A cellular automaton model integrated with CALPHAD-based thermodynamic calculations for ferrite-austenite phase transformations in multicomponent alloys

Abstract

A cellular automaton (CA) model, which is directly coupled with thermodynamic calculations based on the CALPHAD method, is developed for the simulation of ferrite (α)-austenite (γ) phase transformations involving the partitioning and long-range diffusion of both interstitial and substitutional elements in multicomponent alloys. A data management scheme is proposed, which renders the efficient coupling between the kinetic simulation using the CA approach and thermodynamic calculations. The present CA model is applied to simulate the cyclic phase transformations between the α- and γ-phases for a quaternary Fe-0.02C-0.2Mn-0.1Si (wt%) alloy in both one- and two-dimensions. The simulated curves of α-volume fraction varying with temperature clearly illustrate the different α-γ transformation stages. The transformation kinetics and concentration profiles of Mn and Si obtained by the 1-D CA simulation agree well with DICTRA predictions. The 2-D simulation of a single circular grain is found to have shorter stagnant stages and faster transformation kinetics than that of the 1-D planar case. This can be attributed to the different growth geometries of 1-D planar and 2-D circular, and the effect of substitutional element Mn on the migrating interface. The non-uniform distribution of C and residual spikes of Mn and Si can be clearly delineated in the 2-D multi-grain simulation. It is shown that the data management scheme proposed in the present work significantly reduces the simulation time for the required thermodynamic data by two orders of magnitude approximately, which makes it feasible to perform valid simulations of microstructural evolution and solute distributions during α-γ phase transformations in steels with more than three components.