Characterizing Short Necklace States in Logarithmic Transmission Spectrum of Strongly Localized Systems

TL;DR

This paper investigates the properties of short necklace states in localized systems, revealing their statistical characteristics and scaling behavior, and introduces methods applicable to Anderson localization studies.

Contribution

The study defines key quantities and uses two approaches to statistically analyze short necklace states, providing new insights into their properties and scaling in localized systems.

Findings

Short necklace states cause high transmission plateaus in spectra.

The distribution of plateau widths fits a sum of two Gaussians.

The typical plateau width remains nearly constant with increasing system length.

Abstract

High transmission plateaus exist widely in the logarithmic transmission spectra of localized systems. Their physical origins are short chains of coupled-localized-states embedded inside the localized system, which are dubbed as "short necklace states". In this work, we define the essential quantities and then, based on these quantities, we investigate the short necklace states' properties statistically and quantitatively. Two different approaches are utilized and the results from them agree with each other very well. In the first approach, the typical plateau-width and the typical order of short necklace states are obtained from the correlation function of logarithmic transmission. In the second approach, we investigate statistical distributions of the peak/plateau-width measured in logarithmic transmission spectra. A novel distribution is found, which can be exactly fitted by the summation of two Gaussian distributions. These two distributions are the results of sharp peaks of localized states and the high plateaus of short necklace states. The center of the second distribution also tells us the typical plateau-width of short necklace states. With increasing the system length, the scaling property of typical plateau-width is very special since it almost does not decrease. The methods and the quantities defined in this work can be widely used on Anderson localization studies.