Complex ${\mathrm{Ga}}_2{\mathrm{O}}_3$ Polymorphs Explored by Accurate and General-Purpose Machine-Learning Interatomic Potentials

Abstract

${\mathrm{Ga}}_2{\mathrm{O}}_3$ is a wide-bandgap semiconductor of emergent importance for applications in electronics and optoelectronics. However, vital information of the properties of complex coexisting ${\mathrm{Ga}}_2{\mathrm{O}}_3$ polymorphs and low-symmetry disordered structures is missing. In this work, we develop two types of kernel-based machine-learning Gaussian approximation potentials (ML-GAPs) for ${\mathrm{Ga}}_2{\mathrm{O}}_3$ with high accuracy for $\beta$/$\kappa$/$\alpha$/$\delta$/$\gamma$ polymorphs and generality for disordered stoichiometric structures. We release two versions of interatomic potentials in parallel, namely soapGAP and tabGAP, for excellent accuracy and exceeding speedup, respectively. We systematically show that both the soapGAP and tabGAP can reproduce the structural properties of all the five polymorphs in an exceptional agreement with ab initio results, meanwhile boost the computational efficiency with $5\times 10^2$ and $2\times 10^5$ computing speed increases compared to density functional theory, respectively. The results show that the liquid-solid phase transition proceeds in three different stages, a "slow transition", "fast transition" and "only Ga migration". We show that this complex dynamics can be understood in terms of different behavior of O and Ga sublattices in the interfacial layer.

13 pages, 7 figures

References in corpus (8)

• Intrinsic Electron Mobility Limits in beta-Ga2O3
• An Accurate and Transferable Machine Learning Potential for Carbon
• Universal radiation tolerant semiconductor
• Structural, electronic, elastic, power and transport properties of $\beta$-Ga${}_2$O${}_3$ from first-principles
• Simple machine-learned interatomic potentials for complex alloys
• Highly tunable polarization-engineered two-dimensional electron gas in $\epsilon$-AlGaO3 / $\epsilon$-Ga2O3 heterostructures
• Multiscale machine-learning interatomic potentials for ferromagnetic and liquid iron
• Persistent room temperature photodarkening in Cu-doped \b{eta}-Ga2O3

Cited by in corpus (9)

• Threshold displacement energy map of Frenkel pair generation in $\mathrm{G}{\mathrm{a}}_2{\mathrm{O}}_3$ from machine-learning-driven molecular dynamics simulations
• Phase glides and self-organization of atomically abrupt interfaces out of stochastic disorder in $\alpha$-Ga${}_2$O${}_3$
• Dissimilar thermal transport properties in $\kappa$-Ga${}_2$O${}_3$ and $\beta$-Ga${}_2$O${}_3$ revealed by machine-learning homogeneous nonequilibrium molecular dynamics simulations
• Ultrahigh Stability of O-Sublattice in $\beta$-Ga${}_2$O${}_3$
• Exploring the energy landscape of aluminas through machine learning interatomic potential
• Orientation-Dependent Atomic-Scale Mechanism of $\beta$-${\mathrm{Ga}}_2{\mathrm{O}}_3$ Thin Film Epitaxial Growth
• Microtubes and nanomembranes by ion-beam-induced exfoliation of $\beta$-Ga${}_2$O${}_3$
• Improved capabilities of the TurboGAP code for radiation induced cascade simulations: an illustration with silicon
• Machine-learning-guided molecular dynamics simulations of point defect evolution in beta-Ga2O3 during ion implantation and annealing