Reliability of molecular dynamics interatomic potentials for modeling of titanium in additive manufacturing processes
Abstract
We present an approach to study the transferability of molecular dynamics (MD) interatomic potentials (IPs) for materials modeling during metal additive manufacturing (MAM) processes, which involves rapid and cyclic melting, solidification, vaporization, and solid phase transformations. We apply the approach to the ten available IPs of titanium (Ti) as an example, which resembles the MAM process modeling of commercially pure Ti and Ti-base alloys. We consider the capability to produce solid phase change behavior of the alloy in equilibrium and nonequilibrium thermodynamics as the primary required characteristics of an IP for the MAM process modeling. Two-phase α-β, β-liquid, and liquid-vapor coexistence properties are used as the secondary criteria. Finally, we use other relevant single-phase properties as the tertiary criteria, such as the elastic constants at high temperatures, liquid structure factor, self-diffusivity, and shear viscosity. We show there are only three IPs that demonstrate reasonable characteristics for MAM process modeling.
Publication Title
Computational Materials Science
Recommended Citation
Etesami, S., Laradji, M., & Asadi, E. (2020). Reliability of molecular dynamics interatomic potentials for modeling of titanium in additive manufacturing processes. Computational Materials Science, 184 https://doi.org/10.1016/j.commatsci.2020.109883
