Theoretical considerations of the effects of the medium upon molecular energy levels

For several years attempts have been made to allow for the effect of a solvent upon the energy levels of a solute atom or molecule by treating it as a medium with an effective dielectric constant. The solute is normally placed (figure 31) at the centre of a spherical cavity. One obvious advantage of such a model is that it attempts right from the start to deal with the many-body character of the forces which operate at very high gas densities and in liquids. But, as recent work by Bellamy (this Discussion) shows, such a non-specific explanation for the frequency shifts in vibration spectra is inadequate. This is not surprising. Indeed, one of the first requirements is that we should be able to distinguish those situations in which an effective-dielectric-constant approach is likely to be valid from the other situations where it is not. The latter situations are far more numerous than the former.