Investigation of operating parameters in high-performance displacement chromatography

The effect of operational parameters of displacement chromatography was examined in the separation of various mixtures such as that of the main hydrolysis products of methylfurylbutyrolactone, a potential anticancer drug, the diastereoisomers benzoyl-D- and benzoyl-L-phenylalanyl-L-alanyl-L-proline, as well as polyethylene glycol homologues containing 1-10 ethylene oxide units. The chromatograph was assembled from modules generally used in analytical high-performance liquid chromatography (HPLC) and the column effluent was analyzed by an on-line HPLC unit at 30-sec intervals. Octadecyl-silica was used throughout as the stationary phase. Derivatives of ethylene glycol and propylene glycol as well as tetrabutylammonium bromide and n-butanol were used as displacers. The throughput was used as the measure of efficiency. In the absence of axial dispersion, for a given separation various displacers are expected to yield the same efficiency if the slope of the operating line is kept the same by appropriate adjustment of displacer concentrations. In practice, however, the optimum slope of the operating line has to be determined experimentally as most available chromatographic systems depart from ideal behavior. The dependence of the throughput on the flow-rate and feed load also indicated the presence of non-equilibrium phenomena and the optimum value of these parameters was established experimentally. In most cases water was used as the carrier solvent but the separation of poorly soluble peptides required the use of hydro-organic carriers. Results obtained with octadecyl-silicas of different origin and a given displacer were found to vary significantly suggesting that even for stationary phases of the same type the selection of displacer requires special consideration. Most experiments were carried out with columns having dimensions customary in analytical HPLC. Increasing the inner diameter of the column did not result in the expected increase in throughout probably due to poor distribution of the sample at the column entrance. Therefore scaling-up the process requires careful engineering of inlet conditions. Throughput can be increased by connecting a small inner diameter column to the outlet of a large diameter preparative column. As theoretical predictions for ideal displacement chromatography do not hold in practice when axial dispersion is significant, optimization of the process requires experimental support. The results obtained in the separation of a variety of mixtures shed light on the most important operational aspects of displacement chromatography and suggest approaches to find optimum conditions.(ABSTRACT TRUNCATED AT 400 WORDS)