IMF2017

On moving to information technology era and the green-and-clean environment, high capacity, fast to access, and low power consumption digital media has gained great interest. The solid state drive consisting of ferroelectric layers has been chosen as promising candidate. However, due to the complexity of the ferroelectric structure, there are some incomplete pictures in describing/predicting ferroelectric phenomena especially in low dimensions. For instance, the generally considered Landau-Hamiltonian discards thermal fluctuation which is important close to the critical point, and the spin-Hamiltonian lacks of appropriate thermodynamic-limit description of ferroelectric phase transition. This work then combines both Hamiltonians and investigates using Monte Carlo simulation. With applied periodic electric field, the dynamic hysteresis behavior was investigated as functions of films’ thickness, temperature, and field parameters. The dynamic critical point, separating the symmetric from the asymmetric hysteresis loop, was extracted as varying field parameter and films’ thickness to draw dynamic phase diagrams. Artificial Neural Network was used to draw further details of hysteresis profiles among simulated conditions. The results show that dynamic critical point becomes higher with increasing field frequency as more thermal/electrical energy is required to compensate the faster field switching. Further, the critical point also increases with increasing the films’ thicknesses due to the stronger ferroelectric interaction in thicker films. Different dipole interaction ranges show different critical magnitude but agree qualitatively. In addition, the results are in good qualitative agreement with experiments on ferroelectric films, where applicable.