Methods

A major asset of our institute is the competence in developing conceptual and computational methods based on static and time-dependent density functional theory (DFT), and reduced density matrix functional theory (RDMFT), diagrammatic many-body perturbation theory (e.g. GWapproximation), dynamical mean field theory (DMFT) in combination with exact diagonalization schemes to deal with the quantum impurity problem, classical Monte Carlo techniques, and molecular and spin dynamics, paying particular attention to the description of increasingly complex solids and to massively parallel high performancecomputers that will provide petaflop performances.

The complexity dimension of solids and nanostructures addresses their structural and chemical complexity and disorder, the dimensionality of solids, the topological complexity of the electronic and magnetic structure, and solids whose properties are a result of a manifestation of an increasingly complex electron-electron interaction (e.g. van der Waals interaction, orbital ordering).

Die Möglichkeit der Methodenentwicklung bietet eine hervorragende Ausbildung für unsere Studenten. Die Entwicklung neuer Methoden gibt uns die Möglichkeit eine sehr innovativen Forschung. Es ermöglicht auch, das reiche Spektrum an Feststoffen (z. B. Moleküle, Cluster atomarer Skala, Graphen, Oberflächen, Grenzflächen und Heterostrukturen von Halbleitern, Isolatoren, von magnetischen, ferroelektrischen und multiferroischen Materialien) und das reiche Spektrum von Phänomenen (z.B. strukturelle und chemische Unordnung, dynamische Instabilitäten, elektronische Anregungen, nichtkollinearen Magnetismus, Spinwellen, Spinrelaxation und Spindynamik, ballistischen und transversalen Transport, Mott-Übergang und Orbitalordnung).

To achieve these goals we are member of the Institute for Advanced Simulation (IAS). We have developed and are developing a set of computational methods in particular density functional theory based methods that include hybrid functionals, exact exchange in the optimized effective potential formulation, Green-function embedding, or the calculation of  topological invariants or run on BlueGene computers. We collaborate also with experts on numerical mathematics and experts on computer architectures and technologies e.g. of the Exascale Innovation Center.

Many of our methods are available for use by the scientific community and are collected at the page juDFT.