Enabling Ab Initio Material Design of InAs/GaSb Superlattices for Infrared Detection

Abstract

Type-II superlattices (T2SLs) exemplified by $\mathrm{In}\mathrm{As}/\mathrm{Ga}\mathrm{Sb}$ are promising for infrared detection as the cutoff wavelength can simply be tuned by the layer thickness. Computational material design is thus highly desirable, but traditional methods such as tight binding and k\ifmmode\cdot\else\textperiodcentered\fi{}p depend on the choice of parameters. Ab initio density-functional-theory (DFT) calculations, on the other hand, suffer from a dilemma because band-gap accuracy and computational efficiency cannot be reached simultaneously due to band-gap underestimation. Here we show the great potential of the shell DFT-1/2 method in the material design of T2SLs. With a single parameter input, the lattice constant, or even without parameter, the method yields satisfactory electronic structures for a series of $\mathrm{In}\mathrm{As}/\mathrm{Ga}\mathrm{Sb}$ T2SLs considering spin-orbit coupling, with the band-gap mismatch limited to 5% compared with our industry-class samples. Detailed miniband analysis is presented based on the ab initio results. For covalent semiconductors without strong correlation, shell DFT-1/2 proves to be an excellent candidate for accurate band-structure calculation with local-density-approximation-level efficiency. It enables the material design of III-V superlattice materials at the engineering level without introducing empirical parameters.