Optimal design of continuous integrated crystallization-filtration processes

Industrial crystallization is typically followed by filtration to separate the solid particles from the mother liquid and there is often a tradeoff between crystallizer and filter design: Investing more money in a larger crystallizer can increase crystal size and lower filtration costs. Thus, process simulation and optimization are essential to identify the optimal design of such integrated process. However such analyses are rare in the literature due to challenges in estimating cake resistance from a crystal size distribution (CSD). Recently it has been proposed to address this issue by combining the Kozeny-Carman equation with the discrete element method (DEM).

In this work, this approach is applied to optimize continuous integrated crystallization-filtration processes including those where the CSD is modified by product classification or fines removal. Results show that the optimal design depends on crystallization kinetics and slurry viscosity. For systems with low nucleation rate and low slurry viscosity, the optimal design is to minimize the crystallizer cost. For systems with high nucleation rate and high viscosity, a rigorous process model is required to determine the optimal crystallizer volume and fines dissolver design.