Application of the Lagrangian CFD Approach to Modelling of Crystallization in Stirred Batch Reactors Using the Smoothed Particle Hydrodynamics Method

Crystallization phenomena in stirred reactors are influenced by local hydrodynamic conditions and these must be taken into account for successful process scale-up and optimization. In this work, the available state of the art grid-based CFD methods and advantages and disadvantages of their application to mathematical modelling of batch crystallization processes were analyzed. The benefits of the Langrangian meshfree methods were discussed and the Smoothed Particle Hydrodynamics method proposed as an efficient method for a rapid prediction of the global mean flow in stirred reactors. Various aspects of the simulation results were assessed: quality of the fluid prediction, computational requirements, existence of numerical problems and availability of crystal size distribution. The developed Smoothed Particle Hydrodynamics CFD model was successfully coupled with discretised population balance equations to model a cooling batch crystallization process. It has been shown that 200 additional transport equations resulting from the discretisation of the population balance leads to only 50% decrease in computational performance while the same problem is still almost intractable from the computational point of view using the grid-based CFD methods.