Research

Electronic structure work at CPHT focuses on an ab initio (i.e. without adjustable parameters) description of the properties of crystalline materials with strong electronic Coulomb correlations. It extends both to challenging applications in materials physics and to the further development of suitable methods to tackle them. The latter involves advances towards a truly parameter-free description of the electronic Coulomb interactions, in the framework of the ERC project “Predictive electronic structure calculations for materials with strong electronic correlations: Long-range Coulomb interactions and many-body screening” (S. Biermann), as well as the elaboration of new tools to make different properties accessible to calculations.

Examples of materials applications range from questions inspired "real" problems in materials science to puzzles of interest in fundamental science. Illustrative examples include:

  1. Transition metal oxides with partially filled 3d, 4d or 5d shells.
  2. Rare-earth iron alloys for permanent magnet applications.
  3. f-electron materials for applications as pigments.
  4. Spin-crossover molecules on surfaces
  5. Low-dimensional systems such as adatom layers on semiconductor surfaces

 

At the heart of current methodological questions the CPHT scientists are concerned with are two -- interrelated -- issues concerning the concept of combining electronic structure and many-body techniques:

  1. Techniques for systematic (even though in general approximate) downfolding of higher energy degrees of freedom in order to define a double-counting-free interface between the continuum electronic structure and many-body effects at lower energy scales.
  2. Advances to include long-range interactions, most notably ligand-to-correlated-shell interactions and non-local interactions coupling neighboring correlated shells.

 

Promising recent achievements include the development of the combined many-body perturbation theory + dynamical mean field theory scheme ("GW+DMFT"), and its derivatives.  For example, the combination of a screened exchange Hamiltonian with dynamical mean field theory – reviewed in Delange et al., J. Phys. Soc. Jpn. 87, 4, 041003 (2018) – allows for first principles calculations of spectral properties of materials of similar accuracy as "GW+DMFT" but at much cheaper computational cost.

English