Radiation response of the overlay cladding from the decommissioned WWER-440 Greifswald Unit 4 reactor pressure vessel

Nuclear Engineering and Design 286 (2015) 227–236

Contents lists available at ScienceDirect

Nuclear Engineering and Design journal homepage: www.elsevier.com/locate/nucengdes

Radiation response of the overlay cladding from the decommissioned WWER-440 Greifswald Unit 4 reactor pressure vessel Hans-Werner Viehrig ∗ , Eberhard Altstadt, Mario Houska Helmholtz-Zentrum Dresden-Rossendorf (HZDR), Bautzner Landstrasse 400, D-01328 Dresden, Germany

h i g h l i g h t s • • • • •

Tensile and crack extension testing of the austenitic overlay cladding from decommissioned reactor pressure vessels. Engineering crack initiation fracture toughness values according to ASTM E1820 were evaluated. Due to the inhomogeneous structure of the welded overlay cladding a significant scatter was observed in the initiation values. The J–a values show discontinuities caused by fast crack propagation because of low tearing strength or crack jumps. The measured fracture toughness values indicate that the cladding would remain intact during pressurized thermal shock transient.

a r t i c l e

i n f o

Article history: Received 4 November 2014 Received in revised form 24 February 2015 Accepted 25 February 2015

a b s t r a c t The Results of tensile and crack extension testing conducted on irradiated austenitic overlay cladding material are presented. The specimens were machined from three trepans sampled from the decommissioned WWER-440/V-230 reactor pressure vessel of the nuclear power plant Greifswald Unit 4. Crack extension curves were measured with Charpy size SE(B) specimens using the unloading compliance technique according to ASTM E1820-11 at different temperatures. Provisional crack initiation fracture toughness values JQ and KJQ are determined. The highest KJQ values were found in the temperature range from 20 to 80 ◦ C. A significant scatter was observed in the initiation values. This is due to the inhomogeneous structure of the welded overlay cladding. During the loading the crack moves through regions with different tearing strength, thus the J–a values show discontinuities caused by fast crack propagation or crack jumps. The comparison of the measured KJQ values and conservatively estimated stress intensity factors at an assumed surface crack shows that the cladding would remain intact during pressurized thermal shock transient. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Austenitic overlay cladding has been originally designed to protect the low alloyed RPV base and weld metals against corrosion. Usually, the overlay cladding remains out of consideration in the current RPV integrity assessment regulatory codes (Brumovsky et al., 1997; Gillemot et al., 2007). Some codes allow the inclusion of the overlay cladding on condition that its properties are known during the RPV lifetime (Brumovsky, 2003; Gillemot et al., 2007). The fracture behaviour of an operating nuclear RPV, particularly during certain overcooling transients, may strongly depend on the properties of the irradiated overlay cladding (Haggag et al., 1990). Investigations have shown that the integrity of the

∗ Corresponding author. Tel.: +49 351 2603246; fax: +49 351 2602205. E-mail address: [email protected] (H.-W. Viehrig). http://dx.doi.org/10.1016/j.nucengdes.2015.02.008 0029-5493/© 2015 Elsevier B.V. All rights reserved.

overlay cladding strongly influences the loading of a crack in the base or weld metal during an emergency PTS event. An intact overlay cladding clamps an under clad crack and results in a lower stress intensity factor KI at the crack tip. In contrast a surface crack is additionally loaded due to the tension stress in the overlay cladding caused by different thermal properties with respect to the base or weld metal (Abendroth and Altstadt, 2007). Therefore, knowledge about the properties of the overlay cladding, mainly of its fracture toughness, is also necessary for a precise evaluation of the RPV integrity during a PTS event. Overlay cladding material is currently not included in the RPV surveillance specimen programmes. Therefore, mechanical and fracture toughness data are not available from operated RPVs. In the literature there are test results from the overlay cladding investigated within the acceptance tests and irradiations experiments performed in research reactors. Results on the RPV overlay cladding and especially on their irradiation response are limited compared to base and weld metal. Irrespective

228

H.-W. Viehrig et al. / Nuclear Engineering and Design 286 (2015) 227–236

Nomenclature a ai a0 a0q bcc c COD E FEM JIc

JQ KI KJIc KJQ NPP Rp0.2 Rm RPV PTS s SE(B) SIF T UC WWER ˚ 

crack depth (small half axis of the ellipse) actual crack length determined by unloading compliance initial crack length initial crack length estimated according to ASTM E1820 with the ai values up to maximum load body centred cubic large half axes of the ellipse crack opening displacement Young’s modulus finite element method JQ becomes JIc when properly qualified against the criteria proposed in the test standard procedures ASTM E1820 J integral a at 0.2 mm crack extension excluding blunting (according to ASTM E1820) stress intensity factor JIc based fracture toughness according to ASTM E1820 JQ based fracture toughness according to ASTM E1820 nuclear power plant proof strength at 0.2% plastic deformation ultimate tensile strength reactor pressure vessel pressurized thermal shock thickness of the cladding single edge bend specimen stress intensity factor temperature in K unloading compliance technique Russian type pressurized water reactor (water operated and moderated) neutron fluence uniform hoop stress

