Modelling of precipitation process by hybrid large eddy simulation - multizonal approach Application to uranium oxalate

A modelling approach dedicated to precipitation processes has been developed to control the quality of the solid particles formed and to support design of industrial reactors. Precipitation reactions are well known to be very fast and highly sensitive to mixing effects. That is why modelling relies on the coupling of chemistry and hydrodynamics. The precipitation chemistry involves thermodynamic data and kinetic laws while the hydrodynamics inside the precipitator is described using a Computational Fluid Dynamics (CFD) approach. Currently a key issue for modelling a precipitation process arises from turbulence and chemistry coupling. Different methodologies are proposed in the literature ranging from hybrid methods based on multizonal models to a fully coupling, in which the population balance equations are directly implemented into a CFD code. In the hybrid modelling, the multizonal model that comprises several perfectly mixed compartments can be parameterised using CFD. Under these conditions, this multiscale method offers a practical compromise between simplified models based on idealised mixing description and the computationally expensive CFD solution. Consequently we have developed a multizonal approach in which hydrodynamic information is entirely extracted from Large Eddy Simulations. This modelling is applied to simulate the precipitation of actinide oxalates.