摘要：This paper deals with the simulation of microstructure evolution in steels, specifically eutectoid steels, where competitive diffusive (pearlitic) and diffusionless (martensitic) transformations may take place. Diffusion-controlled transformations are modelled by using the classical Johnson-Mehl- Avrami-Kolmogorov law for isothermal transformations, while the martensitic transformation is assumed to obey either the Koistinen-Marburger or the Yu laws. The non-isothermal evolution of diffusive transformations is derived from the isothermal transformation kinetics either by invoking the additivity rule, or by integrating the rate form of the Johnson-Mehl-Avrami-Kolmogorov law in time. The ability of both techniques to build continuous cooling transformation (CCT) diagrams from isothermal transformation (IT) diagrams is evaluated. Microstructure evolution is coupled with the thermal analysis, performed using the finite element method. A finite element analysis of a quench problem is finally carried out to evaluate the performance of the model.
其他摘要：This paper deals with the simulation of microstructure evolution in steels, specifically eutectoid steels, where competitive diffusive (pearlitic) and diffusionless (martensitic) transformations may take place. Diffusion-controlled transformations are modelled by using the classical Johnson-Mehl- Avrami-Kolmogorov law for isothermal transformations, while the martensitic transformation is assumed to obey either the Koistinen-Marburger or the Yu laws. The non-isothermal evolution of diffusive transformations is derived from the isothermal transformation kinetics either by invoking the additivity rule, or by integrating the rate form of the Johnson-Mehl-Avrami-Kolmogorov law in time. The ability of both techniques to build continuous cooling transformation (CCT) diagrams from isothermal transformation (IT) diagrams is evaluated. Microstructure evolution is coupled with the thermal analysis, performed using the finite element method. A finite element analysis of a quench problem is finally carried out to evaluate the performance of the model.
