Validation and application of cellular automaton model for microstructure evolution in IN718 during directed energy deposition

• Published in: Computational materials science 2023-10, Vol.230 (C), p.112450, Article 112450
• Main Authors: Yuan, Lang, Ju, Siyeong, Huang, Shenyan, Spinelli, Ian, Yang, Jiao, Shen, Chen, Mohr, Luke, Hosseinzadeh, Hamed, Bhaduri, Anindya, Brennan, Marissa, Sun, Changjie, Kitt, Alex
• Format: Article
• Language: English
• Subjects: Cellular automaton; Directed energy deposition; Microstructure; Nickel-based superalloy; Nucleation; Solidification
• ISSN: 0927-0256; 1879-0801
• DOI: 10.1016/j.commatsci.2023.112450

[Display omitted] •A highly parallel 3D cellular automation model studies nucleation and grain growth for directed energy deposition.•The calibrated model accurately predicts grain size and misorientation for IN718 with quantitative validation.•Coupled model predicts solidification behaviors, including remelting, epitaxial growth, and columnar-to-equiaxed transition.•Multiple-track multiple-layer simulation captures and explains sandwich-patterned structures with fine grains between layers. In Directed Energy Deposition (DED), the complex and nonuniform thermal history during the laser heating and cooling cycles determines the as-built microstructures and defects and thus affects the ultimate mechanical properties of fabricated parts. In this study, a highly parallel solidification microstructure model based on the Cellular-Automaton method is developed to predict the grain structure evolution in solid-solution strengthened Ni superalloys, IN718. Through coupling with the DED process model via Simufact, the grain structures in single tracks and a multiple-track multiple-layer block are predicted and compared with experiments. Parametric studies are performed to investigate the impact of key parameters in the stochastic nucleation model and calibrate the nucleation parameters, including nucleation density, nucleation undercooling, and its standard derivation. The predictions capture the general grain growth behavior, including remelting, epitaxial growth, and columnar-to-equiaxed transition, with a quantitative agreement with experimental results in terms of grain size and misorientation. This study demonstrates that the 3D grain structural model coupled with a process model has established the capability to model grain evolution for additive manufacturing processes, illustrating a powerful tool to assist the materials design and process development in additive manufacturing.