7-10 March 2005) 9 K. Tsutsumi et al, SCC Growth Rate of Nickel based Alloy 132 Weld Metal in PWR Primary Water (11th International Conference on Environmental Degradation in Nuclear Power Systems - Water Reactors, Stevenson (WA), august 2003)
: 10. F. Vaillant et al, Influence of a Cyclic Loading on Crack Growth Rates of Alloy 600 in Primary Environment : An Overview (11th International Conference of Environmental Degradation of Materials in Nuclear Power Systems - Water Reactors, Stevenson (WA), august 2003) 11 Y. Yamamoto et al (JNES), Outline of Evaluation Technology for SCC Growth of Ni Base Alloys (NiSCC)
Project in Japan and Current Results in PWR Environment 563
Project in Japan and Current Results in PWR Environment 563


Latest revision as of 06:21, 10 January 2025

Official Exhibit - RIV000140-00-BD01 - Francois Vaillant, Et Al., Influence of a Cyclic Loading on the Initiation and Propagation of PWSCC in Weld Metal 182, Proceedings of the 12th International Conference on Environmental
ML15329A273
Person / Time
Site: Indian Point  Entergy icon.png
Issue date: 12/31/2005
From: Couvant T, Vaillant F, Amzallag C, Boursier J, Champredonde J
Govt of France, EDF Generation
To:
Atomic Safety and Licensing Board Panel
SECY RAS
References
RAS 23735, 50-247-LR, 50-286-LR, ASLBP 07-858-03-LR-BD01
Download: ML15329A273 (9)
v • d • e


Text

INFLUENCE OF A CYCLIC LOADING ON THE INITIATION AND PROPAGATION OF PWSCC IN WELD METAL 182 François Vaillant1, Jean-Marie Boursier1, Thierry Couvant1, Claude Amzallag2, Jacques Champredonde3 1EDF/R&D/MMC, Les Renardires Research Center, 77818 Moret sur Loing, France 2EDF/SEPTEN, 12-14 avenue Dutrievoz, 69628 Villeurbanne Cedex, France 3EDF/CEIDRE, 2 rue Ampre, 93206 - Saint-Denis Cedex 01 Keywords: Stress Corrosion Cracking, weld metal 182, primary water, cyclic loading Abstract PWSCC (Primary Water Stress Corrosion Cracking) of nickel based weld metal 182 is now a major issue in Pressurised Water Reactors, since significant cracking was observed in several welds of the primary circuit in some countries. Many questions arise about the stress (or strain) threshold for initiation and on the crack growth rates of SCC, with significant differences depending on the orientation of the loading axis with regard to the dendrites.

The stress threshold for initiation was determined at the level of 350 MPa with constant load tests on as-welded material. This threshold was not significantly modified by a ripple loading at 360°C. The strain threshold for initiation, determined by interrupted SSRTs for various elongations, was found to range between 1.25 and 1.5%.

Preliminary data of crack growth rates (CGRs) at 325°C provided a dependence on (K-9)0.4 for the average values of the CGRmax.

The strongly oriented microstructure led to faster (x 2 to 3) CGRs in the direction of the dendrites. The influence of a trapezoidal loading (R = 0.7, 2.8 10-4 Hz) on CGRs depended on the orientation : no significant modification was found with respect to a pure static loading in the direction TL perpendicular to the dendrites, but some significant increase (x 5 to 7) could be observed in the direction of the dendrites TS. A stress relief treatment (610°C) could lower the CGRs by a factor 2 to 3.5 with respect to the as-welded condition. The results are discussed with regard to available data from literature and compared to the background on alloy 600 : the chemical composition of the weld metal could lead to some variability (factor 3) in the CGRs (4 welds with significantly different chemical compositions).

Tests are scheduled to account for the influences of temperature and composition of the weld metal on the stress threshold and on the crack velocities.

Introduction PWSCC (Primary Water Stress Corrosion Cracking) of nickel based weld metal 182 is now a major issue in Pressurised Water Reactors, since significant cracking was observed in several welds of the primary circuit in several countries. Despite important R&D efforts to determine the influence of the main parameters which could affect PWSCC of alloy 182, many questions arose about the stress- (and strain-) threshold for initiation and on the crack growth rates of SCC, with significant differences depending on the orientation of the loading axis with regard to the dendrites.

