The Mechanism of Chromatographic Retention - Mixed Phases > Page 24

The simple linear relationship is not altogether surprising as the volume fraction of each phase would be expected to determine the probability that a given solute molecule will interact with a molecule of that phase. This is analogous to the partial pressure of a solute in a gas determining the probability that a solute molecule will collide with a gas molecule. Thus, doubling the concentration of one phase doubles the probability of solute interaction with that phase and consequently doubles its contribution to retention.

It is also clear that the linear relationship between distribution coefficient and phase concentration allows the contributions from the interactions of each phase to be summed directly, but their logarithms can not This was, of course, in conflict with some contemporary thinking and cast doubts on the thermodynamic procedure for describing the effect of stationary phase composition on solute retention. The effect of stationary phase composition on retention is often taken into account by inserting an extra term in the expression of the standard free energy of distribution. The results of Purnell et al., indicate that such adjustments to the standard free energy is not correct and retention is linearly not exponentially related to the stationary phase composition.


Purnell's results showed that stationary phases of intermediate polarities can be easily prepared from binary mixtures of a strongly dispersive and a strongly polar stationery phase. This procedure, although probably the easiest and most economic method of fabricating columns having intermediate polarities, is not used nearly to the extent that it could in commercial columns. The reason for this is not clear, but mixed phases are always worth considering as a flexible alternative to the use of a specific proprietary material.

Unfortunately, this simple relationship between phase composition and retention, was found not to be universal and did appear to break down where strong association occurred between the two phases. Initially, the reason for this was not understood and placed the interpretation of Purnell's work under some uncertainty. However, it was pointed out by Katz et al (7), that when strong association between the components of the phase took place, the phase would no longer constitute a simple binary mixture but would also contain the associate of the two phases as a third component. This third component would contribute to retention in a unique manner that differed from those of the unassociated components. It follows, that the simple linear relationship obtained for a binary mixture would not be expected to hold. The effect of association of solvents in binary mixtures will be considered in detail in due course.