PSI - Issue 42

S. Lindqvist et al. / Procedia Structural Integrity 42 (2022) 42–49 Author name / Structural Integrity Procedia 00 (2019) 000–000

43

2

mm). Thus, it does not represent a definitive upper shelf temperature. Historically, one simple definition of upper shelf was the temperature where a Charpy-V impact test leads to a 100 % ductile fracture surface. The old British requirement that the British nuclear reactors have to operate on the upper shelf stems from this definition. The introduction of fracture mechanics and fracture toughness tests has, however, clouded the technological definition of upper shelf. [1] In the USA it is common to define the onset of upper shelf as the temperature where the fracture toughness exceeds the ductile initiation fracture toughness J IC . Similarly, in Britain the onset of upper shelf temperature (OUST) is presently defined as the intersection between the 5% probability curve for fracture in the transition region and the mean curve for ductile fracture initiation at 0.2 mm crack growth. The USA definition does not specify a specific probability for the brittle fracture estimate and the ductile initiation definitions differ slightly. The American J IC corresponds to 0.2 mm ductile tearing in addition to crack tip blunting, whereas the British definition is 0.2 mm total crack tip extension including blunting. For a high toughness material, the British definition may correspond purely to crack tip blunting. These definitions, however, rule in no way out the possibility of brittle fracture after some ductile tearing (fast fracture) at a higher temperature. Also, the OUST temperature depends, besides on the brittle fracture properties of the material, also on the ductile fracture properties. For the same brittle fracture properties, a material with a poor ductile fracture resistance will show a lower OUST than a material with a high ductile fracture resistance. [2] In this study, the effect of rapid cooling on tearing resistance of a rolled steel (J-R curve) is investigated. Presently, there is no data, nor predictions, regarding the J-R curve development during a cooling transient in the upper shelf region. 2. Materials and methods 2.1. Test material, matrix and specimens The investigated material is thermo-mechanically rolled ferritic steel plate S460MC, a 25 mm thick ferritic rolled steel plate. The fracture toughness behaviour during a cooling thermal transient was characterized. with C(T) specimens (Figure 4). The thickness, B, of the C(T) specimens was 25 mm and the width, W, was 50 mm. The fracture toughness specimens are oriented in the L-T orientation where the specimens fracture plane’s normal is in the longitudinal rolling direction and the crack propagates in the transverse direction of the plate. All together 10 The J-R curve fracture toughness testing was done using ASTM E1820-16. Isothermal testing was done at 25, 100, 200 and 300 ° C. The testing was begun at 300 ° C, and the specimen was loaded according to the unloading compliance method. After reaching certain combination of load and displacement, the cooling was started and the specimen was loaded with a constant displacement rate without unloadings. The testing was ended after the temperature of the specimen was between 0 to 25 C. The cooling was started at different values of displacement and the the cooling rate at the center of the specimen was 2 ° C/s. The cooling rates are comparable to the cooling rates observed in Nuclear Power Plants during accident conditions [3,4]. The J-integral values at the beginning of cooling are presented in section 3.1. [5] The specimen was cooled by opening the door of the environmental chamber and by spraying liquid nitrogen directly on the specimen at a close range. The liquid nitrogen was sprayed from a copper coil with several holes and spun around the specimen to obtain as even cooling of the specimen as possible, Figure 1. The cooling rates obtained with this system at different depths was measured by using a calibration specimen with drilled holes of various depths, Figure 1 b. The depth of the holes were 4 mm, 8 mm, 12.5 mm. Thermocouples were placed in the holes and at the surface. The cooling rates at different depth locations of the specimen are presented in section 3.1. During the actual fracture toughness testing, a 10 mm deep hole was drilled at one of the corners to confirm that the cooling rate obtained during testing conforms to the cooling rate of the calibration specimen. The J-integral is calculated from the load displacement record. There are various methods to calculate the J-Integral, as given in ASTM E1820: 1) Unloading compliance method, 2) Basic method, 3) Normalization method. The unloading compliance method is typically used and was the main method for analyses of the isothermal J-R specimens were tested. 2.2. J-R curve testing