An alternative approach for the determination of mean free paths of electron scattering in liquid water based on experimental data

TL;DR

This paper introduces a new first-principles method combining ab-initio calculations and Monte Carlo simulations to accurately determine electron mean free paths in liquid water, addressing discrepancies in previous estimates.

Contribution

It presents a novel approach that integrates ab-initio scattering calculations with experimental data to reliably estimate electron mean free paths in liquid water.

Findings

IMFP is larger than EAL in liquid water.

EMFP exceeds that of gas-phase water.

Both EMFP and IMFP are shorter than in amorphous ice estimates.

Abstract

The mean free paths of low-energy electrons in liquid water are of fundamental importance for modelling radiation damage and many related physico-chemical processes. Neither theoretical predictions nor experimental estimations have so far converged to yield reliable values for these parameters. We therefore introduce a new approach to determine the elastic and inelastic mean free paths (EMFP, IMFP) based on experimental data. We report extensive ab-initio calculations of electron quantum scattering with water clusters, which are brought to convergence with respect to the cluster size. This provides a first-principles approach to condensed-phase scattering that includes both multiple-scattering and condensation effects. The obtained differential cross sections are used in a detailed Monte-Carlo simulation to extract EMFP and IMFP from two recent liquid-microjet experiments that determined the effective attenuation length (EAL) and the photoelectron angular distribution (PAD) following oxygen 1s-ionization of liquid water. For electron kinetic energies from 10 eV to 300 eV, we find that the IMFP is noticeably larger than the EAL. The EMFP is longer than that of gas-phase water and the IMFP is longer compared to the latest theoretical estimations, but both the EMFP and IMFP are much shorter than suggested by experimental results for amorphous ice. The Python module developed for the analysis is available at https://gitlab.com/axelschild/CLstunfti and can be used to further refine our results when new experimental data become available.