13th International Conference on Fracture June 16–21, 2013, Beijing, China -8- Figure 4 : Evolution of the active plastic zone for LCIKF cycles. The history of local mechanical parameters - principal stress σI, cumulated plastic strain p and triaxiality τ - at the physical point where cleavage is triggered (identified by SEM investigations) were computed for each specimen. The results corresponding to the different experiments are shown in Fig. 5 where circles indicate cleavage fracture. The case of an isothermal test was added in Fig. 5 for comparison. For all the experiments, the maximum principal stress and the cumulated plastic strain are increasing at cleavage triggering. This is consistent with the hypotheses of Beremin type models that consider the necessity of plastic flow to initiate microcracks and a stress-controlled propagation. Moreover, we note that the value of principal stress at which the cleavage fracture is triggered is almost the same for all the experiments (with or without warm prestressing), about 2250MPa, whatever the level of plastic strain, reinforcing the idea of a critical stress needed to propagate cracks. Note that this value of critical stress is higher than the one found in [4], as our modeling slightly overestimates stresses at large strains (see inset fig. 3). Compared to the isothermal test, the plastic strain at fracture - and its evolution with stress - is larger for WPS tests. The evolution of the triaxiality vs. stress is also different for WPS tests than for isothermal tests, which suggests that LAF models should include the effect of triaxiality. Figure 5: Evolution of plastic strain as a function of the principal stress (b) and triaxiality (b) at the location where cleavage was found to be triggered from microstructural investigations. Solid lines correspond to WPS experiments, dashed lines to isothermal test at -150°C.
RkJQdWJsaXNoZXIy MjM0NDE=