Spectral Analysis of Correlated One-Dimensional Systems with Impurities

TL;DR

This paper introduces a spectral analysis method for correlated one-dimensional systems with impurities, using an averaging procedure and numerical diagonalization to interpret experimental spectra and impurity effects.

Contribution

It presents a novel averaging approach and a few-pole approximation for analyzing spectral features in disordered one-dimensional spin systems.

Findings

Few-pole approximation effectively describes numerical spectral data.

Disorder-induced pseudo-gap and spectral weight loss are characterized.

Impurity distribution functions can be inferred from spectral features.

Abstract

An averaging procedure is proposed to account for spectral features of correlated one-dimensional systems in the presence of non-magnetic impurities. The dynamical spin structure factor for a corresponding random ensemble of Heisenberg chain segments is calculated by exact numerical diagonalization. It is shown that a few-pole approximation is sufficient to describe the numerical results. A similar analysis is proposed for the discussion of experimental spectra, such as obtained by inelastic neutron scattering measurements on Zn-doped CuO chains. By examination of the disorder-induced pseudo-gap, the loss of spectral weight, and the discrete peak structures due to smallest-cluster contributions, the underlying impurity distribution function can be determined.