Efeitos não-adiabáticos em moléculas leves

Abstract

For light molecules, the main discrepancy between theoretical and experimental data is due to the neglect of adiabatic and non-adiabatic corrections in the calculation of the rovibrational energy levels. While adiabatic corrections do not present a serious problem for small molecules, the computation of non-adiabatic corrections is still a con- siderable challenge. Starting from the separation of the motions of atomic cores and valence electrons, a new non-adiabatic theory has been developed in this thesis. In the underlying physical model, an electronic core mass is added to the nuclear mass when solving the nuclear motion problem. Unlike standard methods, our approach adds no additional cost to standard computations and can be extended easily to polyatomic systems. Applying the method to the diatomic molecules H+2 , H2 and isotopologues, excellent agreement with experimental data is found. For the triatomic ion H+3 , a coordinate-dependent core mass surface has been constructed for the first time, which is capable of producing the rotational transitions of the 1/2 fundamental band with an accuracy of nearly 0:001 cm-1. For the heteronuclear molecule LiH, which is proble- matic due to an interplay of covalent and ionic structures, a mass model has been developed based on valence bond theory. Far away from the avoided crossing region, this mass model improves significantly previous ab initio predictions and provides data in excellent agreement with experiment.