Implementation of population balance models in Computational Fluid Dynamics codes and coupling with Molecular Dynamics

In this work a new approach for calculating the rate of formation of polymer nanoparticles in solvent-displacement precipitation processes is presented. The work is motivated by the fact that the standard expressions obtained from the classic nucleation theory seem not to be adequate to describe nucleation of particles primarily constituted by polymer molecules, due to the many approximations involved. A new expression for the nucleation rate is therefore derived and molecular dynamics simulations (carried out with Gromacs) are used to obtain all the missing parameters involved. This expression is then implemented together with the classical one into a population balance model that accounts for particle formation, molecular growth and aggregation. The population balance model is implemented in a computational fluid dynamics code (i.e. Fluent) and solved with the quadrature method of moments. Resulting in turbulent flow conditions, the model is treated with the Reynolds-averaged Navier-Stokes equation approach and coupled with a micromixing model (i.e. direct quadrature method of moments coupled with the interaction and exchange with the mean model. The model is finally validated against experimental data obtained for a polycaprolactone solvent-displacement precipitation process carried out in a confined impinging jets mixer and using water and acetone as anti-solvent and solvent. Predictions result in good agreement demonstrating therefore the overall validity of the approach.