Potential Surface Studies of Open Shell Systems

The electronic structure of a large number of organic radicals was examined by using the developed unrestricted version of the semiempirical MIND0/3 method. A review of the obtained results is presented. The calculated heats of formation (/I.Hr) are generally in good agreement with available experimental data. The estimated /I.Hr values are usually too low - 40 kJ/mol. This is not a serious drawback and had to be expected from the single determinant approach. However, it only shifts the origin of the energy scale. The relative stabilities of radical isomers are in most cases predicted quite correctly. Theoretical interatomic distances are in reasonably good accordance with experiment. Particular emphasis is placed on the discussion of the reactivity of radicals and reaction paths. Theoretical results are here of utmost importance because the elementary reactionsteps in radical rearrangements are extremely difficult to access by conventional experimental techniques. An original procedure for searching and verifying local minima and saddle points on the energy hypersurfaces is discussed. This approach is applied on the series of w-alkenyl radicals and their routes of cyclisations were throughly investigated. It was found that the experimental findings can be rationalised by a simple orbital overlap model which on the other hand leads to a natural link to the allowed/forbidden concept. These findings are corroborated by the available pieces of experimental evidence. An important outcome of our analyses is, that unlike for closed shell systems, the activated complexes of forbidden reactions for doublet states can be successfully described by single configurations and should therefore be accessible by appropriate semiempirical treatments