Otimização e estudo de casos na modelagem e simulação de processos de separação em leito móvel simulado com gradiente de solvente (LMS-GS)

Abstract

Chromatographic separation processes have been widely used in industry as they were proved able to separate similar components into mixtures usually difficult to separate using simpler physical methods. It can operate in continuous or batch modes. Among available continuous chromatographic processes, the simulated moving bed (SMB) emerges as one technique that efficiently works in complex mixtures mainly due to its countercurrent operation without requiring movement of the solid phase. The SMB process is extensively studied and applied industrially such as in petrochemicals, pharmaceuticals and fine chemicals. The use of solvent within the dessorvent stream to act as an affinity modifier along the columns is one derivative of the traditional mode of operation, known as solvent gradient SMB (SG-SMB). The present work aims to study this latter mode of operation, through computational models built using less rigorous equations that were proven to represent the system when compared to experimental data for amino acid mixtures (tryptophan, phenylalanine and tyrosine). The available experimental data sets are input to an inverse problem for identifying the overall kinetics mass transfer coefficients, used to characterize the SG-SMB unit. It was utilized the Luus-Jaakola routine and following with the results different operating conditions were tested using the X-Equation optimization approach. The use of a design equation called X-Equation has been validated based on simulation results for a SG-SMB unit, where all optimized cases were in agreement with expected results with regard to purity of the streams, productivity and solvent consumption of the process.