Item

Abstract

To be productive the columns are usually overloaded in preparative chromatography. It is well known that concentration overloading is superior to volume overloading. A typical restriction to optimise a certain separation process is typically given by the limited solubility of the sample in the applied mobile phase. A possibility to increase the column load is to use for the injection a stronger solvent. The injection of very concentrated samples now brings the risk of triggering undesired crystallisation processes. Crystallisation might occur in particular in the initial section of the column and cause blocking phenomena. The objective of the presentation is to study experimentally and theoretically the application of a stronger solvent for the injection than for the mobile phase. Of main interest is to quantify the limits of this approach. For two different single component model systems elution and breakthrough experiments were carried out. a) DL-threonine on a NH₂ phase using water-ethanol mixtures as the mobile phase and pure water as the injection solvent, b) Methyl deoxycholate on normal phase silica using ethyl acetate-hexane mixtures as the mobile phase and dichloromethane as the injection solvent. Besides the measurement of the adsorption isotherms, systematic solubility studies were carried out for various solvent compositions and the kinetics of crystal growth were evaluated. In the theoretical part of the work a two phase column model was developed. Essential property of the model is the fact that it can take into account crystallisation in the column provided a certain supersaturation is exceeded. The model was tested using the parameters determined for the model systems. A difficult aspect in applying the approach suggested is the description of nucleation phenomena.