Moreover, because crack growth rates are sometimes extensively determined using periodic unloading reloading procedures, a comparison was proposed for CGRs obtained with this kind of loading with respect CGRs values deduced from a pure static loading, depending on the orientation of specimens.

This contributing paper presents the status of EDF R&D program on the initiation and propagation of SCC in weld metal 182, with a particular emphasize on the influence of the loading mode.

Materials Two welds (D545 and D1054) in alloy 182 were investigated in the as-welded conditions : they were realized by two different manufacturers with SMAW process, using the same batch of electrodes (Soudonel CQ5, diameter 4 mm) to fill a V-shape (10°)

mould (width 50 mm at the bottom, thickness 35 mm, length 350 mm) made of alloy 600. The welding conditions were 110 A, 19 volts, temperature between weld deposition higher than 128°C.

The resulting weld metal consisted in a pile-up of 15 layers with 7 beads.

The chemical compositions and mechanical properties met the RCC-M requirements.

Table I. Chemical composition and mechanical properties of weld metals 182 C

Si Mn Ni Cr Ti Nb + Ta Fe YS20 (MPa)

UTS20 (MPa)

YS350 (MPa)

Requir AWS 5.11-76 d 0.10 d 1 0,6 max 5.0 to 9.5

! 59 13 to 17 d 1

> 1.8 6 to 10 t 250 t 550

> 190 D545 0.029 0.42 7.55 bal.

14.8 0.08 2.13 7.35 395 657 353 D1054 0.026 0.35 6.20 bal 15.0 0.05 2.00 6.90 386 627 347 Proceedings of the 12th International Conference on Environmental Degradation of Materials in Nuclear Power System - Water Reactors -

Edited by T.R. Allen, P.J. King, and L. Nelson TMS (The Minerals, Metals & Materials Society), 2005 557 RIV000140 Submitted: November 9, 2012 United States Nuclear Regulatory Commission Official Hearing Exhibit In the Matter of:

Entergy Nuclear Operations, Inc.

(Indian Point Nuclear Generating Units 2 and 3)

ASLBP #: 07-858-03-LR-BD01 Docket #: 05000247 l 05000286 Exhibit #:

Identified:

Admitted:

Withdrawn:

Rejected:

Stricken:

Other:

RIV000140-00-BD01 11/5/2015 11/5/2015

Experimental procedure Tests and specimens Initiation tests. The tests were mainly performed to determine the stress and strain thresholds for SCC, they included :

x Constant load tests in order to assess a threshold stress on tensile specimens with a diameter of 4 mm and a gauge length of 25 mm. Most of the results obtained on weld D545 were previously reported in [1,2].

x Constant extension rate tests (SSRTs, 5 10-8 s-1) on tensile specimens with a diameter of 4 mm and a gauge length of 85 mm, cut in the weld D1054. They were interrupted after various elongations in order to provide a threshold strain, most suitable for application to components in plants.

x Periodic unloading-reloading at R = 0.9 every hour (frequency 2.8 10-4 Hz), on tensile specimens with a diameter of 4 mm and a gauge length of 25 mm, cut in the weld D545. They have allowed to assess the influence of a ripple load near the stress threshold defined by the previous constant load tests.

Propagation tests. The tests were performed firstly to determine the influence of the main mechanical and metallurgical parameters on the CGRs, secondly to assess any influence of a periodic unloading-reloading procedure (R = 0.7, frequency 2.8 10-4 Hz, hold time 57 min) on the CGRs, with regard to the CGRs measured with a pure static loading. CT specimens from both welds were used, thickness 15 mm, pre-cracked by fatigue in air at Kfmax less than 15 MPam. They were cut in the welds according to the TL orientation (fatigue pre-crack perpendicular to dendrites in a plane containing the axis of the dendrites) and the TS orientation (fatigue pre-crack in the direction of the axis of the dendrites, see Figure 1). They have allowed one to assess the influence of the orientation on the CGRs. The results were compared to previous results obtained both on CT, DCB and WOL specimens [2,5].

Figure 1. Orientation of the CT specimens in the welds Environment and facilities The basic environment contained 1000 ppm B as boric acid and 2 ppm Li as LiOH,H2O, with 30 cc/kg of hydrogen. The control of hydrogen depended on the facility and the temperature.

