Modelling and experimental investigation of micromixing of single-feed semi-batch precipitation in a liquid–liquid stirred reactor

Abstract

A finite-mode probability density function (FM-PDF) model is adopted to simulate the semi-batch precipitation process of barium sulfate from aqueous solutions of barium chloride and sodium sulfate in the presence of kerosene drops in a turbulent liquid–liquid stirred reactor driven by a Rushton turbine. The turbulent kerosene–water flow field is calculated with an anisotropic two-phase explicit algebraic stress model (EASM) and the standard k–ε model based on an Eulerian–Eulerian approach using in-house codes. The flow predictions are validated against macromixing homogenization curves measured from tracer mixing experiments. The particle size distribution has been calculated by solving five moments of the particle size distribution coupled with both nucleation and growth kinetics. The effects of impeller speed and inert dispersed oil volume fraction have been investigated. Agreement between micromixing model predictions and experimental data is satisfactory, which suggests that the time-dependent finite-mode probability density function model is a feasible approach for numerical prediction of the fast precipitation process in turbulent two-phase stirred reactors.