OPTIMAL DESIGN OF SOLVENT GRADIENT SIMULATED MOVING BED CHROMATOGRAPHY FOR AMINO ACID SEPARATION

Abstract

A solvent gradient mode was applied to the simulated moving bed (SMB) chromatography for the separation of two amino acids, phenylalanine and tryptophan, for the first time. The optimal design of such a solvent gradient SMB (SG-SMB) system was performed under a four-zone open loop configuration, in which zone IV is disconnected from zone I. One of the prerequisites for such an optimal design is to obtain the adsorption isotherm and mass-transfer parameters (i.e., intrinsic parameters) of ethanol that functions as a mobile phase modifier. For this task, a multiple frontal experiment was carried out, which showed that ethanol followed a Langmuir-type adsorption behavior. Based on the resultant parameters of ethanol, the SG-SMB for amino acid separation was optimized to maximize the productivity under the constraints of pump capacities and product purities. The inlet and outlet flow rates, the switching time, and a local distribution of ethanol along the chromatographic bed were optimized using non-dominated sorting genetic algorithm with elitism and jumping genes (NSGA-II-JG) and rate-model simulations. The optimal SG-SMB for amino acid separation was found to have 110% higher productivity, 53% lower desorbent consumption, and 71% higher product concentration, compared to the corresponding isocratic SMB.