Intrinsic vacancy chalcogenides as dilute magnetic semiconductors: Theoretical investigation of transition-metal doped gallium selenide

Incorporating magnetism into semiconductors has been a major goal of research efforts aimed at achieving control of both spin and charge of carriers. An entirely new class of electronic materials known as dilutes magnetic semiconductors (DMS) has been synthesized and closely studied. This has led to a deeper understanding of semiconductor physics and the development of new magnetic mechanisms. Unfortunately, most DMS materials retain their magnetic ordering below room temperatures and have only limited solubility and structural compatibility with standard semiconductors. We investigate transition metal doped vacancy-ordered Ga 2Se3 as a material that can address these shortcomings. The intrinsic vacancies of this III-VI, zinc-blende-based semiconductor open possibilities for self-compensation as well as supply highly anistropic and polarizable band edge states. Ga2Se3 is also closely lattice matched to Si and may be grown heteroepitaxially on Si with high quality interfaces. Our first principles computations of X: Ga2Se3 (X = Mn, V, Cr, concentrations 5% to 16%) reveal that X atoms hybridize with neighboring Se in the p-d hybridization typical of III-V and II-VI DMS materials. This hybridization spin-polarizes states near the Fermi level in these T =0 calculations, and lowers the energy of the Se lone-pair orbitals that neighbor vacancies, reducing their prominent role in determining the properties of intrinsic Ga 2Se3. There are distinct differences between substitution on a vacancy or for a Ga. Anisotropic, hole-like conductivity is predicted when X is located in a Ga site, while for X situated in a vacancy, a half-metallic state with an isotropic conductivity appears likely. Our calculations suggest that Mn offers the best choice for the dopant, perhaps because its 3d 5 electronic configuration offers a large (∼0.5 eV) separation of spin up and spin down states near the Fermi level, reducing the metallic densities of states at the Fermi level for all doping concentrations. The large energy splitting suggests that doped Ga2Se3 may be a suitable material for spintronics applications at higher temperatures than these T = 0 initial calculations.