Development of a crack arrest toughness measurement technique using small specimens to evaluate the fracture behavior in the ductile-brittle transition of "ferritic" steels
Fracture of technical ferritic steels is a complex phenomenon that involves mechanisms taking place over length scales covering ten orders of magnitudes. The ferritic steels are characterized by a change of the fracture mode, going from ductile at high temperature to brittle at low temperature. The brittleness is measured by a ductile-to brittle transition temperature (DBTT). When such steels are used in a nuclear environment and experience neutron irradiation, they suffer from irradiation damage, which is manifested by an increase of the DBTT, called embrittlement. Embrittlement of components made of ferritic steels is one of the key parameters limiting the life of the nuclear power plants. Embrittlement is usually quantified with tests on small specimens. Unfortunately, brittle fracture toughness is strongly affected by the size and geometry of the specimens used. Owing to the small volume of available irradiation facilities and the limited number of neutron-irradiated specimens, techniques to measure initiation fracture toughness with small specimens have been developed over the years. The specimen size effects can be quantified and predicted with the application of local approach to fracture models. In the frame of these models, critical stress/strain conditions at the crack tip leading to the initiation of an unstable macro-crack. The sequence of event is link to the capacity of the material to arrest micro-crack in the process zone at the crack tip. The objective of this PhD project was to develop an experimental technique for measuring the apparent arrest toughness of ferritic steels with miniaturized specimens. Small cantilevers made of the 9Cr Eurofer97 ferritic steel were produced with a brittle thin laser-treated layer, which enables crack initiation. The cracks were successfully arrested in the more ductile equiaxed ferritic matrix in a series of tests that were carried out from -123 °C to room temperature with different test configurations. The mechanical tests showed that brittle fracture could be triggered and stopped inside the specimen with a width of 2 mm over a temperature range of more than 100 °C. An apparent arrest toughness was calculated and its temperature dependence was related to the plastic work dissipated by a propagating crack. The technique was then applied on two technical steels: the as-received 9Cr tempered martensitic Eurofer97 steel and the low-allow reactor pressure vessel steel JRQ. Series of crack arrest tests were successfully performed on these two steels. The mechanical tests were supplemented with finite element simulations to determine the stress and strain fields in the specimens and at the crack tip at initiation as well as at arrest. The simulated stress field were compared with the analytical description of the stress field based on the stress intensity factor, which is calculated with a standard equation using the load and the specimen and crack dimensions. Some adjustments of the analytical solutions were proposed to take into account plasticity. We showed that the arrest toughness increases moderately within the transition region and that this increase scales with a corresponding increase of the local fracture stress. The positive dependence of the local fracture stress is linked to the plastic work dissipated during the crack surface formation. Fractographic observations were conducted to gain insight into the crack path in the different microstructures of the materials investigated.
Dr Aïcha Hessler-Wyser (présidente) ; Prof. Philippe Spätig (directeur de thèse) ; Prof. Andreas Mortensen, Dr Peter Hähner, Dr Milan Heczko (rapporteurs)
2024
Lausanne
2024-12-06
10989
204