High-throughput interface prediction and generation scheme: The case of β -Ga2O3/AlN interfaces

Abstract

Solid–solid interfaces play a critical role in determining the performance of materials and devices. However, both experimental and theoretical determinations of interface structures are very challenging. In this Letter, we develop a high-throughput interface prediction and generation (IPG) scheme to predict the atomic structure of interfaces. To demonstrate the performance of the IPG, we apply it to the β-Ga2O3/AlN interface, which has attracted a great deal of attention for its promising application in power devices. Six β-Ga2O3/AlN interfaces are screened from thousands of candidates with the minimum mismatch and area tolerance. All these interfaces are modeled to satisfy the electron counting rules and then optimized by the minimum hopping method. The band alignments at the β-Ga2O3/AlN interfaces are in accordance with the available theoretical and experimental results. The results provide the basis for designing optical and electronic devices based on β-Ga2O3/AlN interfaces.