Prediction of Fracture Toughness for WWER RPV Integrity Assessment on the Basis of the Unified Curve Approach and Surveillance Specimens Testing

For construction of the fracture toughness temperature curve that may be used for WWER RPV integrity assessment on the basis of tests of cracked surveillance specimens, the issues have to be solved as follows. First of all, it is important to determine how fracture toughness varies as a function of temperature, and how the fracture toughness vs. temperature dependence, K_JC(T), changes with in-service material degradation due to neutron irradiation. These variations of K_JC(T) curve are known to be the shift of K_JC(T) curve to higher temperature range and change in the K_JC(T) curve shape. At present, two advanced engineering methods are known that allow the prediction of K_JC(T) curve on the basis of small-size fracture toughness specimens (for example, pre-cracked Charpy specimens), namely, the Master Curve and the Unified Curve methods. Procedures of test result treatment for the Master Curve and the Unified Curve are very similar. The Master Curve method uses the lateral temperature shift condition and, therefore, does not describe possible change in the K_JC(T) curve shape. The Unified Curve method has an advantage as compared with the Master Curve as the Unified Curve describes a variation of the K_JC(T) curve shape when degree of embrittlement increases. This advantage becomes important for RPV integrity assessment when the reference K_JC(T) curve is recalculated to the crack front length of the postulated flaw that is considerable larger than thickness of surveillance specimens. Application of the K_JC(T) curve determined from test results of cracked surveillance specimens to RPV integrity assessment requires also to introduce some margins. These margins have to take into account the type and number of tested specimens and the uncertainty connected with spatial non-homogeneity of RPV materials. Indeed, there is sufficient number of experimental data showing variability in fracture toughness for various parts of RPV. Therefore, situation is possible when the material properties near the postulated flaw will be worse than the properties of surveillance specimens. In the present report, advanced approaches are considered for prediction of fracture toughness for WWER RPV integrity assessment that allow one: • to construct the K_JC(T) curve for irradiated RPV steels with any degree of embrittlement; • to provide transferability of fracture toughness data from cracked surveillance specimens to calculation of resistance to brittle fracture of RPV with a postulated flaw.