of the various cladding technologies the following main features can be summarized (Haggag et al., 1990; Alekseenkov et al., 1997; Brumovsky et al., 1997; Timofeev and Karzov, 2006; Gillemot et al., 2007; Margolin et al., 2010) • The austenitic cladding shows a ductile-to-brittle transition behaviour similar to bcc steels because of the weld structure containing 3–10% of ␦ ferrite. Specimens which failed in brittle mode contain areas of cleavage associated with the ␦ ferrite phase and ferrite–austenite interphases fractured by interphase separation. • The increase of the ultimate tensile strength with decreasing temperature is more pronounced than it is for the yield strength. • The crack initiation fracture toughness vs. temperature curve runs over a peak at room temperature, with unstable crack propagation at low temperature and low tearing strength above 200 ◦ C. • There is a susceptibility to neutron irradiation, which leads to an increase of the yield and ultimate strength, the appearance of brittle crack extension and a decrease of the tearing strength. • Depending on the test temperatures, unstable failure of specimens or crack jumps after preceding ductile crack extension were observed. Margolin et al. (2010) suggests the introduction of a limit temperature above which intercrystalline fracture and brittle crack jumps are not observed. In addition, a critical numerical jump fracture toughness value, Jc , for the brittle crack initiation or crack extension was specified with 65 kJ/m2 for WWER RPVs.

• The progression of crack extension curves measured on the cladding from WWER-440 and WWER-1000 RPVs exhibits discontinuities because of the heterogeneous weld structure (Margolin et al., 2010).

Low fracture toughness and the occurrence of brittle failure at elevated temperatures raise concerns over the potential adverse impact of the austenitic overlay cladding on the integrity of the RPV during a PTS transient. The growth of a surface flaw can lead to the failure of the cladding and significantly increase the probability of vessel failure (Haggag et al., 1990; Abendroth and Altstadt, 2007; Margolin et al., 2010). The initiation of brittle crack extension in the cladding below a threshold fracture toughness value has to be avoided for temperatures above 20 ◦ C (Margolin et al., 2010). This threshold value has to be specified depending on the operation and emergency conditions of the NPP to be assessed. Therefore knowledge about the fracture toughness within the temperature range of a PTS transient and at operation temperature of the RPV are of high importance. The most realistic evaluation of the toughness response of RPV overlay cladding to irradiation may be achieved by directly studying RPV wall samples taken from decommissioned RPVs. Such a possibility is available with the investigation of samples taken from the RPVs of the Greifswald NPP. Four WWER-440/V-230 nuclear reactors representing the first generation of this reactor type were operated between 11 and 15 years and were decommissioned in 1990. The RPVs were designed and manufactured in the former Soviet Union at the end of the 1960s and the beginning of the 1970s. The operation history, the expected neutron fluences and the material conditions of the units 1–4 are presented elsewhere (Konheiser et al., 2006; Viehrig et al., 2009, 2010, 2012a,b). While the RPVs of the units 1 and 2 do not contain an austenitic anticorrosive overlay cladding, the RPVs of the units 3 and 4 are cladded. Trepans were extracted from the RPV of the units 1, 2 and 4. The description of the trepanning device and the sampling procedure and the results of investigations on the base metals and the weld materials of the RPVs were published elsewhere (Konheiser et al., 2006; Viehrig et al., 2009, 2010, 2012a,b, 2015). This paper presents the test results measured on the austenitic overlay cladding of trepans sampled from different locations within the reactor core region of the Greifswald Unit 4 RPV, which was shut down after 11 operating cycles. The main focus is on the measurement of the crack extension curves and the evaluation of initiation fracture toughness values according to the test standard ASTM E1820-11.

2. Material and specimens The RPV of Greifswald Unit 4 is overlay cladded by austenitic stainless materials using automatic submerged welding methods. The overlay cladding is welded with several passes using ribbon material of different composition, and hence represents a multilayer structure. Following requirements have to be met (Alekseenkov et al., 1997; Brumovsky et al., 1997; Timofeev and Karzov, 2006; Gillemot et al., 2007; Margolin et al., 2010):

- Prevention of hot cracks and a martensitic structure in the first layer diluted with base metal, - Prevention of the overlay cladding embrittlement due to the formation of brittle secondary carbides and intermetallic phases, - High resistance against intercrystalline corrosion, and - Prevention of the embrittlement in adjoining zones of base metal because of high temperature thermo strained cycles during the welding of the cladding.

H.-W. Viehrig et al. / Nuclear Engineering and Design 286 (2015) 227–236

229

Fig. 1. Picture and cutting scheme of trepan 4-1.

