Non-Drude behaviour of optical conductivity in Kondo-lattice systems

TL;DR

This paper investigates the optical conductivity in Kondo-lattice systems, revealing a non-Drude $rac{1}{ ext{frequency}}$ tail at high frequencies, which may relate to magnetic scattering in strange metals.

Contribution

It introduces a memory function formalism analysis of optical conductivity in Kondo lattices, highlighting non-Drude behavior linked to magnetic scattering mechanisms.

Findings

High frequency optical conductivity scales as 1/ω, deviating from Drude behavior.

The non-Drude tail is characteristic of strange metals.

Magnetic scattering may be key to understanding anomalous metallic behavior.

Abstract

The optical conductivity in a Kondo lattice system is presented in terms of the memory function formalism. I use Kondo-lattice Hamiltonian for explicit calculations. I compute the frequency dependent imaginary part of the memory function ($M^{\p\p}(\om)$), and the real part of the memory function $M^{\p}(\om)$ by using the Kramers-Kronig transformation. Optical conductivity is computed using the generalized Drude formula. I find that high frequency tail of the optical conductivity scales as $\sigma(\om) \sim \frac{1}{\om}$ instead of the Drude $\frac{1}{\om^2}$ law. Such a behaviour is seen in strange metals. My work points out that it may be the magnetic scattering mechanisms that are important for the anomalous behaviour of strange metals.