Fatigue Life Evaluation and Prediction of Inconel 718 Manufactured by Selective Laser Melting

Abstract

The prevalence of additive manufacturing (AM), or 3D printing, has grown in recent decades as a method of manufacturing complex geometries, prototype parts, or high tuned microstructures. One method of additive manufacturing is selective laser melting (SLM), a powder bed fusion technique which utilizes a laser to melt a metal alloy powder layer by layer to form the final geometry. Inconel 718 is a nickel-based superalloy commonly used in high temperature applications due to its superior mechanical and corrosion resistance properties. Owing to a difficult in machining, additive manufacturing has become particularly advantageous method for the production of Inconel 718. The SLM process is associated with a greater degree of process defects as compared to a traditional production method, in particular a poor surface condition, porosity and an anisotropic microstructure. These surface defects pose an issue in response to cyclic loading, as they serve as stress concentrators and can become crack initiation sites. As a result, post-processing treatments are typically required to improve the quality and mechanical properties. This master work focused on two primary aims. First, the characterization of SLM Inconel 718 processed with different layer thicknesses, build directions and post-processing surface treatments and second, the development of a prediction model for fatigue life based on process defects. Smaller layer thickness was found to result in lower surface roughness, which in turn resulting in fatigue performance as compared to larger thicknesses. Differences in surface quality and properties were identified on samples built at an incline. The plane opposite the build direction (the ‘downskin’ side) exhibited higher surface roughness and increased near surface hardness. Mechanical polishing of the surface led to a significant improvement in fatigue performance associated with dramatic reduction in surface roughness and near-surface voiding. Electropolishing as a post-processing treatment did not have a significant impact on surface quality or on fatigue performance. Surface defects were identified as the location of fatigue crack initiation. Removal of these defects by post-processing would therefore provide the greatest improvement in the fatigue performance of SLM Inconel 718. The Murakami-Endo method of fatigue life prediction showed promising results in predicting fatigue life of a sample population based on defects, with good agreement to experimental data.