Modeling and Optimization of a Suspension Crystallization Separation Process for para-Xylene Purification

The morphology and DSC methods were used for measuring the crystallization kinetics of para-xylene. Solubility data from DSC experiments were fitted using the activity coefficient model and classical solid-liquid phase equilibrium theory, while the rates of nucleation, growth, aggregation, and breakage (functions of supersaturation) were built from morphological experiments. The particle size distribution was modeled via a population balance equation and validated by experimental data on a suspension crystallization separation process. The effects of the crystallizer operating temperature and residence time on product yield were also investigated; although the rates of nucleation, growth, breakage, and aggregation increase with decreasing temperature and increasing residence time, the breakage and aggregation are dominant at low temperatures and long residence times. Therefore, the crystallization process should be operated at an optimal temperature and residence time. The models built for this work can be used further in additional process design and optimization.

Automated summary

The crystallization kinetics of para-xylene was measured using morphology and DSC methods. Solubility data from DSC experiments were fitted using activity coefficient model and solid-liquid phase equilibrium theory, while the rates of nucleation, growth, aggregation, and breakage were determined from morphological experiments. A population balance equation was used to model particle size distribution and validated with experimental data. The effects of crystallizer operating temperature and residence time on product yield were investigated, suggesting an optimal temperature and residence time for the crystallization process. The models developed can be applied in process design and optimization.