Optimization And Application Of Interatomic Potential Parameters For Perovskite Ferroelectrics

Perovskite ferroelectrics are very important for a wide range of technological applications in optoelectronics, waveguides, laser frequency doubling devices, high capacity computer memory cells, etc. With the development of the computer technology and algorithm, simulation method has been widely used to study the structure and properties of the perovskite-type ferroelectrics for more than 80 years. Molecular dynamics (MD) simulation is one of the most important computer simulation techniques. Interatomic potentials are the key to determining the reliability of MD simulations. The accuracy and reliability of the simulation results depend critically on the quality of the interatomic potentials employed. Because there are a number of interatomic potential parameters for perovskite ferroelectrics, it is a demanding task to optimize these parameters efficiently.In this paper, a more effective optimization method is proposed. The sensitivity analysis is applied to find the leading parameters which affect structures and properties mostly in all the potential parameters. If the unimportant parameters are ignored, the dimension of the optimization is substantially reduced. Using the genetic algorithm to optimize the leading parameters, we optimized the Rigid-ion model potential parameters and shell model potential parameters. For the Rigid-ion model, the potential parameters in the literature and our optimized potential parameters were applied to calculate the lattice constant and physical properties of cubic BaTiO₃ and SrTiO3 by GULP, respectively; for the shell model, the potential parameters in the literature and our optimized potential parameters were applied to calculate the lattice constant and physical properties of cubic and tetragonal BaTiO₃ and PbTiO3 by GULP, respectively. The results show that the optimized structure and physical properties of perovskite ferroelectrics are closer to experimental measurement. That is to say, using the genetic algorithm to optimize the leading parameters, we can obtain very ideal results.Finally, we apply our optimized shell model potential parameters in the simulation of the thin film growth of BaTiO₃. This Paper presented a Kinetic Monte Carlo method which was suited to simulate Perovskite ferroelectric thin film growth, applied it to analyze the influence of substrate temperature and deposition rate on the nucleation rate, island density and bonding rate and the influence of deposition atoms' kinetic energy on the three dimensional growth mode and void rate. It will help optimizing the process parameters for high-performance ferroelectric thin films preparation.