Geerd HF Diercksen

Max-Planck-Institut für Astrophysik, D-85740 Garching, GERMANY

Computational methods have become widely accepted as a research tool to study the properties and processes of atoms and molecules, both in the academic environment and increasingly in industrial research and development. The properties of isolated, small molecules in their electronic ground state may be investigated reliably and with spectroscopic accuracy using quantum mechanical models. For studying the properties of larger molecules including about ten up to some hundred of atoms more approximate quantum mechanical models are required. Provided some care is taken in choosing and parametrizing the theoretical models used for describing the studied molecules valuable information about various properties may be obtained.

For the theoretical study of the properties and processes of atoms and molecules in their excited electronic states much less experience is available than for studying the electronic ground state. This is partly due to the fact that the reliable description of excited molecular states is more elaborate than that of the electronic ground state and, in general, requires a fair understanding of the electronic structure of various interacting excited electronic states prior to the computational study.

Finally, the theoretical investigation of the properties and processes of molecules in solution is, in general, not feasible by a purely quantum mechanical model. In such case, usually, the properties of the solute molecule are treated by a quantum mechanical model while the effect of the solvent on the solute is taken into account approximately by classical mechanics. Again, developing and applying adequate theoretical models and interpreting the computed results requires careful understanding of the physics and chemistry underlying the studied properties and processes.

By using the case of Hydrogen/Rare-gas Rydberg clusters the sensible and reliable use of quantum mechanical models for studying Rydberg states of small molecules and clusters will be demonstrated and results will be presented. In view of the increasing use of computational methods in theoretical chemistry by novices in the field, due to low cost computing power and freely available software, it is argued that intelligent software is required to assist novice users in the sensible choice and application of computationally feasible theoretical models and for interpreting the results. The possible structure of such intelligent software will be indicated.