Use of miniaturized compact tension specimens for fracture toughness measurements in the upper shelf regime

In the nuclear field, the importance of direct fracture toughness measurements on reactor pressure vessel (RPV) materials has been nowadays widely recognized, as opposed to Charpy-based estimations. However, sample dimensions have to be kept small in order to optimize the use of available material (often in the form of previously broken Charpy specimens) or, in the case of new irradiations, make effective use of the limited space available inside irradiation facilities. One of the most appealing geometries for fracture toughness measurements is the miniature Compact Tension specimen, MC(T), which has the following dimensions: B = 4.15 mm, W = 8.3 mm, cross section 10 × 10 mm2. Four MC(T) specimens can be machined out of a broken half Charpy, and in the case of irradiation ten MC(T) samples occupy approximately the same volume as a full-size Charpy specimen. A comprehensive investigation is presented in this paper, aimed at assessing the applicability of MC(T) specimens to measure fracture toughness in fully ductile (upper shelf) conditions. In this study, 18 1TC(T) and 20 MC(T) specimens have been tested at different temperatures from three RPV steels and one low-alloy C-Mn steel. The results obtained clearly show that MC(T) samples exhibit lower fracture toughness properties, both in terms of initiation of ductile tearing (according to various test standards) and resistance to ductile crack propagation (J-R curve). The reduction of tearing resistance might be attributed to work hardening prevailing over loss of constraint in the uncracked ligament for a side-grooved specimen, or to the inadequacy of J-integral to represent ductile crack extension in very small specimens. Both arguments need to be verified with further investigations.