The magnitude of (K) depends on the relative strengths of the interactive forces between the solute molecules and those of the two phases. The net effect of interactions between the solute molecules and those of the mobile phase (gas) are relatively very week, partly because the size of the mobile phase molecules are small and partly because the probability of collision between the solute molecules and the phase molecules is much less in a gas than in a liquid. As already stated, those solutes having a large distribution coefficient with respect to the stationary phase are held very strongly in the stationary phase and will be retained longer in the chromatographic system. Consequently, the stationary phase must be chosen to interact strongly enough with the solutes to retain them for a reasonable time but not so strong as to irreversibly hold them in the column. Nevertheless, it is clear that in any GC separation, it is the stationary phase that needs to be modified to adjust retention and resolution, not, as in liquid chromatography, the mobile phase. Molecular interaction is the result of intermolecular forces between molecules and it follows that it is necessary to understand the nature of these intermolecular forces before proceeding further to consider their relative magnitudes.
All intermolecular forces are electrical in nature. Although magnetic and gravimetric forces may also be present, it appears (to date) that their contributions to the total interactive force is minimal and, so far, have not been measurable. There are basically three different types of intermolecular force, dispersion forces, polar forces and ionic forces. All interactions between molecules are due to one, or a composite of two or all three of these types of molecular force.
This simple understanding of intermolecular forces has been clouded and confused by the introduction of different terms that are not based on force type, but on force strength. For example, strong dipole - dipole interactions which have almost the energy of a chemical bond have been termed hydrogen bonding but hydrogen bonding is not a different type of interaction, it is still only a strong form of polar interaction. At the other extreme, interaction between dipoles and induced dipoles (which will also be discussed later) have been termed (p)-(p) interactions, but they are simple polar interactions albeit weak: they are not a unique type of interaction. The strengths of polar interactions cover a wide range, and many names have been conjured up to describe the different interactive strengths. It must be understood, however, that they are all polar in nature (forces between charges on the molecule) and, from a molecular interaction standpoint, vary only in strength.
Finally, it must again be emphasized that there are only three basic types of molecular interactive forces, dispersive, polar and ionic. In CG, by selecting a stationary phase that will provide one or more of these forces having appropriate individual interactive strengths, the very subtle selectivity's necessary for entantiomeric separations can often be readily achieved.