How to characterise the anisotropy of laser powder bed fusion-processed parts? Towards a surrogate, non-destructive indentation-based approach

Abstract

This work proposes a non-destructive indentation (NDI)-based surrogate mechanical testing framework for anisotropy and mechanical property evaluation of LPBF processed Inconel 718 superalloy. A Knoop-hardness tester was employed to develop the surrogate framework. Further post-heat treatments (PHTs) were performed at various cooling rates. The as-printed and PHT (SA, SB, SC, and SD) samples were characterized using X-ray diffraction (XRD) optical microscopy (OM), scanning electron microscopy (SEM), and electron back-scattered diffraction (EBSD) techniques. The yield locus plots show strong anisotropy for the as-printed sample. The IPF maps and pole figures for the as-printed samples showed elongated columnar grains with strong fiber texture along the (001) plane parallel to the build direction. They corroborated sufficiently with XRD, grain boundary map (GBM), and Kernel average misorientation (KAM) analysis. The yield locus shape of samples subjected to homogenization treatment becomes a nearly uniform ellipse with a reduction in the difference between σx and σy values, which can be attributed to the recrystallization of grains as observed by the transition from low-angle grain boundaries (LAGBs) to high angle grain boundaries (HAGBs). The PHT samples show a further increase in yield loci size due to the precipitation of γ’ and γ” strengthening phases due to aging treatment and corroborated with higher Taylor factor and dislocation density than the as-printed samples. The yield loci were sensitive to the PHT's cooling rates. Notably, among PHT ones, ‘SA’ samples exhibited a strong correlation (R2 ≈ 97.5 %) between the Taylor factor and yield locus. It was attributed to slower cooling rates (furnace cooled) with noticeably enhanced precipitation of strengthening phases, an increase in the volume fraction of recrystallized grains (from grain orientation spread (GOS) map), corroborating with higher dislocation density due to local strain as inferred from EBSD analysis. The proposed NDI surrogate approach was initially validated on an as-printed Inconel 718 stator blade. The yield locus showed a strong correlation between the stator blade and the as-printed sample, indicating the effectiveness of the proposed methodology.