Relationships Between Charpy and Fracture Toughness Transition Temperatures for ASME Code Use

Abstract General relationships have long been recognized between transition temperature metrics determined via Charpy V-notch testing (T41J) and the fracture toughness Master Curve transition temperature (T0) for the steels used to fabricate nuclear reactor pressure vessels (RPV). As plants having higher embrittlement levels consider service life extension, Master Curve data is sometimes used to demonstrate continued plant operating safety. A quantitative conversion between T41J (or ΔT41J) and T0 (or ΔT0) supports on-going efforts to develop a Code Case (N-914) that aims to provide parallel and consistent pathways to use whatever data is available, be it Charpy or Master Curve data, for embrittlement assessment. In this paper we use T41J and T0 data collected from the same materials and compiled in NUREG/CR-6609 to develop quantitative relationships between T41J and T0, and between ΔT41J and ΔT0, where the change indicated by the Δ symbol is that due to neutron irradiation embrittlement. This study produced an algebraic equation that allows estimation of T0 or ΔT0 from Charpy transition temperature (T41J) and upper shelf energy (USE) values. The fit demonstrates that RPV steels having lower upper shelf energies exhibit values of ΔT0 of similar magnitude to ΔT41J, while steels with higher upper shelf energy exhibit larger ΔT0 values larger (on average) than ΔT41J values. The practice of using a hyperbolic tangent (tanh) function to fit all Charpy data from lower shelf through upper shelf links the T41J and USE metrics. For low upper shelf steels this linkage generates an additional ΔT41J shift caused by the reduction in transition slope, a reduction not manifest in the transition slope of fracture toughness (KJc) data. Thus, this increase in Charpy transition temperature for low upper shelf steels is an artifact of tanh fitting unrelated to embrittlement, causing the approximate equality of ΔT41J and ΔT0 observed for low upper shelf materials. The larger values of ΔT0 relative to those of ΔT41J observed in high upper shelf materials can most likely be attributed to differences in loading rate and crack-tip constraint between Charpy and fracture toughness specimens.