Cellular Automata Model of Low Carbon Steel Recrystallization Using a Refined Dislocation Density Calculation

Cellular automata (CA) recrystallization models offer an alternative to conventional Johnson-Mehl-Avrami-Kolmogorov (JMAK) models. CA models can be used to predict the final microstructure after an annealing treatment. A critical input for this kind of recrystallization model is the dislocation density. In order for this model to properly represent the recrystallization kinetics, an accurate calculation of the dislocation density is needed. In this paper, modified Williamson-Hall and Warren-Arvenbach methodologies were used to calculate the dislocation density of low carbon steel from X-ray diffraction patterns. This density was compared and validated with transmission electron microscopy (TEM) measurements. The calculated dislocation density was used in the CA recrystallization model implemented in the finite element model (FEM) software DEFORM-3D. Validation of this model was carried out by annealing experiments. This model was used to study the recrystallization kinetics and final microstructure obtained during continuous annealing and post-annealing heat treatments of low carbon steel. The results were compared with JMAK predictions and experimental results and discussed in terms of the effect of a proper dislocation density in the predicted microstructure.