Initiation tests. They were conducted in static autoclaves at 360°C, as shown in Figure 2 and Figure 3. The SSRT facility was described with details in [3] (Figure 2), deaeration was performed by evaporation of 20% of the water volume at 125°C, then hydrogen was introduced and controlled by a AgPd thimble at 360°C.

Figure 2. Autoclave for SSRTs at 360°C Constant load and ripple load tests were performed in a static autoclave placed in the load line of a creep machine. The load was applied by dead weights to reach the maximum stress level (350 or 380 MPa), a part (10%) of the load was applied by a volume of water admitted in a tank installed on the dead weights, at regular intervals with a constant flow rate in order to reach the desired trapezoidal loading (rising time 90 s, hold time 57 mn).

Figure 3. Facility for SCC tests with constant load or periodic unloading-reloading Propagation tests. They were performed at 325°C in an autoclave equipped with a load line containing 4 CT specimens, mounted on a creep machine in which the load was applied by using dead weights (static loading) or an air-jack (trapezoidal loading, rising time 90 s, hold time 57 mn), Figure 4. The autoclave was constantly fed in primary water with a Cormet loop to ensure the required water chemistry (conductivity and oxygen control) and hydrogen was monitored by an Orbisphere sensor to meet a content of 30 cc/kg.

558

Figure 4. Facility for SCC CGRs measurements with constant load or periodic unloading-reloading Evaluation of the crack depth Initiation tests. During the test, a LVDT sensor allowed a record of the elongation of the load line. The crack extension was assessed by examination of fracture surface (optical and/or scanning electron microscope). For unbroken specimen, the examination of the surface was realized optically.

Propagation tests. The true extension of the crack was performed on the fracture surface (optical and/or scanning electron microscope), along the longest crack branch in the case of macrobranching) : it has provided maximum value ('amax), and mean value ('amean) of the crack depth, in the latter case after measurement on 15 equally-distributed points of the crack front.

The crack growth rate (CGRmax or CGRmean) was deduced from the crack depth and from the duration of the test (generally 1300 h to 2000 h).

Results Initiation of SCC Threshold stress Vth. From previous constant load tests performed at 350°C on tensile specimens (EDF and CEA) and capsules (Framatome ANP) [1], it was deduced that for reasonable durations of tests, the stress threshold Vth for PWSCC of alloy 182 was near 350 MPa. Since that time, this assumption was reinforced by a rupture by SCC observed after 23771 h at 350 MPa on weld metal D 545 at 350°C [4]. Assuming an activation energy of 185 kJ/mol, it corresponded to a time for failure next to 13700 h at 360°C, which will be the reference result in the followings.

Strain threshold. In order to improve the use of such data for plant components, the threshold strain was investigated on both welds using interrupted SSRTs (5 10-8 s-1). Preliminary tests on weld metal D545 have demonstrated that no cracking occurred after a (plastic) elongation of 0.9%and a rupture by SCC was noted for an elongation ranging between 3.4% and 4.9%. Complementary data were obtained on weld metal D1054, without any cracking for 1.0 and 1.25% elongation, while a very significant cracking was noted for 1.5% elongation (Figure 5). It meant that the threshold strain ranged between 1.25 and 1.5%, which was roughly consistent with the threshold stress (350 MPa), due to the extremely severe SSRTs. For these reasons, this threshold appeared slightly lower than the value of 2% determined from capsules tests in [1].

Figure 5. SC crack after an interrupted SSRT at 1.5% elongation at 360°C (weld D 1054)

Influence of a ripple loading on Vth. Finally, the influence of a ripple loading (R = 0.9, 2.8 10-4 Hz) on the stress threshold was investigated at 360°C on weld metal D545. At the maximum stress of 380 MPa, a significant crack occurred after 7372 h under a ripple loading but the result on a pure static loading at 380 MPa was not available at that time. At the maximum stress of 350 MPa, a rupture occurred by SCC after 14034 h under a ripple loading (Figure 6). This result was very similar to the reference result obtained with a pure static loading at 350 MPa (13700 h). Despite the limited number of available results, a ripple loading did not seem to induce a significant influence on the stress threshold of alloy 182. A definitive conclusion will be drawn with regard to the influence of a ripple loading at the end of the test at 380 MPa.

