Towards Large-Scale and Spatio-temporally Resolved Diagnosis of Electronic Density of States by Deep Learning

TL;DR

This paper presents a deep learning approach for large-scale, spatio-temporally resolved diagnosis of electronic density of states, enabling efficient and accurate analysis of complex nonequilibrium states in materials.

Contribution

The study introduces a deep neural network model that accurately predicts electronic density of states and reveals local atomic environments during dynamic processes.

Findings

DNN accurately captures DOS in complex systems.

Atomic contributions to DOS serve as robust phase indicators.

Efficient large-scale, time-resolved DOS diagnosis achieved.

Abstract

Modern laboratory techniques like ultrafast laser excitation and shock compression can bring matter into highly nonequilibrium states with complex structural transformation, metallization and dissociation dynamics. To understand and model the dramatic change of both electronic structures and ion dynamics during such dynamic processes, the traditional method faces difficulties. Here, we demonstrate the ability of deep neural network (DNN) to capture the atomic local-environment dependence of electronic density of states (DOS) for both multicomponent system under exoplanet thermodynamic condition and nonequilibrium system during super-heated melting process. Large scale and time-resolved diagnosis of DOS can be efficiently achieved within the accuracy of ab initio method. Moreover, the atomic contribution to DOS given by DNN model accurately reveals the information of local neighborhood for selected atom, thus can serve as robust order parameters to identify different phases and intermediate local structures, strongly highlights the efficacy of this DNN model in studying dynamic processes.