Multiscale simulation of grain refinement induced by dynamic recrystallization of Ti6Al4V alloy during high speed machining

Abstract

During high speed machining (HSM), the strong thermal-mechanical coupling can lead to the microstructure evolution in the deformation zone of workpiece. Grain refinement may occur, which has great effects on the mechanical behavior, and even on the fatigue strength and corrosion resistance of the machined surface. The development of multiscale models to predict the microstructure evolution is gaining rising interest. This study aims to investigate the grain refinement induced by dynamic recrystallization (DRX) occurring in HSM of Ti6Al4V, through finite element (FE) and cellular automata (CA) methods. An orthogonal cutting model for HSM of Ti6Al4V is developed combining a modified Johnson-Cook constitutive model (TANH) and Johnson-Mehl- Avrami-Kolmogorov (JMAK) DRX model. The CA model is proposed considering dislocation density evolution, grain nucleation and growth. The 2D mesoscopic microstructure evolution is simulated successfully by the CA model in which the input deformation parameters come from the FE simulations of the orthogonal cutting process. Finally, the grain size and microstructure morphology calculated by both FE and CA methods are compared with those characteristics obtained experimentally by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). Identical microstructure predictions from both CA and FE methods show a reasonable agreement with the TEM results, on the condition that twinning and phase transformation are not considered in the simulations. This work proves that the combination of FE and CA methods is an effective approach to achieve a more comprehensive understanding of the microstructure evolution and its effect on mechanical behavior during HSM. It shows that the rise of both DRX volume fraction and DRX grain size finally results in the slightly decreasing of average grain size of serrated chips with the increase of cutting speed, which leads to the strain softening phenomenon of flow stress.