A graphic in the paper entitled "Device-scale atomistic modeling of phase-change memory materials". 

A research paper entitled "Device-scale atomistic modeling of phase-change memory materials" by Xi'an Jiaotong University (XJTU) and the University of Oxford was published online in Nature Electronics on Sept 25. 

Computer simulations can play a central role in the understanding of phase-change materials and the development of advanced memory technologies. However, direct quantum-mechanical simulations are limited to simplified models containing a few hundred or thousand atoms. 

To this end, Professor Zhang Wei from the Center for Alloy Innovation and Design of the State Key Laboratory for Mechanical Behavior of Materials at XJTU and Professor Volker L. Deringer from the Department of Chemistry at the University of Oxford collaborated to develop a machine-learning-based potential model that is trained using quantum-mechanical data and can be used to simulate a range of germanium-antimony-tellurium compositions-typical phase-change materials under realistic device conditions.

The model combines the computational efficiency of traditional force fields with the computational accuracy of quantum mechanics. It accurately describes the complex structural characteristics and reversible phase-change processes of germanium-antimony-tellurium alloys with arbitrary compositions along the pseudo-binary line. It also has high scalability, allowing for the simulation of the effects of memory device interfaces on the phase-change process, which can help optimize the size and structure of memory devices. 

This work was completed by doctoral student Zhou Yuxing as the first author, under the guidance of professors Zhang Wei (corresponding author), Ma En, and Volker L. Deringer (corresponding author). 

Currently, all the databases and potential functions obtained from this work have been open-sourced. The open-source data can be freely downloaded from the platforms of Zenodo and Alkemie.