Improved calculations of mean ionization states with an average-atom model

TL;DR

This paper improves the calculation of mean ionization states in dense plasma and warm dense matter by developing and comparing three novel approaches within average-atom models, showing enhanced accuracy over traditional methods.

Contribution

It introduces three new methods for computing the mean ionization state in average-atom models, including a novel electron localization function approach and a Kubo-Greenwood conductivity extension.

Findings

New methods outperform canonical approach in accuracy

Electron localization function shows particular promise

Extensions align well with experimental data

Abstract

The mean ionization state (MIS) is a critical property in dense plasma and warm dense matter research, for example as an input to hydrodynamics simulations and Monte-Carlo simulations. Unfortunately, however, the best way to compute the MIS remains an open question. Average-atom (AA) models are widely-used in this context due to their computational efficiency, but as we show here, the canonical approach for calculating the MIS in AA models is typically insufficient. We therefore explore three alternative approaches to compute the MIS. Firstly, we modify the canonical approach to change the way electrons are partitioned into bound and free states; secondly, we develop a novel approach using the electron localization function; finally, we extend a method which uses the Kubo-Greenwood conductivity to our average-atom model. Through comparisons with higher-fidelity simulations and experimental data, we find that any of the three new methods usually out-performs the canonical approach, with the electron localization function and Kubo-Greenwood methods showing particular promise.