To meet these requirements, the cladding was welded in twolayers overlay with three welding passes (Alekseenkov et al., 1997; Timofeev and Karzov, 2006): - First layer (one pass) with a thickness of 3–4 mm: ribbon of Sv07Cr25Ni13 and OF-10 flux, and - Second layer (two passes) with a thickness of 6–9 mm: ribbon of Sv-08Cr19Ni10Mn2Nb and OF-10 flux. Consequently, the total thickness of the cladding amounts to 8–10 mm after mechanical machining. The welding of the cladding and the assembling of the forged rings that form the RPV were performed simultaneously, with the aim to reduce the number of heat treatments. Within these technological processes the cladding in the core region of the RPV, which includes the forged base metal ring 0.3.1 and the welding seam SN0.1.4, was tempered five times in the temperature range 660–675 ◦ C with a total duration of 50 h (Timofeev and Karzov, 2006). The overlay cladding contains up to 8% ␦-ferrite to avoid hot cracks. The chemical composition of the overlay cladding is summarized in Table 1. The overlay cladding materials investigated were extracted from the decommissioned WWER-440/V-230 Unit 4 RPV of the Greifswald NPP. The trepans were taken from the circumferential beltline welding seam SN0.1.4 (trepans 4-4 and 4-6) and from the forged base metal ring 0.3.1 in the region of the maximum neutron flux (trepan 4-1). The trepans were cut into discs over the RPV wall thickness using a travelling wire electric discharge machine. The cutting scheme is exemplarily depicted for the trepan 4-1 in Fig. 1. As the first work step, a vertical cut was made to transfer the former orientation of the trepan in the RPV to the discs to be cut.

In the following step the curvature at the inner side of the trepan was cut to get a plane reference surface followed by the cutting of the discs as shown in Fig. 1. The first disc towards the inner RPV wall contains the overlay cladding. From this disc 14 Charpy size SE(B) specimens in L-S orientation (specimen axis along vessel circumferential direction and crack growth direction in the vessel radial direction) according to ASTM E1823 were machined. In addition, sub-sized rectangular flat tensile specimens were machined from broken halves of the tested Charpy size SE(B) specimens as schematically depicted for the trepan 4-1 in Fig. 2. From one half 2 specimens from the first layer and 4 specimens from the second layer of the overlay cladding were machined. The orientation of the tensile specimens is L (specimen axis along vessel circumferential direction) according to ASTM E1823. The neutron flux through the RPV was calculated with TRAMO (Barz et al., 1998), which is a multi-group Monte Carlo code for neutron and gamma transport calculations. Table 2 summarizes the accumulated neutron fluences in the centre of the overlay cladding. The accumulated neutron fluences in Table 2 correspond to about one quarter of the designed fluence for 30 years operation (IAEATECDOC-1442, 2005).

3. Testing and evaluation The irradiated specimens were tested remote controlled with a servohydraulic testing system installed in a hot cell laboratory. Tensile testing was conducted with the sub-sized rectangular flat specimens as displayed in Fig. 2 in a temperature range of −75 to 270 ◦ C. The specimens were loaded with an actuator speed of 0.25 mm/min. Because of the situation in the hot cell and the small

230

H.-W. Viehrig et al. / Nuclear Engineering and Design 286 (2015) 227–236

Table 1 Chemical composition of the cladding materials (wt.%). Material Sv-07Cr25Ni13a Sv-08Cr19Ni10Mn2Nbb Sv-08Cr19Ni10Mn2Nb (trepan 4-1)c Sv-08Cr19Ni10Mn2Nb (trepan 4-6)c a b c

C <0.09 <0.10 – –

Cr

Ni

Mn

Cu

P

Nb

23.0–26.5 17.5–20.5 18.7 18.3

11.0–14.0 8.5–11.0 10.20 9.85

0.8–2.0 1.3–2.5 1.42 1.37

– – 0.24 0.23

<0.03 <0.03 0.031 0.026

0.65 0.62

1st layer. 2nd layer according to the specification (Brumovsky et al., 1997; Timofeev and Karzov, 2006). Chemical analysis of the material from the 2nd layer.

Table 2 Positions and accumulated neutron fluences of the investigated cladding materials (Viehrig et al., 2012a,b; Rindelhardt et al., 2009). Code of the trepan and disc

4-1.1 4-4.1 4-6.1

Location of the trepan in the Unit 4 RPV

Distance of the crack tip location from inner surface mm

˚ (E > 0.5 MeV) 1018 n/cm2

Base metal ring no. 0.3.1

9.3 7.8 7.5

53.73 41.04 41.06

Beltline welding seam no. SN0.4.1

size of the specimens, the travel was measured with an extensometer attached on the clamping device. Crack extension curves were measured with Charpy size SE(B) specimens using the unloading compliance technique according to ASTM E1820-11. Following test conditions for the UC testing were applied: • • • • • • • • • • •

Charpy size SE(B) specimens, 20% side-grooving, a/W = 0.5, Radius of the side grooves: 0.4 mm, Integral knifes for COD-Clip-On extensometer, Pre-cracking with a servohydraulic testing machine (Kmax : √ √ 25 MPa m, Kend : 15 MPa m), Test temperature range: −100 ◦ C to 270 ◦ C, Loading velocity: 0.2 mm/min, Partial unloading: 25% of the actual load, Relaxation time: 30 s, Distance between the unloading sequences (COD): 0.04 mm, Unload and reload ramp rate: 100 N/s, and End of test criteria: 2 mm crack extension, 4 mm COD and when the load reached a value 50% below of the maximum load, whatever was exceeded first.

