DFT beyond the ground state

Density functional theory (DFT) with its fundamental mathematical background in the Hohenberg-Kohn theorem has proven its success for more than 30 years for the parameter-free description of solids and molecules leading to the Nobel prize of Walter Kohn in 1998. Especially the properties related to the total energy of any solid like equilibrium volumes, lattice parameters, atomic positions, and elastic constants can be predicted very precisely. The deviance from experimental values is in the range of a few percent only. While the ground state properties are described very reliably by DFT, the treatment of excited states is a complex task. On the one hand, the theoretical background of describing excited states within DFT is not yet well settled, while, however, a number of different promising approaches is presently being developed. On the other hand, the use of ground state quantities for the description of excited states has been quite successful for a variety of materials and properties while it has failed badly for others. In this context, it is often not very clear to which extent errors are due to the over-interpretation of ground state properties or caused by the approximations used in the ground state calculation itself. The ability of treating excited states from first principles opens new research perspectives: Optical properties of solids are a major topic in basic research as well as in industrial applications. One of the main tasks in materials research is the tailoring of material properties to meet special needs such as the tuning of spectroscopic properties. Applications in displays, computer screens, lasers, or smart windows are mentioned only as a few out of many examples. Since the synthesis of new materials is usually very time-consuming and expensive, the experimental effort of finding a new compound suitable for a special purpose could be dramatically reduced if theory could reliably predict how to modify a certain crystal. By this procedure the synthesis of functional materials can be guided by theory. Also in order to be able to analyse measured spectra in their full range, experiments have to be accompanied by theoretical investigations. For these most challenging reasons a reliable theoretical tool for describing and predicting optical properties of solids is urgently needed. The goal of the present project is to develop tools to reliably describe and predict spectroscopic properties of solids. The work to be carried out consists of theoretical considerations, the development of codes and their application to a variety of materials. The problems to be investigated comprises the electron-hole interaction in molecular crystals, liner and non-linear optical properties of metals and semiconductors, Raman scattering in semiconductors, and photo-emission of superconductors. The work is strongly related to the Research and Training Network EXCITING funded by the European Community.

Density functional theory (DFT) with its fundamental mathematical background in the Hohenberg-Kohn theorem has proven its success for more than 30 years for the parameter-free description of solids and molecules leading to the Nobel prize of Walter Kohn in 1998. Especially the properties related to the total energy of any solid like equilibrium volumes, lattice parameters, atomic positions, and elastic constants can be predicted very precisely. The deviance from experimental values is in the range of a few percent only. While the ground state properties are described very reliably by DFT, the treatment of excited states is a complex task. On the one hand, the theoretical background of describing excited states within DFT is not yet well settled, while, however, a number of different promising approaches is presently being developed. On the other hand, the use of ground state quantities for the description of excited states has been quite successful for a variety of materials and properties while it has failed badly for others. In this context, it is often not very clear to which extent errors are due to the over-interpretation of ground state properties or caused by the approximations used in the ground state calculation itself. The ability of treating excited states from first principles opens new research perspectives: Optical properties of solids are a major topic in basic research as well as in industrial applications. One of the main tasks in materials research is the tailoring of material properties to meet special needs such as the tuning of spectroscopic properties. Applications in displays, computer screens, lasers, or smart windows are mentioned only as a few out of many examples. Since the synthesis of new materials is usually very time-consuming and expensive, the experimental effort of finding a new compound suitable for a special purpose could be dramatically reduced if theory could reliably predict how to modify a certain crystal. By this procedure the synthesis of functional materials can be guided by theory. Also in order to be able to analyse measured spectra in their full range, experiments have to be accompanied by theoretical investigations. For these most challenging reasons a reliable theoretical tool for describing and predicting optical properties of solids is urgently needed. The goal of the present project is to develop tools to reliably describe and predict spectroscopic properties of solids. The work to be carried out consists of theoretical considerations, the development of codes and their application to a variety of materials. The problems to be investigated comprises the electron-hole interaction in molecular crystals, liner and non-linear optical properties of metals and semiconductors, Raman scattering in semiconductors, and photo-emission of superconductors. The work is strongly related to the Research and Training Network EXCITING funded by the European Community.