P3: Many-body approaches to realistic quantum impurity models

Supervision:Gerd Czycholl, Tim Wehling

Motivation and state of the art: Open-shell transition metal impurities in 2d materials and at oxide surfaces provide localized magnetic moments and hold promises in the context of novel solid-state information storage as well as quantum information processing. The theoretical description of these systems is, however, extremely challenging since it requires us to account for electronic correlations, spin-orbit coupling, crystal fields and hybridization effects. Combinations of ab-initio techniques with Anderson impurity models (AIM) can in principle address strong electronic correlations and real material aspects on equal footing. However, realistic transition metal systems require multi-orbital AIMs, which are notoriously difficult to solve particularly in cases of low symmetry and/or strong spin-orbit coupling. Thus, the theoretical understanding of key quantities like magnetic anisotropies or spin-relaxation times remains an open problem. To change this situation, new and improved treatments for AIMs will be developed and applied.

Aims and work plan:The project will develop a theory of ground state properties, electronic excitations, magnetic anisotropies and spin-dynamics of single magnetic adatoms on /impurities in 2d materials and at oxide surfaces. To this end, we will develop approximations for realistic AIMs by mapping them on exactly solvable reference systems containing correlation / interaction terms using variational principles. New bath discretization schemes for exact diagonalization (ED) treatment of realistic AIMs shall be developed. We will implement calculations of total energies, excitation spectra and transport properties for magnetic impurities in and adsorbates on complex transition metal oxide (P6, P8, P9, P10) and dichalcogenid (P1, P4, P5, P11) hosts. ED approaches will be advanced to the treatment of optically excited systems using Green function based approaches in collaboration with P5. Applicability of AIM based approaches to describe chemical reactions at oxide surfaces via dynamical mean field theory will be explored together with P8.