Dynamical coherent-potential approximation approach to excitation spectra in 3d transition metals

TL;DR

This paper develops a dynamical CPA method incorporating higher-order corrections to analyze excitation spectra in 3d transition metals, explaining experimental spectra and revealing effects of dynamical correlations and screening.

Contribution

The paper introduces an advanced dynamical CPA approach with higher-order corrections for accurate excitation spectra in 3d transition metals, improving upon previous static methods.

Findings

Dampening of main peaks in density of states due to dynamical effects.

Reduction of thermal spin fluctuation-induced band broadening.

Explanation of satellite features and spectral data matching experiments.

Abstract

First-principles dynamical CPA (Coherent-Potential Approximation) for electron correlations has been developed further by taking into account higher-order dynamical corrections with use of the asymptotic approximation. The theory is applied to the investigations of a systematic change of excitation spectra in $3d$ transition metals from Sc to Cu at finite temperatures. It is shown that the dynamical effects damp main peaks in the densities of states (DOS) obtained by the local density approximation to the density functional theory, reduce the band broadening due to thermal spin fluctuations, create the Mott-Hubbard type bands in the case of fcc Mn and fcc Fe, and create a small hump corresponding to the `6 eV' satellite in the case of Co, Ni, and Cu. Calculated DOS explain the X-ray photoelectron spectroscopy data as well as the bremsstrahlung isochromat spectroscopy data. Moreover, it is found that screening effects on the exchange energy parameters are significant for understanding the spectra in magnetic transition metals.