Isolating the effects of deviatoric and hydrostatic stress on damage evolution using cold extrusion experiments

• Experimental approach to extrude material under different Lode parameters but the same triaxiality. • Isolated effect of the Lode parameter on damage mechanisms, void area fraction and void volume. • Shift from inclusion cracking for L = −1 to debonding for L = 0 in 16MnCrS5. • Extrusion under L = 0 results in anisotropic damage. • Determination of inclusion-matrix-interface stresses in RVE-simulations. Ductile damage evolution is known to be dependent on stress triaxiality and Lode parameter. An experimental isolation of these two influencing factors is challenging, especially at large plastic strains, since a variation of the Lode parameter is usually accompanied by a variation in the triaxiality. In this study, forward rod extrusion and forward hollow extrusion are utilized to induce Lode parameters ( L ) of −1 and 0, respectively, while keeping the triaxiality and the plastic equivalent strain the same in both processes. For the triaxiality this is accomplished by superposing hydrostatic pressure by a counterpunch during forward rod extrusion. This allows for the first time, to the author’s knowledge, to experimentally study the isolated effect of the Lode parameter on void evolution at large plastic strains, and the resulting effects on product performance like impact toughness. Mass density measurements and scanning electron microscopy (SEM) on extruded samples of the case hardening steel 16MnCrS5 reveal an increased damage evolution under L = 0 as compared to L = −1. The SEM investigations show a distinct switch in void nucleation mechanisms from cracking of manganese sulphide inclusions for L = −1 to cracking and debonding of inclusions and surrounding matrix for L = 0. The debonding occurs in a preferred direction, leading to anisotropic void area fractions. The locations of inclusion-matrix-debonding are verified by evaluation of interface stresses between matrix and inclusion in representative volume element simulations. While the impact toughness of specimens extracted along different orientations is consistently different, this cannot be conclusively attributed only to the corresponding microscopic damage anisotropy.