The recorded data of the load, the deflection and COD were evaluated according to the test standard ASTM E1820-11. Accordingly,

JR curves (J integral vs. a) were determined. Engineering crack initiation fracture toughness values, JQ , were evaluated from the JR curves. The majority of the crack initiation JQ values could not be qualified as JIc , because they did not adhere to the validity criteria given in the test standard ASTM E1820-11. The JQ values were converted into their equivalents in units of stress intensity factors KJQ . For some tests all J–a data between the 0.15 mm and 1.5 mm exclusion line could not be used for the fitting of the power law regression line because of their unsteady progression. Otherwise the regression line would intersect the 0.2 mm offset line away from the data points, which results in an unrealistic JQ value. 4. Results and discussion Fig. 3 shows the yield and ultimate tensile strength vs. test temperature of the investigated overlay cladding materials. As depicted in Fig. 3 there is a minor difference between both layers, in terms of the development of the yield and ultimate tensile strength. On average the values of the layer 1 lie above layer 2. Eqs. (1)–(4) describe the strength vs. temperature behaviour of the 1st and the 2nd layer of the overlay cladding: 1st layer :

Rp0.2 = 629.8 · e−0.00108·T in MPa

(1)

1st layer :

Rm = 417.0 + 1830.79 · e−0.00712·T in MPa

(2)

2nd layer :

Fig. 2. Scheme of the machining of sub-size tensile specimens from broken halves of Charpy size SE(B) specimens.

Rp0.2 = 290 + 410.2 · e−0.0041·T in MPa

(3)

Fig. 3. Tensile strength characteristics of the cladding layers 1 and 2 vs. temperature.

H.-W. Viehrig et al. / Nuclear Engineering and Design 286 (2015) 227–236

231

Fig. 4. Overview of the structure of the investigated overlay cladding from the individual trepans and the location of the crack tip.

2nd layer :

Rm = 415.2 + 3402 · e−0.009·T in MPa

(4)

The room temperature values are calculated as follows: • 1st layer: Rp0.2 and Rm are 460 MPa and 645 MPa, respectively, • 2nd layer: Rp0.2 and Rm are 414 MPa and 623 MPa, respectively. These values are about 30% higher than the yield strength of 314 MPa and ultimate tensile strength of 490 MPa for the initial condition of the WWER-440 overlay cladding as reported by Timofeev and Karzov (2006).

As mentioned before, a disc with a thickness between 1.5 and 4 mm was cut from the inner side of the of the trepan, therefore the later crack tips in the SE(B) specimen were at different locations in the cladding for the individual trepans. Fig. 4 gives an overview about the crack tip locations in the structure of the investigated overlay claddings. In the trepans 4-4 and 4-6 from the welding seam SN0.1.4, the crack tips are located in the 2nd layer welded with the Sv-08Cr19Ni10Mn2Nb ribbon, and the cracks moves towards the RPV inner wall surface during the fracture mechanics testing. The crack tips in the specimens from trepan 4-1 are located in the 1st

Table 3 Summary of the crack initiation J values, JQ , and the converted fracture toughness values KJQ of the investigated overlay cladding from the welding seam SN0.1.4 of the Greifswald Unit 4 RPV (trepans 4-4 and 4-6). Specimen code

T (◦ C)

JQ (kJ/m2 )

√ KJQ (MPa m)

4-4.1.1 4-4.1.2 4-4.1.3 4-4.1.4 4-4.1.5 4-4.1.6 4-4.1.7 4-4.1.8 4-4.1.9 4-4.1.10 4-4.1.11 4-4.1.12 4-4.1.13 4-4.1.14

22 50 100 150 200 268 22 50 100 150 200 270 0 0

92 152 162 111 63 119 117 104 69 92 98 76 125 109

141 181 187 154 116 160 159 149 122 141 145 128 164 153

a

Unstable failure of the specimen Jc and KJc .

Specimen code 4-6.1.2 4-6.1.3 4-6.1.4 4-6.1.5 4-6.1.6 4-6.1.7 4-6.1.8 4-6.1.9 4-6.1.10 4-6.1.11 4-6.1.12 4-6.1.13 4-6.1.14

T (◦ C) −25 −50 −75 −25 0 −37 −100 100 50 150 25 200 270

JQ (kJ/m2 )

√ KJQ (MPa m)

171 117 49 103 133 102 40a 104 170 126 188 121 93

192 159 103 149 169 148 92a 150 191 165 201 162 142

232

H.-W. Viehrig et al. / Nuclear Engineering and Design 286 (2015) 227–236

Table 4 Summary of the crack initiation J values, JQ , and the converted fracture toughness values KJQ of the investigated overlay cladding from the forged base metal ring 0.3.1 of the Greifswald Unit 4 RPV (trepan 4-1). Specimen code

T (◦ C)

4-1.1.1 4-1.1.2 4-1-1-3 4-1.1.4 4-1.1.5 4-1.1.6 4-1.1.7 4-1.1.8 4-1.1.9 4-1.1.10 4-1.1.11 4-1.1.12 4-1.1.13 4-1.1.14

25 −25 −50 0 −75 100 200 270 50 75 150 250 125 50

a

JQ (kJ/m2 ) 171 117 76 115 33a 243 116 96 125 307 179 154 121 131

√ KJQ (MPa m) 192 159 128 157 84a 229 158 144 164 257 196 182 161 168

Unstable failure of the specimen Jc and KJc .

