Condensed Matter Seminar: The Road to a Predictive Theory of Strongly Correlated Electron Materials
Prof. Gabriel Kotliar, Rutgers University and Brookhaven National Laboratories
Abstract:
The prediction of the physical properties of solids without experimental input is a longstanding challenge in theoretical physics. Weakly correlated materials, are well understood in terms of the Fermi Liquid paradigm and their properties can be accurately computed with modern implementations of density functional theory and perturbation theory in the screened Coulomb interactions, to the point that designing materials with desirable properties is now possible. This paradigm fails for a class of compounds, known as strongly correlated systems.
Correlated electron systems are materials for which the previous paradigm fails dramatically. They continue to be discovered accidentally and continue to surprise us with their exceptional physical properties, ranging from high temperature superconductivity to anomalous thermoelectricity, and their potential for new applications. They pose one of the most difficult non-perturbative challenges in physics.
In this colloquium I will give an elementary introduction to the field of strongly correlated electron materials and Dynamical Mean Field Theory (DMFT) a non perturbative method which provides a zeroth order picture of the strong correlation phenomena in close analogy with the Weiss mean field theory in statistical mechanics. Applications materials containing f and d electrons will be presented. We will show how the anomalous properties of correlated materials emerge from their atomic constituents and conclude with an outlook of the challenges ahead and the perspectives for a rational material design.
