Integrating micropillar compression testing and crystal plasticity modelling to unravel slip system activity and deformation mechanisms in P91 steel

The plastic behaviour and slip activity of P91 steel were studied at room temperature using integrated micropillar compression tests and crystal plasticity (CP) modelling. Micropillars were fabricated from martensitic blocks with distinct crystallographic orientations to activate specific slip plane families in the body-centered cubic (BCC) structure. Scanning electron microscopy (SEM) analysis of the slip traces revealed that all three slip families, {110}, {112}, and {123}, contributed to plastic deformation. While most slip activity was associated with the highest Schmid factor, deviations occurred in approximately 12 % of micropillars. Compression tests were performed under both load-controlled and displacement-controlled conditions, which revealed that load-controlled tests showed lower strain hardening but provided more accurate yield stress measurements compared to displacement-controlled tests. Post-deformation cross-sectional electron backscatter diffraction (EBSD) analysis uncovered the grain structure beneath the top surface of the micropillars, which highlighted the critical role of sub-surface microstructures in analysing slip traces. The CP model was calibrated using stress-strain data from single-grain micropillars under both loading conditions, incorporating the effects of machine-dependent material response. The CP simulations accurately predicted slip trace morphology and location in single-grain micropillars. For polycrystalline micropillars, slip traces were found to be more localized in larger grains compared to smaller ones. The present work provides a basis for understanding how martensitic steel microstructure affects mechanical performance. This study highlights some limitations of phenomenological CP simulations for predicting slip activity and highlights the need for incorporating sub-surface microstructures and machine-dependent material response effects in future studies.