layer of the cladding so that the cracks will propagate within that layer from the inner RPV wall surface towards the base metal. The reason for this change of the direction was to maintain the crack propagation in the same material during the testing. Otherwise the crack would move from the first to the second layer. Generally, it has to be taken into account that the locations of the crack tips vary within one disc because of the curvature of the RPV and differences of the fatigue crack length. The difference can amount up to 0.25 mm. Tables 3 and 4 summarize the provisional initiation J values, JQ , and the equivalent fracture toughness values, KJQ , evaluated according ASTM E1820-11 of the investigated overlay cladding materials. Figs. 5 and 6 depict the KJQ values vs. temperature. Generally, the scatter of the KJQ values is large, which is caused by the crack moving through the heterogeneous structure of the overlay cladding. Furthermore, the J–a data determined from the first partial unloading sequences tend to scatter, which has an impact on the evaluated crack initiation JQ values. The trend of the KJQ vs. temperature values can be approximated to a polynomial fit. Both figures show that the KJQ values increase with temperature to a maximum between 25 ◦ C and 80 ◦ C and decrease again towards higher temperatures. Above ambient temperature the fracture toughness √ values, KJQ , are ≥116 MPa m. Such progression of the initiation fracture toughness is also reported in the literature (Haggag et al., 1990; Margolin et al., 2010; Timofeev and Karzov, 2006). At ambi√ √ ent temperature KJQ values of 170 MPa m and 183 MPa m were approximated for the overlay cladding from the welding seam and the forged ring, respectively. The specimens from the cladding of

Fig. 5. Crack initiation fracture toughness values, KJQ , vs. temperature of the investigated overlay cladding from the welding seam SN0.1.4 of the Greifswald Unit 4 RPV.

Fig. 6. Crack initiation fracture toughness values, KJQ , vs. temperature of the investigated cladding from the forged base metal ring 0.3.1 of the Greifswald Unit 4 RPV.

trepans 4-1 and 4-6 fail unstably at −75 ◦ C and −100 ◦ C, respectively. Some specimens showed crack jumps or unstable failure after ductile crack extension within the 0.15 mm and 1.5 mm exclusion lines. In this case only data up to the crack jumps were used to fit the power law regression curve. In the summary two fracture mechanisms were observed: • Transcrystalline ductile fracture and • Intercrystalline fracture. Fig. 7 shows the J–a values of specimens tested at temperatures from room temperature till −100 ◦ C. Generally, the slopes of the J–a values increase with the test temperature which results in higher JQ values (Table 3). The transcrystalline ductile fracture appears as dimple fracture, whose size and shape is influenced by the structure of the cladding and the test temperature. Fig. 8a shows the dimples of the specimen 4-6.1.12 (Table 3) tested at 25 ◦ C in the vicinity of the fatigue crack front. The dimples change their shape and size (Fig. 8b) when J–a values reach the flat region of the graph as depicted in Fig. 7. The diameter of dimple in the area of low tearing strength is smaller compared to that in the area of

Fig. 7. J vs. a data and JR curves measured on specimens from the overlay cladding of the welding seam SN0.1.4 (trepan 4-6) of the Greifswald Unit 4 RPV in the temperature range from −100 ◦ C to 25 ◦ C.

H.-W. Viehrig et al. / Nuclear Engineering and Design 286 (2015) 227–236

233

Fig. 8. Fractured surfaces of the SE(B) specimen 4-6.1.12 (Table 3) (a) in the vicinity of the fatigue crack and (b) in the area of low tearing strength (Fig. 7).

Fig. 9. Intercrystalline fracture of the SE(B) specimen 4-6.1.8 (Table 3).

high tearing strength. In addition, the specimen 4-6.1.3 (Table 3) shows small jumps of intercrystalline cracks, whose initiators are identified as cleavage microcracks in brittle phases (␦-ferrite and ␴phase) located on crystallite boundaries (Margolin et al., 2010). The specimen 4-6.8 (Table 3) tested at −100 ◦ C failed completely by an intercrystalline fracture (Fig. 9). The increasing fracture toughness shown in the Figs. 5 and 6 is the result of the reducing intercrystalline fracture with increasing temperature and the reduction above 80 ◦ C is caused by the decreasing yield strength (Fig. 3). The location of the crack tip and the scatter of the J–a data are very important in terms of the evaluation of the initiation values JQ according to ASTM E1820-11. Slight differences in the progression of the J–a data within the range of the JR curve fit (between the 0.15 mm and 1.5 mm exclusion lines) can cause a large difference of the evaluated JQ values. Fig. 10 shows the J–a curves of two specimens from trepan 4-4 tested at 100 ◦ C as an example. Both specimens show a steep unsteady slope of the J–a data at the beginning which becomes more shallow with increasing crack extension. The large deviation of the JQ values results from the different progression of the J–a data at the beginning of the curves. Differences in the initial crack length (a0 ) can cause that the crack tips are located in different structures of the overlay

Fig. 10. J vs. a data and JR curves measured on specimens from the overlay cladding of the welding seam SN0.1.4 (trepan 4-4) of the Greifswald Unit 4 RPV.

