Isolated Defects in 2D-3D: Real Space Green's Functions Approach
Joe Sink, Ph.D.
The behavior of dilute atomic or molecular defects, particularly in the limit of a single or a few interacting impurities within a host material, while challenging, is critical not only for developing methods to mitigate undesired effects[1] but also to allow better control in systems where these defects may be used as single atom/molecular quantum devices: such as magnetometers[2], transistors[3] and quantum emitters[4]. Ab initio density functional theory (DFT) methods-such as those implemented in Quantum ESPRESSO (QE) and the Vienna Ab initio Simulation Package (VASP)-are both indispensable and widely employed to investigate these systems. One key advantage of DFT is that, in principle, the only required inputs are the atomic species and an initial geometry. However, this convenience comes with significant computational cost, approximately scaling as $\mathcal{O}(Volume^3)$. Additionally, DFT calculations must be carefully converged with respect to the size of the simulation cell to minimize artificial interactions or hybridization between periodic images of the defect.
As an alternative, some defects are well treated as a localized perturbation to the pristine (bulk) system[5,6]. Crucially, because this perturbation is introduced in a post-processing step, the expensive DFT calculation can be confined to the primitive unit cell of the host material. One such method involves computing real-space Green’s functions derived from tight-binding models-either semiempirical or constructed via Wannierization of DFT results [citations]. I will present ongoing work applying this formalism to a variety of systems, including vacancies in wide-band gap III–V semiconductors and magnetic transition metal impurities on superconducting thin films. In cases where cross-sectional scanning tunneling microscopy (STM) data is available, we find that the simulated defect wave functions exhibit excellent agreement with experimental observations.
[1]IEEE TRANSACTIONS ON NUCLEAR SCIENCE, V.69, N.3 (2022)
[2]Scientific Reports volume 6, 37077 (2016)
[3]Nature Nanotechnology volume 7, 242–246 (2012)
[4]Phys. Rev. Applied 20, 014058 (2023)
[5]J. Vac. Sci. Technol. 19, 502–507 (1981)
[6]Phys. Rev. Lett. 92, 216806 (2004)