such as squalane. The operating temperature would be chosen so that the kinetic energy of the dissolved solutes molecules was sufficiently high to provide adequate partial vapor pressure for each and thus permit elution in a reasonable time. Interactions in the mobile phase are extremely weak in GC, (5) and are not employed to influence selectivity. In LC, an appropriate dispersive stationary phase might be a bonded phase with a long aliphatic chain. To ensure that the selectivity resided predominantly in the stationary phase, a complementary polar and weakly dispersive mobile phase would be used. In LC, it is usual to allow one type of interaction to dominate in the stationary phase while a different type of interaction remains controlling in the mobile phase. Separations Based on Dispersive Interactions Separations based solely on dispersive interactions in GC must employ a nonpolar stationary phase such as a hydrocarbon or an alkyl silicone

between the alkane chains and the dispersive (hydrophobic) groups of the protein can be strong enough to cause structural deconformation; (i.e., the protein becomes denatured). Reducing the chain length of the bonded material, the dispersive forces can be reduced significantly and the deconformation diminished. In practice, carbon chains only two or four carbon atoms long are among those most commonly used for separating labile proteins. Stationary Phase Limitation by Chiral Selectivity. The extent to which an enantiomer can interact with the stationary phase depends on how close it can approach the molecules of the stationary phase. If the stationary phase is also chiral in nature, it is likely that one enantiomer in the sample will fit closely to the stationary phase surface whereas the other will be stearically excluded and thus have less stationary phase with which to interact. The first chiral separations in GC were reported by Gil-Av et al. as in

nbsp; It is clear that there is yet another limitation of which to be aware when exploring the effect of solvent composition on retention and selectivity. It is important to examine the effect of solvent composition over a range of temperatures, to ensure that the true effect of solvent composition on selectivity is disclosed. If the distribution system is evaluated at, or close to that temperature where the separation ratio remains constant and independent of solvent composition, then the potential advantages that can be gained from an optimized solvent mixture will never be disclosed. Any evaluation of either a particular stationary phase, or solvent mixture, for the separation of closely eluting solutes must be carried out over a range of temperatures.       Optimum

is distinctly more selective than the b material. It has been employed in the separation of a very wide range of compound classes, and from very small to very large molecules. Yet another stationary phase has been synthesized by substituting the cyclodextrin hydroxyl groups with pure the 'S' hydroxypropyl groups followed by permethylation. As a result, the size selectivity of the material is reduced but more polar (hydrophilic) groups are introduced. The b material has a greater chiral selectivity than the a or g phases. This material provides a good general purpose column. It is clear that there are many possibilities for derivatizing the cyclodextrins to provide unique interactive character; there are a large number commercially available and many more are likely to be synthesized in the future.   A number of examples of some specifically derivatized cyclodextrin stationary phases have been reported. In one example, the positions 2 and 6 are alkylated (pentylated) thus

or mixtures of both) so that polar selectivity remains dominant in the stationary phase. Unfortunately, all substances are not simply polar or dispersive or ionic but can exhibit various combinations of all three interactive characteristics. It follows, that the stationary phase will also need to have the appropriate mixture of interactive properties to maximize selectivity and in contrast the interactive character of the mobile phase will need to oppositely balanced to ensure maximum selectivity still resides in the stationary phase. Under some circumstances the selectivity of the mobile phase can also be exploited to interact more strongly with the neighboring substance and thus elute it more quickly. It is clear, that some considerable effort must be made to identify the bests phase system for the isolation of a substance from a hitherto unknown mixture. This work can be carried out on an analytical column and its success will determine the maximum load that can be used,

in competing for the benzene against the dispersive interactions of the n-hexadecane. Chiral Stationary Phases There are basically five general types of chiral stationary phase in common use in LC. The first is the protein based stationary phase. These stationary phases usually take the form of natural proteins bonded to a silica matrix. As they are proteins, they contain a large number of chiral centers and are known to interact strongly with small analytes exhibiting strong chiral selectivity. There are specific interactive sites that provide chiral selectivity, but there are many more sites that only contribute to general retention. These other sites can be deactivated by mobile phase additives (e.g. octylamine) which reduces the overall retention and increases the chiral selectivity. The second type consists of relatively small molecular weight chiral substances bonded to silica 9 Pirkle (37). Each bonded group has a limited number of chiral centers available but, due