234

H.-W. Viehrig et al. / Nuclear Engineering and Design 286 (2015) 227–236

Fig. 11. J vs. a data and JR curves measured on specimens from the overlay cladding of the welding seam SN0.1.4 (trepan 4-6) of the Greifswald Unit 4 RPV.

Fig. 13. J vs. a data and JR curves measured on specimens from the overlay cladding of the forged ring 0.3.1 (trepan 4-1) of the Greifswald Unit 4 RPV (filled symbols mark the UC sequence after that a crack jump occurred).

cladding, which also influences the course of the first J–a data and consequently the JQ values. The initial crack tip locations of both specimens in Fig. 10 differ by 0.11 mm. As shown in Fig. 10 the slope of the first J–a data of specimen 4-4.1.3 is steeper over a larger range compared to specimen 4-4.1.9., which results in the higher JQ value. The J–a data of many specimens exhibit an unsteady trend caused by the crack propagation through different structures. Fig. 11 shows examples from the overlay cladding of trepan 4-6. The J–a progressions of the specimens tested at −25 ◦ C differ completely. The scatter and the different slope of the first J–a data cause a large difference in the JQ values. Fig. 12 depicts the UC sequences of these specimens. The load vs. COD data of specimen 4-1.1.2 reveal a considerable discontinuity, which results in a large amount of crack extension (Fig. 11). This indicates that the crack runs through structures of low tearing strength in the overlay cladding. On the other hand, discontinuities in the J–a progression of specimen 4-6.1.5. (Fig. 11) are not obviously reflected in the UC sequences (Fig. 12). The majority of the test records of specimens from the trepans 4-4 and 4-6 (welding seam) tested below 0 ◦ C show large a steps between two UC sequences. Larger crack jumps which are indicated as load drops in the load vs. COD diagram occur for the cladding of the trepan 4-1 from the forged ring 0.3.1. The neutron fluence accumulated by the overlay cladding of this trepan is about 30% higher than that of the trepans

from the welding seam SN0.1.4 (Table 2). Furthermore the crack tip of the SE(B) specimens is located in the first layer of the cladding (Fig. 4). As shown in Fig. 6, the KJQ vs. temperature progression is in principle comparable to the overlay cladding from the weld metal trepans. However, this overlay cladding shows crack jumps during the loading above room temperature. In some cases strong load drops, caused by crack jumps, initiated the end of test criteria and finished the test after the following UC sequence. Fig. 13 shows the J–a curves of specimens tested in the temperature range from 50 ◦ C to 100 ◦ C. These specimen show crack jumps up to 0.4 mm after 0.8 mm ductile crack extension. At higher temperatures no crack jumps were observed till the end of test criteria was reached. The load drops connected with the crack jumps are clearly visible in the load vs. COD diagrams exemplarily depicted for the specimens tested at 50 ◦ C and 75 ◦ C in Fig. 14. The loading of both specimens was automatically stopped after the subsequent UC sequence as a predetermined end of test criteria was exceeded. Fig. 15 shows the transition from the ductile dimple fracture to the intercrystalline crack jump on the fractured surface of specimen 4-1.1.10. As shown in Figs. 5 and 6, low KJQ values were determined on specimens tested at 270 ◦ C, which is about the inlet water temperature of the WWER-440 reactors. At this test temperature the span of √ the KJQ values reaches from 128 to 160 MPa m (see Tables 3 and 4), in which the lower values were measured on the specimens from the overlay cladding of the welding seam SN0.1.4. The JR curves of

Fig. 12. Load vs. COD data of the specimens 4-6.1.2 and 4-6.1.5 from the overlay cladding of the welding seam SN0.1.4 (trepan 4-6) of the Greifswald Unit 4 RPV.

Fig. 14. Load vs. COD data of the specimens 4-1.1.9 and 4-1.1.10 from the overlay cladding of the forged ring 0.3.1 trepan 4-1 of the Greifswald Unit 4 RPV.

H.-W. Viehrig et al. / Nuclear Engineering and Design 286 (2015) 227–236

235

Table 5 SIFs in dependence on a/s calculated according to VERLIFE (Brumovsky, 2003) and (Dufresne, 1981). a/s

√ KI a (MPa m)

√ KI b (MPa m)

0.2 0.5 0.8

33.9 63.7 103.7

33.2 60.4 95.0

a b

According to VERLIFE (Brumovsky, 2003). According to Dufresne (1981).

5. Integrity assessment The integrity of the overlay cladding during a PTS scenario is assessed by a conservative estimation of the stress intensity factor (SIF) that could occur at a flaw. The leading scenario for the WWER440 is described by a stuck open pressurizer relief valve (which initiates the emergency core cooling system) followed by an inadvertent re-closure after 1 h (Pistora and Kral, 2003). In Abendroth and Altstadt (2007) and Viehrig et al. (2012a) fracture mechanics analyses of the RPV wall based on this scenario were performed. It was shown that the equivalent stress in the cladding, resulting from tension in axial and hoop direction, reaches the yield strength during the cold water injection. This is due to the larger thermal expansion coefficient of the cladding material compared to the ferritic RPV material. Therefore we assume the following conditions for the estimation of the SIF:

Fig. 15. Fractured surfaces of the SE(B) specimen 4-1.1.10 (Table 3) in the transition from ductile dimple fracture to the intercrystalline crack jump.

the specimens tested at 200 ◦ C are summarized in Fig. 16. This figure poses again the problem with the JR testing of overlay cladding material using the UC technique. The first J–a values measured with the first UC sequences strongly scatter which has an impact on the JQ values. As mentioned above the scatter is caused by the accuracy of the load and COD measurement and in addition by the tearing strength of the inhomogenous cladding structure in the vicinity of the crack tip. Therefore, the low JQ value of specimen 4-4.4.5 might represent a lower shelf value caused by the low tearing strength of the overlay cladding in the vicinity of the crack tip.

