Comparison of solvent models for characterizing stationary phase selectivity in gas chromatography

✍ Scribed by Salwa K. Poole; Brian R. Kersten; Colin F. Poole

• Publisher: Elsevier Science
• Year: 1989
• Tongue: English
• Weight: 772 KB
• Volume: 471
• Category: Article
• ISSN: 1873-3778
• DOI: 10.1016/s0021-9673(00)94158-x

✦ Synopsis

Two experimental approaches to the measurement of stationary phase selectivity using the thermodynamic models proposed by Rohrschneider-McReynolds and Golovnya-Poole are tested theoretically and experimentally to establish their reliability. The retention index difference of Rohrschneider-McReynolds is demonstrated to incorrectly determine selectivity since this difference is largely determined by the difference in solubility of the n-alkane retention index markers in the polar and non-polar phases. Also, the assumption that the contribution of dispersion to the index value is equivalent to the retention of a hypothetical n-alkane on squalane with the same retention time as the test solute on the polar phase fails to take account of the differences in the free energy of solution per index unit on the compared phases. These inconsistencies are not found when differences in the partial molar Gibbs free energy of solution for a series of test solutes are used to determine stationary phase selectivity. A general equation relating the free energy differences to retention index differences is derived and indicates that there is no simple relationship between the two models which, therefore, predict very different selectivity changes for the same test solutes for a group of compared phases. It is concluded that the ordering of stationary phases with respect to their ability to interact selectively with a particular test solute should be determined from free energies of solution (determined from gas-liquid partition coefficients corrected for interfacial adsorption) and that the use of the McReynolds phase constants be abandoned for this purpose.