Figure 6. Fracture surface of specimen in weld metal D 1054 after a ripple loading at R = 0.9, 2.8 10-4 Hz at 360°C Propagation of SCC Fracture surface examination. Fracture surface of PWSCC on weld metal 182 appeared intergranular / interdendritic. The crack front was uneven and a strong macro-branching at 30-45° was observed for the TL-orientation (Figure 7), while the crack front was most regular and the crack propagated roughly in the plane of 559

the fatigue pre-crack for the TS-orientation (the direction of the axis of the dendrites), Figure 8.

Figure 7. SCC at 325°C : fracture surface in the TL orientation (specimen TL9, D1054 AW, 41.4 MPam, R = 0.7)

Re-assessment of previous results : da/dt versus K. Previous CGR data obtained by EDF, CEA and ETH-Zürich [5] with a static loading were reassessed on weld metal D545, in the TL orientation :

x

different loading modes have been applied (constant load (CT) or constant deflection (WOL or DCB)).

Consequently, the initial K values were re-calculated taking into account the anisothermal stress relaxation between 20°C and 325°C for constant deflection tests, roughly 18%.

Figure 8. SCC at 325°C : fracture surface in the TS orientation (specimen 1471-TS1, D1054 AW, 22.9 MPam, R = 0.7) x

CGRs measured between 310°C and 330°C were considered and re-calculated at 325°C using an activation energy of 130 kJ/mol. This could be a reason for the typical behaviour of the 3 CEA results within a narrow range of K values (2 values at 330°C and one value at 310°C).

The mean curve for the maximum values of CGR of weld D545 as a function of the initial value of K became :

(da/d)max = 0.5114.(K-9)0.42 using K values ranging between 18 and 41 MPam (Figure 9).

This K-dependence was not so far from the law reported in the case of alloy 600, with the exponent between 0.3 and 0.5, see [6].

Significant data are required to better define the curve at low K values.

D545 & D1054 - AW - TL - static 0,01 0,10 1,00 10,00 0

10 20 30 40 50 60 K (MPa.m0.5)

CGRmax325 (E-10 m/s ETH D545 AW TL EDF D545 AW TL CEA D545 AW TL CGRmax325 D545 AW TL EDF D1054 AW TL D545 : CGRmax325 = 0,5114(K-9)^0,42 Figure 9. Mean curve of the max. CGRs as a function of K on alloy 182 (weld D545) in the TL orientation at 325°C The new results obtained in the TL orientation with weld D1054 were in close agreement with these data (Figure 9) and demonstrated a good reproducibility for tests performed at the same K-value. With the available data, the proposed law for PWSCC of 182 was : (da/d)max = D(material).(K-9)0.4.

Influence of the orientation. Though true crack lengths have been considered in the case of macro-branching (TL) or straight propagation in the direction of the axis of dendrites (TS), the present results demonstrated that the orientation induced significant changes in CGRs. The CGRs measured in the TS orientation were generally higher than in the TL orientation : 2 times for D545 with a static loading, 2.5 to 3 times for D1054 with a trapezoidal loading at R = 0.7.

Influence of a trapezoidal loading (R = 0.7). The influence of a trapezoidal loading on the CGRs was examined on both welds in the TL and TS orientations.

560

On weld D545, the data from Westinghouse [7] obtained with a trapezoidal loading (R = 0.7) were compared to the above-mentioned reference curve for a static loading, in the TL orientation. It could be concluded that no significant influence of the loading procedure was observed in the TL orientation, while the CGRs obtained with a trapezoidal loading in the TS orientation were 5 to 7 times higher than the reference (TL) curve in static loading (Figure 10).

D545, Static / dynamic 0,01 0,10 1,00 10,00 0

10 20 30 40 K (M Pa.m0.5)

CGRmax325 (E-10 m/s)

West A W TS R = 0.7 West A W TL R = 0.7 CGRmax325 AW TL s tatic CGRmax325 = 0,5114(K-9) ^0,42 Figure 10. CGRs on weld D545 : comparison between static and trapezoidal loadings On weld D1054, the CGRs were also very similar in static and in trapezoidal loading in the TL orientation, slightly higher than for weld D545 (Figure 11). By contrast, CGRs measured under a trapezoidal loading with the TS orientation were 3 to 5 times higher than those deduced from the reference (TL) curve.