• Uniform temperature in the cladding: T = 60 ◦ C • Uniform hoop stress in the cladding:  = 408 MPa (corresponds to the yield strength of the cladding at T = 60 ◦ C) • Axially oriented semi-elliptical surface crack in the cladding with an aspect ratio a/c = 0.3 Two analytical formulae are used to calculate the SIF. The first one is according to the VERLIFE code (Brumovsky, 2003): KI =  ·

√

a·



2 − 0.82 · a/c



1 − (0.89 − 0.57

3

a/c) · (a/s)

3/2

3.25

(5)

and the second one according to a work of Dufresne (1981): KI =  ·

√



2

(a/s) a a · 1.14 − 0.48 · + 1.2 c 0.2 + 4.9 · (a/c)



.

(6)

Both equations are based on fits of FEM simulation results and calculate the SIF at the deepest point of the semi-elliptical crack. The SIFs calculated for different a/s ratios are listed in Table 5. The formula according to the VERLIFE code (Brumovsky, 2003) is more conservative. Even in the case of the deep crack (a/s = 0.8) the estimated SIF is well below the KJQ values measured in the temperature range of the PTS scenario (Tables 3 and 4). Hence, it can be stated that the investigated Greifswald Unit 4 RPV overly cladding containing a flaw would not have failed during a PTS event. 6. Conclusions

Fig. 16. J vs. a data and JR curves measured on specimens from the cladding of the forged ring 0.3.1 and welding seam SN0.1.4 trepans 4-1, 4-4 and 4-6 of the Greifswald Unit 4 RPV.

The investigation of the overlay cladding from the decommissioned first generation WWER-440 RPV of Greifswald Unit 4 contributes to the understanding in terms of its stability under real operation conditions. Greifswald Unit 4 was operated for 11 cycles and the RPV received about 25% of the designed neutron loading. As the initial strength and fracture toughness data were missing the radiation susceptibility of the investigated overlay cladding could not be assessed. Tensile strength data in the literature (Timofeev and Karzov, 2006) indicate an irradiation induced increase of the

236

H.-W. Viehrig et al. / Nuclear Engineering and Design 286 (2015) 227–236

yield strength by 30%. The main focus here was put on the measurement of J vs. a curves by using the unloading compliance technique and the evaluation of initiation fracture toughness values according to the test standard ASTM E1820-11. The fracture toughness is reduced by intercrystalline fracture at low temperature and the decreasing yield strength at high temperature. This results in a maximum of the initiation fracture toughness values, KJQ , in the temperature range between ambient temperature and 80 ◦ C. The significant scatter is caused by the crack moving through the heterogeneous structure of the overlay cladding. Furthermore, the J–a data determined from the first partial unloading sequences tend to scatter, which has a strong impact on the evaluated crack initiation JQ values. Crack jumps and unstable failure were found after 1 mm ductile crack extension above ambient temperatures and below −50 ◦ C, respectively. Above ambient temperature the lowest initial frac√ ture toughness values, KJQ , were determined with 144 MPa m and √ 116 MPa m for the overlay cladding from the forged base metal ring 0.3.1 and the beltline welding seam SN0.1.4, respectively. These values are above the stress intensity factors estimated for a postulated surface crack in the overlay cladding during a PTS transient. It can be stated that the overlay cladding of the Greifswald Unit 4 RPV would remain intact and increase confidence in the safety of the structural integrity of the RPV after the operation up to 25% of the designed neutron loading for 30 full power years. An increase of the neutron fluence can shift the range of unstable crack extension and the occurrence of crack jumps to lower fracture toughness values and higher temperatures, respectively, which might affect the integrity of the RPV. References Abendroth, M., Altstadt, E., 2007, June. Fracture Mechanical Analysis of a Thermal Shock Scenario for a VVER-440 RPV. Scientific Technical Reports of Forschungszentrum Dresden-Rossendorf FZD-474 (ISSN 1437-322X). Alekseenkov, N.N., Amaev, A., Gorynin, I., Nikolaev, V.A., 1997. Radiation Damage of Nuclear Power Plant Vessel Steels. American Nuclear Society, La Grange Park, IL, USA. ASTM E1820-11: Standard Test Method for Measurement of Fracture Toughness. Annual Book of ASTM Standards, vol. 03.01. ASTM International, West Conshohocken, PA. ASTM E1823-07: Standard Terminology Relating to Fatigue and Fracture Testing. Annual Book of ASTM Standards, vol. 03.01. ASTM International, West Conshohocken, PA. Barz, H.-U., Boehmer, B., Konheiser, J., Stephan, I., 1998. High-precision Monte Carlo calculations, experimental verification and adjustment of fluences in the pressure vessel cavity of a VVER-1000. In: Proceedings of the ANS Radiation

