Experimental study of the fracture toughness transferability to pressurized thermal shock representative loading conditions

Fracture toughness testing on standard specimens in the ductile to brittle transition regime is well established and was first standardized by ASTM in 1997. However, its applicability to structural components and its potential conservatism, due to the high constraint condition in standard specimens, remain a subject of concern. In this work a study of the transferability of the fracture toughness measurements from laboratory specimens to structural components is performed. The structural component of interest is the Reactor Pressure Vessel subjected to an accidental loading condition called Pressurized Thermal Shock (PTS). As the actual component cannot be reasonably tested, an original experimental set-up is developed to simulate PTS representative loading conditions. A semi-elliptical crack is introduced by fatigue in a disk shape specimen which is biaxial loaded in the ductile to brittle transition regime. The developed disk specimen is called PTS-D specimen. Master Curve tests on 15 PTS-D specimens are performed and compared to standard size specimens having deep and shallow cracks. The reference temperature obtained on PTS-D is equivalent to the one obtained on half-inch compact tension specimens. To support the experimental results, finite element calculations are performed. It turns out that, in the tested range, constraint in PTS-D specimens is high. However, loss of constraint appears earlier than in standard specimens. It is found that the Beremin approach combined with analytical tool allows rationalizing all test results.