Unfortunately, the data for the TS orientation with a pure static loading was not yet available, but it could be expected from the available CGRs on the two (similar) welds that data with a trapezoidal loading from [7] and from the present paper in the TS orientation, would be in good agreement (those from [7]

were the highest). Finally, it seemed that the influence of trapezoidal versus static loading on the SCC CGRs of alloy 182 depended on the orientation : no significant influence in the TL orientation, but a detrimental factor 2 to 7 could be observed with the TS orientation.

0,01 0,1 1

10 0

10 20 30 40 50 60 K (MPa.m0.5)

CGRmax325 (E-10 m/s)

D545 AW TL static D1054 AW TS R 0,7 D1054 AW TL R 0,7 D1054 AW TL static D545-TL D1054-TL D 1054-TS Figure 11. CGRs on weld D1054 : comparison between static and trapezoidal loadings Influence of the stress relief treatment. The CGRs obtained at 325°C with welds D545 under a static loading and D1054 under a trapezoidal loading in the stress-relief condition (610°C-20 h) were 2 times lower than the CGRs measured in the as-welded condition (at the limit of the scatterband for reproducibility but always in the lowest part).

Discussion Some comparisons could be realized with available data in relation to the influence of some major parameters on PWSCC of alloy 182.

561

Comparison of da/dt versus K curves between laboratories The K-dependence of CGRs was investigated by several laboratories. The Figure 12 compares the da/dt versus K curves proposed by EPRI and the preliminary reference curve proposed in this paper. It should be remind that EPRI-MRP 115 curve, with assumed KISCC = 0, was obtained as the best fit for a set of international average values from different welds, at 75%

of the cumulative distribution ((da/dt)ave325 = 1.5 10-12.K1.6), with static or trapezoidal loadings. By contrast, the proposed reference curve in this paper has assumed a KISCC value of 9 MPam, and the proposed law ((da/d)max = 0.5114.(K-9)0.42 for welds D545 or D1054) for a pure static loading was rather similar to the law proposed for alloy 600 [6].

0,01 0,1 1

10 0

10 20 30 40 50 60 K (MPa.m0.5)

CGRmax325 (E-10 m/s)

D545 EDF AW TL static D1054 AW TS R 0,7 D1054 AW TL R 0,7 D1054 AW TL static EPRI MRP 115 132 JNES AW TS static Studsvik lab data [8]

Ringhals field data [8]

Ringhals curve TS EDF on D545-TL CGRmax325 = 0,5114(K-9)0,42 D1054-TL D 1054-TS R 0,7 EPRI MRP 115 CGRave325 = 1,5 10-12 K1,6 Rhingals curve :

K < 25 : CGRmax325 = 7,22 10-23 K9,3 K > 25 : CGRmax325 = 6 10-10 Figure 12. Reference curves for weld metal 182 by EPRI, Ringhals and EDF from laboratory. Comparison with JNES results The Ringhals curve in the TS orientation presented a part with a dependence on K9.3 at K lower than 25 MPam, and a plateau regime at 6 10-10 m/s at K values higher than 25.1 MPam, which was consistent with EDF values on weld metal D1054 in the same (TS) orientation and with a trapezoidal loading. The results they obtained in laboratory under a static loading and from the field [8] were in close agreement with the curve suggested by EDF.

Trapezoidal loading versus static loading It should be noted that the data generated in this paper on weld D1054 in the TS orientation with a trapezoidal loading were slightly higher than the EPRI-curve which should account for this kind of loading. The differences in the shape and the level of EPRI and EDF curves could be the consequence of the differences in the loading procedures. It should be noted that Tsutsumi et al [9] did not recommend the use of a trapezoidal loading to measure CGRs on weld metal 132, unless hold times higher than 105 s would be used or unless pre-cracking of the specimen at the beginning of the test.

This kind of conclusion was still brought on alloy 600 in primary water : no change in CGRs was observed on very susceptible heats, while a significant increase of CGRs under a trapezoidal loading with respect to a static loading could be noted for heats with a low susceptibility to SCC [10]. Finally, it was not possible to explain the reason why no similar influence was observed for weld metal 182 in the TL orientation while a significant increase could be noted with a trapezoidal loading in the most sensitive (TS) orientation.

