Multi-orbital Anderson models and the Kondo effect

TL;DR

This paper investigates multi-orbital Anderson models to understand the Kondo effect in transition metal compounds, introducing a semi-analytical impurity solver that reveals how model parameters influence spectral features and ground state regimes.

Contribution

It presents a new semi-analytical impurity solver for multi-orbital Anderson models, enabling detailed analysis of spectral features and ground state crossovers beyond traditional methods.

Findings

Spectral features depend critically on model parameters.

A crossover between different ground state regimes is identified.

The approach improves understanding of correlation effects in transition metal compounds.

Abstract

The low energy region of certain transition metal compounds reveals dramatic correlation effects between electrons, which can be studied by photoelectron spectroscopy. Theoretical investigations are often based on multi-orbital impurity models, which reveal modified versions of the Kondo effect. We present a systematic study of a multi-orbital Anderson-like model, based on a new semi-analytical impurity solver which goes beyond simple modifications of the well known NCA. We discuss one-particle excitation spectra and in particular the role of level positions and Coulomb-matrix elements. It is shown that the low-energy region as well as the overall features of spectra critically depend on the model parameters and on the quality of the approximations used. Recent photoelectron experiments and corresponding existing calculations are put into perspective. An interesting crossover scenario between different regimes of ground states with characteristically different local correlations is uncovered.