ABSTRACT
A method for selecting mobile phases for either one-dimensional (1-D) or two-dimensional (2-D) planar chromatography is described and is applied to the separation of steroids by overpressured layer chromatography [Formula: see text] a form of forced-flow thin-layer chromatography [Formula: see text] using both normal- and reversed-phase chromatography. Two metrics are used for evaluating the separation quality of simulated chromatograms for each of 100 (or more) subsets of a set of 30 steroids in each of 15 1-D, and 105 2-D systems. The subsets vary in size between five and 25 steroids. Butyl acetate/toluene on silica gel and aqueous 2,2,2-trifluoroethanol are, on average, the highest and second highest ranked 1-D systems, respectively, for separating all subset sizes. These two systems are the constituent members of the system that is, on average, the highest ranked 2-D system for all subset sizes. The probability of the above systems being highest ranked decreases with decreasing subset size. There is only a 17% probability of butyl acetate/toluene on silica being the best system for separating a subset of five steroids, while there is a 3% probability of this being the worst system for this subset size. The spot capacity of each system can be estimated by considering 100 subsets of each size and noting the largest subset size that yields an acceptable value of one of the metrics used for measuring separation quality. The mobile phase selectivity may be quantified using the actual values of either of the two separation metrics, or by a nonparametric approach. The latter is used in such a way that a difference of unity in the ranking (for a given subset size) of two systems corresponds to a 95% probability that the higher ranked system will yield the better separation.