JNES data obtained in the TS orientation at 325°C (calculated from 340°C) under a pure static loading ranged between EPRI and EDF curves, they referred to a law similar to (K-9)1.16, Figure 12, [11].

Influence of the chemical composition of the weld In this paper, two weld metals from the same batch of electrodes and deposited in similar processes by two different manufacturers have provided very similar data, as expected.

However, the scatter on CGR results appeared to be somewhat larger on weld D545 than on weld D1054. This difference could come from laboratory-to-laboratory test procedures, since weld D545 has been tested by 3 different laboratories and weld D1054 by EDF only. On weld D545, the scatter was mainly observed within the K-range of 15-25 MPam, in which the influence of K on CGRs was significant, while EDF has investigated a broader range of K (20-42 MPam) on weld D1054, for which the influence of K was the lowest.

Besides, some CGRs measured at 330°C by CEA on 3 other welds in alloy 182 with C contents ranging from 0.022% to 0.089%, and Si contents from 0.27% to 0.79% were found to be somewhat similar : calculated values of the CGRmax at 325°C were 1.4 to 1.5 10-10 m/s for high or low C low Si weld, 1.8 to 562

2.35 10-10 m/s for the high C high Si weld, and 0.75 to 4.5 10-10 m/s on weld D545 [5].

Finally, weld to weld variability led to a factor 3 on the CGRs for the same TL orientation.

Influence of the stress relief-treatment CGRs measured on weld metal D545 or D1054 in the SR conditions were generally 2 to 3.5 times lower than in the AW conditions in most of the laboratories (EDF, CEA, ETH-Zürich) at 320-330°C, just above the limit for significance according to the reproducibility factor of 2 for this kind of tests. This slight beneficial effect of SR treatment at 610°C could not be explained by any significant change in the microstructure or density of dislocations [12] but could rely on some contribution of residual stresses to the true K level in the case of the AW condition. It should be noted that a significant beneficial effect of SR was noted by [13] at 290°C (factor 3 to 10 in the TL orientation).

For comparison, some significant beneficial influence of the SR treatment was observed on the resistance to the initiation of SCC in the case of highly susceptible materials (high C, high Si), the effect was limited in the case of materials with the lowest susceptibility to SCC [1].

Finally, the root causes of the major beneficial influence of SR treatment in plants rely partly on the reduction in the residual stresses which could reach 50% of the initial stresses and, in the case of initiation, on a possible recristallization of the surface cold worked layer. The intrinsic influence of SR treatment on the initiation of SCC is low on specimens with low C and Si.

Conclusion The initiation and growth of PWSCC of alloy 182 were investigated with regard to the influence of the loading parameters.

The stress threshold for initiation of SCC at 360°C was found close to 350 MPa in constant load tests. With the limited available data, it seemed that a ripple loading did not modify this value. The strain threshold was determined by interrupted SSRTs : SCC occurred for an elongation between 1.25 and 1.5%.

Preliminary data of crack growth rates (CGR) at 325°C provided a dependence on (K-9)0.4 for the average values of the CGRmax.

The strongly oriented microstructure led to faster (x 2 to 3)

CGRs in the direction of the dendrites (TS) than in the perpendicular direction (TL). The influence on the CGRs of a trapezoidal loading (R = 0.7, 2.8 10-4 Hz) with regard to a pure static loading depended on the orientation : no significant modification in the direction perpendicular to the dendrites (TL),

but some significant increase (x 2 to 7) could be observed in the direction of the dendrites (TS). A stress relief treatment (610°C) could lower the CGRs by a factor 2 to 3.5 with respect to the as-welded condition. The chemical composition of the weld metal could lead to some variability in the CGRs (factor 3).

Work is still in progress to assess the influence of temperature and chemical composition of the weld on the stress threshold for the initiation of SCC. For propagation, the influence of low values of K (less than 20 MPam) on the da/dt versus K curve, temperature, stress relief treatment and chemical composition of the weld metal will be investigated.

