Projective Quantum Monte Carlo Method for the Anderson Impurity Model and its Application to Dynamical Mean Field Theory

TL;DR

This paper introduces a projective quantum Monte Carlo algorithm for accurately computing ground state properties of the Anderson impurity model, enabling reliable zero-temperature solutions within dynamical mean field theory.

Contribution

The authors develop a new projective quantum Monte Carlo method of the Hirsch-Fye type for the Anderson impurity model and demonstrate its effectiveness in solving DMFT equations at zero temperature.

Findings

Rapid convergence to ground state enabling zero-temperature analysis

Reconfirmation of the Mott-Hubbard transition results

Method applicable to solving impurity problems in strongly correlated systems

Abstract

We develop a projective quantum Monte Carlo algorithm of the Hirsch-Fye type for obtaining ground state properties of the Anderson impurity model. This method is employed to solve the self-consistency equations of dynamical mean field theory. It is shown that the approach converges rapidly to the ground state so that reliable zero-temperature results are obtained. As a first application, we study the Mott-Hubbard metal-insulator transition of the one-band Hubbard model, reconfirming the numerical renormalization group results.