CFD-coupled Modeling And Optimization Of Crystallization For Production Of Paracetamol

Crystallization is a complex process of multi-phase coexistence and multi-process coupling,involving mass and heat transfer among multiple phases and the transformation from continuous phase to discrete phase.Traditional crystallization process studies rely heavily on experimentation and experience,and process optimization is aimless and inefficient.Crystallization process simulation is based on the establishment of a quantitative crystallization kinetics models.Computer-aided computation can be used to model and predict a crystallization process,help understanding crystallization principles,and accomplish process optimization and control.In this thesis,the batch and continuous crystallization process of a model compound,paracetamol,was modeled and optimized through computer simulation and computational fluid dynamics.Firstly,the batch cooling crystallization process of paracetamol was modeled,and the moment method was used to solve population balance equation.The effects of seed particle size,seed amount,and cooling curve on the final product particle size were studied.It was found that the particle size of seed crystals had a small effect on the particle size of the final product,and the optimal amounts of the seeds were about 10% of mass of the resultant product.The cooling curve was optimized to obtain a cooling profile producing the largest average particle size.Then,a computational fluid dynamic simulation was performed on a continuous,stirred tank.Several turbulence models were compared,and Euler multi-phase model was applied to investigate the suspension status of solid particles of different sizes.The effects of stirring rates on the flow conditions and solid-liquid distribution were investigated.It was found that in a 5L stirred tank equipped with a Ruston impeller two flow circulations can be formed when the rotation speed beyond 100 rpm,and the mixing effect is greatly enhanced.When the particle diameter is larger than 100 ?m,even if the stirring rate is greater than 300 rpm,there will be a significant concentration difference between upper and lower circulations.Finally,a novel coupled CFD-PBE model was developed in which the population balance equation(PBE)was discretized using a high-resolution central scheme and solved simultaneously with CFD equations in ANSYS FLUENT 15.0 utilizing userdefined scalars(UDS).The spatial distribution of slurry density,nucleation rate and particle size distribution(PSD)was simulated.The effects of stirrer rates,temperature and residence time on continuous cooling crystallization process was investigated.It was found that the higher stirrer rate leads to narrower but smaller PSD,and as the average temperature gets higher,or the residence time gets longer,the influence of mixing on final PSD gets smaller.