On the automated characterisation of inclusion-induced damage in 16MnCrS5 case-hardening steel

Manganese sulphide inclusions are commonly found in steels and known to facilitate the formation of deformation-induced damage sites in the form of voids during cold forming. These damage sites either exist as cracks, splitting the inclusion in two parts, or as delamination, separating the inclusion from the surrounding steel matrix. Both negatively influence the longevity of components, especially under cyclic loading. The analysis of damage is inherently scale-bridging, ranging from deteriorated global mechanical properties of the finished part, over the damage behaviour of individual inclusions, to the local description of individual voids. In this work, we set out to devise an analysis approach gathering information on all these scales. To this end, we conducted in-situ tensile tests while acquiring high resolution SEM panoramic images and analysed them with two neural networks, trained for this work, to detect damage sites with respect to the inclusions at which they nucleated. We find that the main damage mechanism during tensile deformation parallel to the length of inclusions is cracking and that damage evolution is equally influenced by void nucleation and void growth in the observed range of deformation. By focussing on the damaging behaviour of different inclusions, we show that the position of inclusions in the microstructure influences the resulting damage evolution and that the vicinity of pearlite bands leads to decreased damage formation.