Text

theses

This thesis contains several investigations of magnetoelectric effects and ferroelectricity in complex oxides studied via first-principles calculations. We start by reviewing the mechanisms of ferroelectricity and magnetoelectric e ffects, and then we give a brief introduction to the first-principles computational methods that are involved. Next, our investigations are divided into two parts. The first half focuses on the magnetoelectric effects, while the second half is mainly on ferroelectricity. The first half aims to examine the lattice contribution to the magnetoelectricity by investigating the dynamical magnetic charge tensors induced by different mechanisms. Through the study of Cr2O3 and a fictitious material KITPite, we find that the dynamical magnetic charges driven by exchange striction are more significant than the ones induced by spin-orbit coupling. Since the lattice contribution to the magnetoelectric effect is proportional to the dynamical magnetic charges, we also study the magnetic charges and the magnetoelectric coupling in hexagonal manganite RMnO3 and ferrite RFeO3. Our results further confirm the importance of the exchange-striction mechanism in inducing large magnetic charges, but we also notice that the magnetoelectric contributions from various phonons tend to cancel each other, leading to a great reduction of the total coupling. These investigations not only provide a prediction of the magnetoelectric coupling constant in RMnO3 and RFeO3, but also emphasize the importance of phonons in magnetoelectric coupling. In the second half of the thesis, we focus on predicting new ferroelectrics in the family of corundum derivatives. Many new corundum derivatives have been synthesized recently; these are automatically polar, and many are magnetic as well. However, a polar material is only called ferroelectric if the polarization is reversible by an external field, and it is not yet clear whether or not this is the case for these new materials. Motivated by this question, we use a structural constraint method to study the ferroelectric reversal path and energy barrier of several corundum derivatives. As a result, we predict several FE candidates with insulting reversal paths and low barrier energies. Since the hysteresis behavior of ferroelectrics is attributed to the ferroelectric domain wall motion, we further investigate the formation and motion of ferroelectric domain walls in corundum derivatives. Our study predicts the atomic structure and orientation of the ferroelectric domain wall, as well as the shape of ferroelectric domains. In addition, we fi nd novel properties at domain walls, including a strong magnetoelectric coupling and an interlocking between chirality and polarization. Moreover, we use the structural constraint method to study the barrier energy of ferroelectric domain wall reversal. Our results suggest that the barrier energy is linearly correlated with the bond valence sum, which can be used as a guide to find new ferroelectrics in the family of corundum derivatives.

ETD_7597

Ph.D.

Includes bibliographical references

by Meng Ye

doi:10.7282/T3ZK5JZQ

ETD doctoral

The author owns the copyright to this work.