Semiconductor device technology has greatly developed in complexity since discovering the bipolar transistor. In this work, we developed a computational pipeline to discover stable semiconductors by combining generative adversarial networks (GAN), classifiers, and high-throughput first-principles calculations. We used CubicGAN, a GAN-based algorithm for generating cubic materials and developed a classifier to screen the semiconductors and studied their stability using first principles. We found 12 stable AA\({}^{\prime}\)MH₆ semiconductors in the F-43m space group including BaNaRhH₆, BaSrZnH₆, BaCsAlH₆, SrTlIrH₆, KNaNiH₆, NaYRuH₆, CsKSiH₆, CaScMnH₆, YZnMnH₆, NaZrMnH₆, AgZrMnH₆, and ScZnMnH₆. Previous research reported that five AA\({}^{\prime}\)IrH6 semiconductors with the same space group were synthesized. Our research shows that AA\({}^{\prime}\)MnH₆ and NaYRuH₆ semiconductors have considerably different properties compared to the rest of the AA\({}^{\prime}\)MH₆ semiconductors. Based on the accurate hybrid functional calculations, AA\({}^{\prime}\)MH₆ semiconductors are found to be wide-bandgap semiconductors. Moreover, BaSrZnH₆ and KNaNiH₆ are direct-bandgap semiconductors, whereas others exhibit indirect bandgaps.

自从发现双极晶体管以来，半导体器件技术的复杂性有了很大的发展。在这项工作中，我们开发了一种计算管道，通过结合生成对抗网络 (GAN)、分类器和高通量第一性原理计算来发现稳定的半导体。我们使用了 CubicGAN，这是一种基于 GAN 的生成立方材料的算法，并开发了一个分类器来筛选半导体并使用第一原理研究它们的稳定性。我们在F-43m空间群中发现了12个稳定的AA \({}^{\prime}\) MH _{6半导体，包括BaNaRhH 6}、BaSrZnH ₆、BaCsAlH ₆、SrTlIrH ₆、KNaNiH ₆、NaYRuH ₆, CsKSiH ₆ , CaScMnH ₆ , YZnMnH ₆ , NaZrMnH ₆ , AgZrMnH ₆ , 和 ScZnMnH ₆。先前的研究报道合成了五个具有相同空间群的 AA \({}^{\prime}\) IrH6 半导体。我们的研究表明，AA \({}^{\prime}\) MnH ₆和 NaYRuH ₆半导体与其他 AA \({}^{\prime}\) MH ₆半导体相比具有显着不同的特性。基于精确的混合泛函计算，AA \({}^{\prime}\) MH ₆半导体被发现是宽带隙半导体。此外，BaSrZnH ₆和KNaNiH ₆是直接带隙半导体，而其他则表现出间接带隙。