Dynamic vs. quasi-static shear failure of high strength metallic alloys: experiment and modelling
Résumé
In many industrial processes, such as e.g. cutting and punching, or accidental events, such as e.g. crash or impact, cracks may initiate and propagate under low stress triaxiality. Dealing with ductile fracture at low stress triaxiality is consequently of major interest, while being not trivial and still remaining a challenge in terms of experimental investigation, constitutive modelling and numerical simulation. Starting from experimental observations, the present work aims at reproducing via a unified model the consequences of two deterioration mechanisms occurring under low and high strain rate shear loadings, respectively, namely void growth induced damage and adiabatic shear banding. The material considered is a high strength Ti6Al4V titanium alloy. In order to study the underlying mechanisms at the origin of shear failure in the material at stake, we have carried out experiments involving shear-pressure combined loading at low and high strain rates. To observe the current and ultimate deterioration states, some tests were interrupted before fracture and other ones were conducted until ultimate failure. At low strain rate, involving quasi isothermal conditions, the material failure has been seen to result from void growth and further dimple formation, in spite of the pressure applied. At high strain rate, involving quasi adiabatic conditions, the material failure has been found to result from the adiabatic shear banding (ASB) localisation mechanism, as expected for this class of alloys. On the modeling side, the back mean stress concept, see [1], and the embedded band based approach, see [2], have been incorporated in a unified constitutive framework in order to describe the behaviour and shear failure of high strength Ti6Al4V titanium alloys. The former concept is linked to void growth induced damage, when strain rate is low, the latter approach to adiabatic shear banding induced deterioration, when strain rate is high enough. The performance of numerical simulations using the model is evaluated in view of the experimental results mentioned. [1] LONGÈRE P., DRAGON A., 2013, Description of shear failure in ductile metals via back stress concept linked to damage-microporosity softening, Eng. Fract. Mech., 98, pp.92-108 [2] LONGÈRE P., DRAGON A., DEPRINCE X., 2009, Numerical study of impact penetration shearing employing finite strain viscoplasticity model incorporating adiabatic shear banding, J. Eng. Mat. Tech., ASME, 131, pp.011105.1-14