Predicting microstructural evolution is a decisive step in the study of aging processes in alloys, especially under irradiation. The results of ab initio calculations of electronic structures can be used to parameterize kinetic methods such as Atomic Kinetic Monte Carlo simulations that allow reproducing quantitatively atomic diffusion and the resulting microstructure. Their use is however limited by their computational cost. Mesoscopic simulations are less concerned by such limitation, but suffer from the lack of reliable parameterization method to use data from simulations at lower scales that leads to a limited prediction capacity. A simulation method called Cellular Kinetic Monte Carlo is developed in this work to bridge the gap scales between atomic and mesoscopic scale simulation of diffusion. A crystal is there modeled as a. This method is based on a description of the crystal as a lattice of cells described by the discrete number of solute atoms they represents. The properties are then obtained by a controlled coarse-graining procedure based on Atomic Kinetic Monte Carlo simulations. It allows reproducing quantitatively macroscopic equilibrium for all cell sizes and has been applied to the Iron-Copper alloy. In order to describe kinetic properties at these scales, a generic computational tool has been developed to compute the Onsager matrix of alloys, based on the Self Consistent Mean field method. Diffusion and precipitation simulations have been done and the results are presented and assessed by a systematic comparison with Atomic Kinetic Monte Carlo simulations.
