Quench-rate and size-dependent behaviour in glassy Ge2Sb2Te5 models simulated with a machine-learned Gaussian approximation potential

Mocanu, Felix C.; Konstantinou, Konstantinos; Elliott, S. R.

doi:10.1088/1361-6463/ab77de

Cited by 30 publications

(29 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…285 Subsequently, simulations with simulation-cell sizes up to 24,300 atoms were carried out using the same potential, systematically addressing the role of the simulation-cell size as well as that of the quench rate on the resulting structures. 298 …”

Section: Applications (I): Force Fieldsmentioning

confidence: 99%

Gaussian Process Regression for Materials and Molecules

et al. 2021

View full text Add to dashboard Cite

We provide an introduction to Gaussian process regression (GPR) machine-learning methods in computational materials science and chemistry. The focus of the present review is on the regression of atomistic properties: in particular, on the construction of interatomic potentials, or force fields, in the Gaussian Approximation Potential (GAP) framework; beyond this, we also discuss the fitting of arbitrary scalar, vectorial, and tensorial quantities. Methodological aspects of reference data generation, representation, and regression, as well as the question of how a data-driven model may be validated, are reviewed and critically discussed. A survey of applications to a variety of research questions in chemistry and materials science illustrates the rapid growth in the field. A vision is outlined for the development of the methodology in the years to come.

show abstract

Section: Applications (I): Force Fieldsmentioning

confidence: 99%

Gaussian Process Regression for Materials and Molecules

et al. 2021

View full text Add to dashboard Cite

show abstract

“…However, a previous analysis on Ge 2 Sb 2 Te 5 shows minor changes in the structural properties of the amorphous phase by doubling the quenching rate from 5 K/ps to 10 K/ps. 43 Reducing the quenching rate to 1 K/ps lead instead to partial crystallization as this material is a very poor glass former. 43 In another work, 44 it was shown that two amorphous models of GeTe generated by quenching from 1000 K to 300 K in either 100 ps or 3 ns featured very similar structural properties provided that in the long quench (3 ns) the quenching rate was faster (≥ 1 K/ps) in the 700-500 K range to prevent crystallization.…”

Section: Computational Detailsmentioning

confidence: 99%

“…43 Reducing the quenching rate to 1 K/ps lead instead to partial crystallization as this material is a very poor glass former. 43 In another work, 44 it was shown that two amorphous models of GeTe generated by quenching from 1000 K to 300 K in either 100 ps or 3 ns featured very similar structural properties provided that in the long quench (3 ns) the quenching rate was faster (≥ 1 K/ps) in the 700-500 K range to prevent crystallization. Several simulations of the supercooled liquid phases of GeTe and Ge 2 Sb 2 Te 5 show that the time scale of few tens of ps used in our production runs is much shorter than the typical incubation time (several hundreds of ps) for the formation of crystalline nuclei in the temperature range 700-500 K where nucleation rate and crystal growth velocities are high.…”

Section: Computational Detailsmentioning

confidence: 99%

Metal-semiconductor transition in the supercooled liquid phase of the Ge2Sb2Te5 and GeTe compounds

Cobelli

Dragoni

Caravati

et al. 2021

Phys. Rev. Materials

View full text Add to dashboard Cite

The Ge2Sb2Te5 and GeTe compounds are of interest for applications in phase change memories. In the reset process of the memory the crystal is rapidly brought above the melting temperature Tm by Joule heating and then the liquid phase rapidly cools down leading to the formation of the amorphous phase. Since the liquid above Tm is metallic and the amorphous phase is semiconducting a semiconductor-to-metal transition occurs in the supercooled liquid. Based on density functional simulations, we estimated the metal-semiconductor transition temperature TM−SC by monitoring the opening of a band gap in the supercooled liquid phase. Due to previous evidence on the importance of the van der Waals (vdW) interaction in describing the liquid phase of these materials, we used both the revised Vydrov-van Voorhis functional which includes vdW non-local interactions and the Perdew-Burke-Ernzerhof functional without vdW corrections. The estimated TM−SC is about 100-150 K higher with the former than with the latter framework for both compounds. By including vdW interactions the estimated TM−SC is closer to Tm than to the glass transition temperature for both systems. The analysis of the structural properties as a function of temperature suggests a correlation between the metal-semiconductor transition and a Peierls distortion. However, the data support more a continuous structural transformation than the presence of a first order liquid-liquid phase change associated the metal-semiconductor transition.

show abstract

“…[ 6 ] It has also been used to simulate very large models of g‐ GST (up to 24 300 atoms), and glassy models of GST quenched from the liquid at rates as slow as 1 K ps −1 . [ 7 ]…”

Section: Introductionmentioning

confidence: 99%

Simulation of Phase‐Change‐Memory and Thermoelectric Materials using Machine‐Learned Interatomic Potentials: Sb₂Te₃

Konstantinou

Mavračić

Mocanu

et al. 2020

Physica Status Solidi (b)

Self Cite

View full text Add to dashboard Cite

Density‐functional‐theory (DFT)‐based, ab initio molecular dynamics (AIMD) simulations of amorphous materials generally suffer from three computer‐resource‐related limitations due to their O(N3) cubic scaling with model system size, N. They are limited to a maximum model size of N ≈500 atoms; they are limited to time scales <1 ns; and, usually, only a single model can be simulated in any one investigation. This article discusses a machine‐learned, linear‐scaling (O(N)), DFT‐accurate interatomic potential (a Gaussian approximation potential, GAP), originally developed by Mocanu et al. [J. Phys. Chem. B 2018, 122, 8998] using a Gaussian process regression method for the ternary phase‐change‐memory material Ge2Sb2Te5 (GST). The chemical transferability of this GAP potential is explored in an application to the case of simulating amorphous models of the phase‐change‐memory and thermoelectric material Sb2Te3, an end‐member of the GST compositional tie‐line GeTe–Sb2Te3. The GAP‐model results are compared with those obtained from conventional DFT‐based AIMD simulations.

show abstract

Quench-rate and size-dependent behaviour in glassy Ge₂Sb₂Te₅ models simulated with a machine-learned Gaussian approximation potential

Cited by 30 publications

References 53 publications

Gaussian Process Regression for Materials and Molecules

Gaussian Process Regression for Materials and Molecules

Metal-semiconductor transition in the supercooled liquid phase of the Ge2Sb2Te5 and GeTe compounds

Simulation of Phase‐Change‐Memory and Thermoelectric Materials using Machine‐Learned Interatomic Potentials: Sb₂Te₃

Contact Info

Product

Resources

About

Quench-rate and size-dependent behaviour in glassy Ge2Sb2Te5 models simulated with a machine-learned Gaussian approximation potential

Cited by 30 publications

References 53 publications

Gaussian Process Regression for Materials and Molecules

Gaussian Process Regression for Materials and Molecules

Metal-semiconductor transition in the supercooled liquid phase of the Ge2Sb2Te5 and GeTe compounds

Simulation of Phase‐Change‐Memory and Thermoelectric Materials using Machine‐Learned Interatomic Potentials: Sb2Te3

Contact Info

Product

Resources

About

Quench-rate and size-dependent behaviour in glassy Ge₂Sb₂Te₅ models simulated with a machine-learned Gaussian approximation potential

Simulation of Phase‐Change‐Memory and Thermoelectric Materials using Machine‐Learned Interatomic Potentials: Sb₂Te₃