References 1

C. Amzallag et al, Stress Corrosion Life Assessment of 182 and 82 Welds used in PWR Components (10th International Conference on Environmental Degradation in Nuclear Power Systems - Water Reactors, South Lake Tahoe (NV), august 2001) 2 C. Amzallag et al, Stress Corrosion Life Experience of 182 and 82 welds in French PWRs (Contribution of Materials Investigation to the Resolution of Problems Encountered on Pressurized Water Reactors, Fontevraud V, France, September 2002) 3 Francois Vaillant, Chedly Braham, Jean-Marie Gras,

<< Etude de la corrosion sous contrainte en milieu primaire des alliages X-750 et X-750 modifié 19% de chrome laide de lessai de traction lente. Influence du traitement thermique >>,

(EDF report HT-45/PV D 736A), january 1990)

[4] << La corrosion sous contrainte des alliages de nickel dans leau haute température >>, F. Vaillant et al Editors, Reference book, (EDF report HT-29/04/077/A, 2005)

5.

S. Le Hong et al, Measurements of Stress Corrosion Cracking Growth Rates in Weld Alloys 182 in Primary Water of PWR (10th International Conference of Environmental Degradation of Materials in Nuclear Power Systems - Water Reactors, South Lake Tahoe (NV), august 2001) 6 F. VAILLANT et al, Crack Growth Rates in Thick Materials of Alloy 600 and Weld Metals of Alloy 182 in Laboratory Primary Water. Comparison with Field Experience (Contribution of Materials Investigation to the Resolution of Problems Encountered on Pressurized Water

Reactors, Fontevraud V, France, September 2002, pp 107-116) 7.

John Hickling, << Materials Reliability Program. Crack Growth Rates for Evaluating Primary Water Stress Corrosion Cracking (PWSCC) of Alloy 82, 182 and 132 welds (MRP-115) (EPRI report 1006696, November 2004) 8 A. Jenssen et al, A Swedish Perspective on PWSCC of Alloy 182 (2005 International PWSCC of Alloy 600 Conference and Exhibit Show, Hyatt Regency Tamaya Resort, Santa Ana Pueblo (NM), 7-10 March 2005) 9 K. Tsutsumi et al, SCC Growth Rate of Nickel based Alloy 132 Weld Metal in PWR Primary Water (11th International Conference on Environmental Degradation in Nuclear Power Systems - Water Reactors, Stevenson (WA), august 2003)

10. F. Vaillant et al, Influence of a Cyclic Loading on Crack Growth Rates of Alloy 600 in Primary Environment : An Overview (11th International Conference of Environmental Degradation of Materials in Nuclear Power Systems - Water Reactors, Stevenson (WA), august 2003) 11 Y. Yamamoto et al (JNES), Outline of Evaluation Technology for SCC Growth of Ni Base Alloys (NiSCC)

Project in Japan and Current Results in PWR Environment 563

(2005 International PWSCC of Alloy 600 Conference and Exhibit Show, Hyatt Regency Tamaya Resort, Santa Ana Pueblo (NM), 7-10 March 2005) 12 C. Cayron et al, << Etude par microscopie électronique en transmission de la microstructure de différentes nuances dalliages déposés 182 >> (Technical report CEA DTEN N° 127/2001) 13 Ruth Magdowski, Markus O. Speidel, << Stress Corrosion Crack Growth of Various Materials Exposed to Simulated PWR Water>> (Technical report ETH-Zürich, contract ND 3367-RE, april 1997) 564

Session Name: Ni-Based Alloys - I, II Session Day/Time: Thursday 8/18, 8am - noon Influence of a Cyclic Loading on the Initiation and Propagation of PWSCC in Weld Metal 182 Presenter: Thierry Couvant Name of Person Asking Question: John Hickling Affiliation of Person Asking Question: EPRI Question: You mentioned the extensive macrobranching in the T-L samples. Did you find any difference in the extent of this crack branching between specimens under static constant load and those exposed with periodic unloading?

Response: We didnt notice any difference. Additional examinations could be performed to precise this point.

Name of Person Asking Question: Jun Peng Affiliation of Person Asking Question: Babcock & Wilcox Company Question: Whereby do you check the effects of a spike in cyclic loading on da/dt (Crack Growth Rate)?

Response: No we dont.

565

Retrieved from "https://kanterella.com/w/index.php?title=ML15329A273&oldid=5179769"