(719z) Automated Iterative Refinement of Adsorption Isotherm Models for Simulated Moving-Bed Operation Based on Cyclic Steady State Data

Enantiomeric separations by simulated moving-bed (SMB) chromatography are usually carried out under strongly nonlinear conditions. The accurate determination of the competitive adsorption equilibrium of the enantiomeric species is thus of fundamental importance to allow computer-assisted optimization or process scale-up. This is especially true for the SMB, because its operating conditions cannot be designed conveniently without knowledge of the adsorption isotherms of the feed components.

The inverse method of isotherm determination is currently becoming very popular as a quick procedure for estimating the adsorption equilibrium data necessary for designing SMB separations. It derives the isotherm from overloaded band profiles of individual solutes or of their mixture. In recent studies, reporting comparisons of frontal analysis and inverse methods, it is concluded that the inverse method gives accurate estimates of the competitive isotherm parameters up to the maximum elution concentration of the overloaded bands. However, it is only moderately accurate from the maximum elution concentration up to the injected concentration.

In this work we describe and validate experimentally a procedure in which the inverse method is applied directly to the cyclic steady-state (CSS) concentration profiles of the running SMB process to update the parameters of the prescribed adsorption isotherm model. The operating conditions are then optimized for the newly determined isotherm parameters and applied to the running SMB process. This process is iterated and automated in our monitoring and control software.

As a proof of concept of the proposed methodology, it has been successfully applied to the chiral separation of Reboxetine enantiomers on Chiralpak AD, using a mixture of Hexane-Ethanol-DEA as solvent. The system was operated near the solubility limit of the the racemic mixture. Reboxetine is an antidepressive NRI drug. Only the (R,R)- and (S,S)-pair is present as a racemic mixture in the active principle and commercial formulations. Recent studies support the hypothesis that the (S,S)-enantiomer is a more potent inhibitor than the (R,R)- and that it is responsible for the vasomotor and cardiac side effects of Reboxetine.