Protection and Shielding Division Topical Conference: Technologies for the New Century, vol. 1, Nashville, Tennessee, USA, April 19–23, 1998, pp. 447–454. Brumovsky, M., Novosad, P., Vacek, M., 1997. Cladding properties changes during operation. In: Proceedings of the IAEA Specialists’ Meeting on Irradiation Effects and Mitigation, IWG-LMNPP-92/2, Vladimir, Russia, September 15–19, 1997. IAEA, Vienna, Austria. Brumovsky, M., 2003. Unified procedure for lifetime assessment of components and piping in WWER NPPs “VERLIFE”. Technical report. In: European Commission, 5th Euratom Framework Programme 1998–2003, Brussels. Dufresne, J., 1981. Probabilistic application of fracture mechanics. Advances in Fracture Research. Fracture 81 (2), 517–531. Gillemot, F., Horvath, M., Uri, G., Fekete, T., Houndeffo, E., Acosta, B., Debarberis, L., Viehrig, H.-W., 2007. Radiation stability of WWER RPV cladding materials. Int. J. Press. Vessels Pip. 84, 469–474, http://dx.doi.org/10.1016/j.ijpvp.2007.04.004. Haggag, F.M., Corwin, W.R., Nanstad, R.K., 1990. Effects of irradiation on the fracture properties of steel weld overlay cladding. Nucl. Eng. Des. 124, 129–141, http://dx.doi.org/10.1016/0029-5493(90)90359-6. IAEA-TECDOC-1442, 2005. Guidelines for Prediction of Irradiation Embrittlement of Operating WWER-440 Reactor Pressure Vessels. IAEA, Vienna. Konheiser, J., Rindelhardt, U., Viehrig, H.-W., Böhmer, B., Gleisberg, B., 2006. Pressure vessel investigations of the former Greifswald NPP: fluence calculations and Nb based fluence measurements. In: ICONE14/FEDSM2006 Proceedings, Contribution ICONE 14-89578, 2006M. Margolin, B.Z., et al., 2010. TAREG 2.01/03 Neutron Irradiation Embrittlement Assessment and Validation of Embrittlement Models for WWER RPVs – Task 6: Investigation Regarding Cladding Metal Properties. European Commission, Brussels, Belgium. Pistora, V., Kral, P., 2003. Evaluation of pressurised thermal shocks for VVER440/213 reactor pressure vessel in NPP Dukovany. In: Transactions of the 17th International Conference on Structural Mechanics in Reactor Technology (SMIRT-17), 2003 Paper G01-3. Rindelhardt, U., Viehrig, H.-W., Konheiser, J., Schuhknecht, J., Noack, K., Gleisberg, B., 2009. RPV material investigations of the former VVER-440 Greifswald NPP. Nucl. Eng. Des. 239 (5), 1581–1590, http://dx.doi.org/10.1016/ j.nucengdes.2008.07.018. Timofeev, B.T., Karzov, G.P., 2006. Assessment of the WWER-440/V-213 reactor condition. Int. J. Press. Vessels Pip. 83 (3), 216–226, http://dx.doi.org/10.1016/ j.ijpvp.2005.11.007. Viehrig, H.-W., Altstadt, E., Houska, M., Valo, M., 2012a. Fracture mechanics characterisation of the beltline welding seam of the decommissioned WWER-440 reactor pressure vessels of nuclear power plant Greifswald Unit 4. Int. J. Press. Vessels Pip. 89, 129–136, http://dx.doi.org/10.1016/j.ijpvp.2011.10.016. Viehrig, H.-W., Schuhknecht, J., Rindelhardt, U., Weiss, F.-P., 2009. Investigation of the beltline welding seam of the Greifswald WWER-440 unit 1 reactor pressure vessel. J. ASTM Int. 6 (5), http://dx.doi.org/10.1520/JAI101925. Viehrig, H.-W., Houska, M., Arora, K.S., Rindelhardt, U., 2010. What one learns from reactor pressure vessels of decommissioned nuclear power plants. In: Proceedings of International Symposium FONTEVRAUD 7: Contribution of Materials Investigations to Improve the Safety and Performance of LWRs, French Nuclear Energy Society, Avignon, France, September 26–30, 2010. Viehrig, H.-W., Houska, M., Altstadt, E., Kuechler, R., 2012b. Fracture mechanics characterisation of forged base metal ring of the decommissioned reactor pressure vessel of NPP Greifswald WWER-440 Unit 4. In: Yamamoto, T., Sokolov, M., Hanson, B. (Eds.), ASTM Symposia Effects of Radiation on Nuclear Materials. ASTM STP 1547, 2013, vol. 25. ASTM International, http://dx.doi.org/10.1520/STP104056. Viehrig, H.-W., Houska, M., Altstadt, E., 2015. Radiation and annealing response of WWER 440 beltline welding seams. J. Nucl. Mater. 456, 334–343, http://dx.doi.org/10.1016/j.jnucmat.2014.10.004.