Optimal Design of Integrated SMB-Crystallization Hybrid Separation Process Using a Binary Solvent

Hybrid separation processes combining simulated moving bed chromatography and crystallization have been known to reduce the capital cost for enantiomer separation, but past studies have been limited to systems of single-component pure solvent. In this study, a model-based design strategy using a binary solvent is presented. Separation of d- and l-phenylalanine in a mixture of methanol and water was employed as a case study. Experimental data for solid–liquid equilibrium as well as chromatographic separation were obtained at various compositions of the solvent mixture to model the process. The experimental data were used to model and optimize the design of the hybrid separation process. The trade-offs between the solubility, temperature, chromatographic separation, and eutectic compositions were analyzed using the model-based optimization approach. In a case study, it was found that the productivity can be increased by over 18% when the temperature is increased from 10 to 50 °C because of the lower eutectic purity.