Nc. Dias et Cf. Poole, Compliance of retention data on inorganic oxide adsorbents with the solvation parameter model, J PL CHROM, 14(3), 2001, pp. 160-174
The solvation parameter model has been used to characterize the retention p
roperties of a varied group of solutes in silica gel thin-layer chromatogra
phy (TLC) and in silica gel and alumina column liquid chromatography. The m
odel was unable to describe retention on silica get TLC plates with the R-M
value as the dependent variable for five single-solvent mobile phases. The
results were improved by fitting the retention data to the simple competit
ion model and separating the solute and solvent contributions (denoted S an
d e(o)A(s), respectively) to the free energy of adsorption on the inorganic
oxide. Separate models were then constructed to enable estimation of value
s of S and A(s) from structure. These models were successful in describing
retention in column liquid chromatography on silica gel with mixtures of me
thyl t-butyl ether and hexane as a mobile phase. This approach proved less
reliable for calculating retention in TLC, probably because of non-equilibr
ium in the separation system. Evidence is presented that neither the solute
adsorption parameter (S) nor the solute cross-section (A(s)) as used in th
ese studies is unambiguously defined. Further refinements aimed at establis
hing clearly defined solute and solvent adsorption terms, and possibly incl
uding selective solute-solvent interactions in the mobile phase as a second
ary contribution to retention, could result in improved model performance.
The approach described here should be considered preliminary and thought of
as a stepping stone in the direction towards a comprehensive model for str
ucture-driven method development in normal-phase separations, which are cur
rently less developed than models available for reversed-phase separations.