Atomistic Switching Mechanism of Ferroelectric Wurtzites

Ferroelectric wurtzite materials have gained immense interest in recent years for their potential in nonvolitile ferroelectric memory due to their CMOS compatibility, high thermal conductivity, and lack of a curie temperature. However, due to the novelty of their discovery, very little is known about the mechanism of the polarization switching. Here, we use a machine learned classical force field to simulate large scale atomistic switching events in bulk aluminum nitride. We uncover a nucleation and growth mechanism that defies traditional Avrami kinetics for ferroelectric switching and propose methods for designer device engineering to reduce coercive field and improve device applicability.