Quantitative Texture Prediction of Epitaxial Columnar Grains in Alloy 718 Processed by Additive Manufacturing

Abstract

The lack of a reliable theoretical model of the processing-microstructure relationship of AM (Additive Manufacturing) material is preventing AM technology from being widely adopted by the manufacturing community. The goal of this work is to establish the link between the microstructure (texture) and the process parameters of metal AM processes. A quantitative method based on the epitaxial growth of columnar grains within and across melt pools is proposed to predict the texture formation during a metal AM process. The state-of-the-art CALPHAD-informed FEM (finite element method) simulation has been used to predict the geometry and thermal profile of the quasi-steady melt pool. The thermal gradient distribution within the 3D melt pool determines the crystallography direction and growth direction of the columnar grains within each deposited single tracks. The single tracks with the predicted geometry are amalgamated together to represent the bulk part, and the epitaxial growth of grains across the boundary of neighboring tracks are quantitatively modeled. The proposed method is calibrated and validated by experimental studies of metal AM processed Alloy 718.