Pseudo-potentials based first-principles approach to the magneto-optical Kerr effect: from metals to the inclusion of local fields and excitonic effects

TL;DR

This paper introduces a first-principles DFT-based method to accurately predict the magneto-optical Kerr effect, incorporating local fields and excitonic effects, validated on metals like Fe, Co, Ni.

Contribution

It provides a formal expression for the dielectric tensor in terms of the density-density correlation function, enabling improved modeling of the Kerr effect including local and excitonic effects.

Findings

Pseudo-potential calculations accurately predict Kerr parameters for Fe, Co, Ni.

Derived an exact dielectric tensor expression for systems with an electronic gap.

Identified limitations in metals where anomalous Hall effect terms are neglected.

Abstract

We propose a first-principles scheme for the description of the magneto-optical kerr effect within density functional theory (DFT). Though the computation of Kerr parameters is often done within DFT, starting from the conductivity or the dielectric tensor, there is no formal justification to this choice. As a first steps, using as reference materials iron, cobalt and nickel we show that pseudo-potential based calculations give accurate predictions. Then we derive a formal expression for the full dielectric tensor in terms of the density-density correlation function. The derived equation is exact in systems with an electronic gap, with the possible exception of Chern insulators, and whenever the time reversal symmetry holds and can be used as a starting point for the inclusion of local fields and excitonic effects within time-dependent DFT for such systems. In case of metals instead we show that, starting from the density-density correlation function, the term which describes the anomalous Hall effect is neglected giving a wrong conductivity.