CFD Model of a Semi-batch Reactor for the Precipitation of Nanoparticles in the Droplets of a Microemulsion

Precipitation inside the droplets of a microemulsion is a promising technology for the production of nanoparticles with tailored properties, like particle size or shape [1]. In this work, a water-in-oil (w/o)-microemulsion consisting of water, cyclohexane and the non-ionic technical surfactant Marlipal O13/40 is used to synthesise BaSO4 nanoparticles. The reaction is initiated by the mixing of two microemulsions, one containing the first reactant BaCl2 and the other containing the second reactant K2SO4, in semi-batch operation mode in a standard Rushton tank (V = 300 ml). A narrow particle size distribution (Fig. 1) and particle sizes between 4 and 40 nm have been achieved by experiments [2]. [GRAPHICS] A population balance model (PBM) assuming a discrete two-dimensional Poisson distribution for the dissolved Ba2+ and SO42- ions has been derived [3]. Rate constants for nucleation and growth kinetics proposed by Baldyga et al. [4] for bulk precipitation were fitted to take the slow-down effect of the droplet exchange into account. The obtained results showed a qualitatively good agreement with the experimental data (Fig. 1), but bigger deviations between the model and the experiments were observed for cases with a high concentration difference between the two microemulsions. The reason might be that the precipitation strongly depends on the supersaturation and that especially high concentration differences near the feed lead to a locally varying supersaturation profile, which is responsible for the deviations. To analyse this behaviour the zero-dimensional (homogeneous) PBM with over 600 ODEs is reduced into a set of 4 ODEs and implemented into the Computational Fluid Dynamics (CFD) code FLUENT 6.2 via user-defined scalars and functions. Hence, it became possible to investigate the reaction process for various three-dimensional inhomogeneous hydrodynamic conditions with reasonable computation times.