Unraveling screening mechanisms in Kondo impurities using an NRG-MPS-based method

TL;DR

This paper introduces a new NRG-MPS-based method to identify and analyze Kondo screening mechanisms in impurity models, enhancing the study of complex strongly-correlated systems.

Contribution

The authors develop a straightforward methodology combining NRG and MPS techniques to better understand Kondo singlet formation and screening in impurity models.

Findings

Effective on single and two-level Anderson impurity models

Potential to generalize to multichannel and multiorbital impurities

Provides a robust framework for complex Kondo systems

Abstract

The Kondo effect is a hallmark of strongly-correlated systems, where an impurity's local degrees of freedom are screened by conduction electrons, forming a many-body singlet. With increasing degrees of freedom in the impurity, theoretical studies face significant challenges in accurately identifying and characterizing the underlying mechanisms that screen the impurity. In this work, we introduce a straightforward yet powerful methodology for identifying the formation of Kondo singlets and their screening mechanisms, by utilizing the numerical renormalization group (NRG) combined with the matrix product states (MPS) technique. We demonstrate the effectiveness of our method on the single and two-level Anderson impurity models (AIM). Furthermore, we discuss potential generalizations of the method to multichannel and multiorbital Kondo impurities. Harnessing advanced tensor network techniques, our approach extends to complex impurity systems, offering a robust and versatile framework for studying Kondo physics.