Relevance of a Size Effect Correction on a Minimum Toughness Value in the Ductile-to-Brittle Transition of Ferritic Steel to Study the French PWR RPV Resistance

Abstract

The integrity of Reactor Pressure Vessel (RPV) which ensures radioactivity confinement is reevaluated regularly to take into account updated knowledge and to verify the high level of safety required for this component. The French 900-MWe RPV integrity demonstration was reviewed by the French Safety Authority (ASN) and the Institute for Radiological Protection and Nuclear Safety (IRSN). This paper focuses on the validity of the fracture criterion used in this demonstration realized by the operator. In the RPV operating temperature range, ferritic steel can have a ductile-to-brittle behaviour due to irradiation embrittlement. In this range, toughness results from normalized test specimens exhibit a size effect that could be related to the crack length. According to this, the operator has considered that the toughness of the RPV containing a given crack is equal to the minimum toughness defined on the RCC-M curve —similar to the ASME curve— reduced by a factor function of the crack length. For IRSN, this criterion is not relevant. This conclusion is based on the statistical analysis of toughness results available in the literature which shows that although taking into account a crack length correction is relevant on the average toughness, it is not the case for the minimum toughness which is the value used in the mechanical analysis: There is no size effect on the minimal toughness. Following the IRSN’s recommendation, the RPV service ability of the 900-MWe reactors has been demonstrated by the operator without applying a crack length correction on the toughness.