Litcius/Paper detail

Threshold displacement energy map of Frenkel pair generation in <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" altimg="si51.svg" display="inline" id="d1e526"><mml:mi>β</mml:mi></mml:math>-Ga2O3 from machine-learning-driven molecular dynamics simulations

Huan He, J. Zhao, Jesper Byggmästar, Ru He, K. Nordlund, Chaohui He, Flyura Djurabekova

2024Acta Materialia26 citationsDOIOpen Access PDF

Abstract

β-gallium oxide (β-Ga2O3) shows great promise for electronics applications, particularly, in future space operating devices exposed to harsh radiation environments for extended times. This study focuses on crucial, yet not fully explored, aspects of radiation damage in this material, such as threshold displacement energies and formation of various radiation-induced Frenkel pairs. Analyzing over 5,000 molecular dynamics simulations based on our machine-learning potentials, we conclude that the threshold displacement energies for two Ga sites, the tetrahedral (22.9 eV) and octahedral (20 eV) ones, differ stronger than the same values for three different O sites, which range only between 17 eV and 17.4 eV. Mapping of threshold displacement energies unveils significant differences in displacements for all five atomic sites. Our newly developed defect identification methodology successfully classified multiple Frenkel pair types in β-Ga2O3, with over ten different Ga and two primary O ones with a predominant O split interstitial at the O1 site. Finally, the calculated recombination energy barriers suggest that O Frenkel pairs are more likely to recombine upon annealing than Ga. These insights are pivotal for understanding the radiation damage and defect formation in Ga2O3, providing the basis for design of Ga2O3-based electronics with high radiation resistance.

Topics & Concepts

Displacement (psychology)Scalable Vector GraphicsMaterials scienceEnergy (signal processing)Computer graphics (images)PhysicsComputer scienceWorld Wide WebQuantum mechanicsPsychotherapistPsychologyGa2O3 and related materialsElectronic and Structural Properties of OxidesMachine Learning in Materials Science