Computational modeling of grain growth in self-reinforced silicon nitride

A model is proposed for studying grain growth in silicon nitride by computer simulation. The simulation is based on the Monte Carlo method and uses the Potts model with additional features peculiar to silicon nitride ceramics, such as anisotropy of grain boundary energies and the presence of liquid phase. The simulation successfully produced self-reinforced microstructures. It was found that the self-reinforcement phenomenon occurred only within a particular range of liquid phase fractions. The influence of anisotropy in interface energy on inhomogeneous grain growth was also examined. When interface energies are isotropic, the microstructure was essentially homogeneous. The results of these simulations suggest that anisotropy in interface energy and the presence of an appropriate amount of liquid phase are two important factors controlling the self-reinforcement process in silicon nitride materials.