Project 2–6: Multiscale Modeling of Materials

Principal Investigator: Eric Darve (Stanford University)

The dielectric constants of ferroelectric materials such as barium titanate can be “tuned” to a specified value by applying an electric field. This property is of great interest for such communications technologies as phased array antennas, where the effective radiation pattern of the array can be reinforced or suppressed in any specified directions. Thus, a single radar system can serve for multiple tasks, such as surface and air detection and tracking and missile uplink capabilities. A wide field of view and significant reduction in overall array size are possible using phased array antennas.

Current molecular dynamics-based modeling methods for barium titanate crystals can account for dipole charge interactions, but they have difficulties accurately modeling phase transitions and vacancy diffusion in the crystal lattice. In this project, a unique set of tools based on genetic algorithms is being used to create a novel model potential tailored specifically for barium titanate. This enables the behavior of the crystal to be modeled with a high degree of fidelity under a diverse range of physical and electrical loading conditions.