Effects of post-processing route on fatigue performance of laser powder bed fusion Inconel 718

Abstract

Fatigue performance of Inconel 718 (IN718) parts built using laser powder bed fusion (L-PBF) is often poorer than that of the cast and wrought alloy parts. This is because of detrimental effects arising from the microstructures such as brittle phases and anisotropy as well as porosities. To improve fatigue performance of L-PBF IN718, two types of post processes namely hot isostatic pressing (HIP) (performed at 980 °C, 100 MPa and 4 h) and shot peening (performed at 0.45 mmA intensity and 200 % coverage) were employed individually as well as sequentially on different specimens in addition to a standard processing route consisting of L-PBF build followed by solution annealing and double aging heat treatment. Porosities were considerably reduced by HIP-ing (from 0.39 % to 0.08 % porosity volume ratio) but this did not translate to improvements in fatigue performance. This can be attributed to the presence of inclusions and yield strength reduction resulting from γ” solutionisation. High compressive residual stresses of close to 1 GPa were recorded following shot peening regardless of whether HIP-ing was carried out prior to the shot peening. Improvement in fatigue lives particularly at high stress levels was observed on shot peened specimens without HIP-ing (H + SP) which can be explained by the compressive residual stresses that delay the onset of fatigue crack initiation. Shot peening applied on the HIP-ed specimens (H + HIP + SP) however, resulted in fatigue performance deterioration. This is related to the harmful tensile stresses that must exist alongside the compressive residual stresses in a shot peened surface, as evident from the location of the sub-surface crack initiations near the residual stress transition region approximately 300 μm from the free surface. In this case, detrimental effects arising from tensile stresses appear to have outweighed the benefits from compressive stresses on fatigue performance.