Fatigue in a corrosive environment has been identified as highly detrimental to the mechanical strength of metallic alloys . In the case of the 316L stainless steel grade, pitting corrosion is observed, leading to localised sharp defects critical to its high cycle fatigue behaviour . Indeed, cracks have a tendency to initiate on such defects on prone to pit materials.
The work hereby presented focuses on 316L specimens produced by Laser Powder Bed Fusion (L-PBF) additive manufacturing and subsequently machined and polished. This preparation was carried out to ensure the characterization of the core and not the raw surface of the elements. Metal powder was horizontally layered, meaning, perpendicular to the loading axis of the specimens. As consequence of the L-PBF process, defects such as lacks of fusion are present in the material .
The fatigue response of three batches corresponding to three configurations of surface integrity are investigated: (i) No corrosion before fatigue testing, (ii) corroded samples under potentiodynamic anodic polarization conditions in a neutral NaCl solution before fatigue testing and (iii) samples with hemispherical machined defects. Fracture surface observations on the first batch showed in every sample initiation from a lack of fusion surface pore. Fatigue strength obtained was in good agreement with the literature. In the second batch, some cracks initiated on corrosion pits depending on the severity of the corrosion applied to each sample and its initial population of defects. Correlation between the square root of the projected area of the initiating defect  and the fatigue strength was observed for both pits and lacks of fusion on a Kitagawa-Takahashi diagram. Hemispherical machined defects seemed less harmful than pits and pores from a fatigue life perspective. Defect morphology didn't appear to be the only driving force for crack initiation as a pit is closer to a machined defect than a flat lack of fusion in terms of shape. The competition between lack of fusion, corrosion pit and machined defects is studied and further discussed in this paper.
 Q.Wu et al., 2017.
 M. El May et al., 2013.
 R. Shrestha et al., 2016.
 Y. Murakami and M. Endo, 1994.