(690a) Eulerian-Lagrangian Modelling of Particle-Liquid Flow and Mixing in a Stirred Vessel

Mechanically agitated vessels are widely used for various mixing operations within a wide range of industries including the chemical, pharmaceutical, food and petroleum industries. They are used for liquid blending, solid-liquid mixing, gas dispersion in liquids, heat/mass transfer enhancement and chemical reaction. Mixing is intrinsically a Lagrangian process and, whilst Eulerian data are essential, Lagrangian information is necessary for its complete description. Possible approaches of generating Lagrangian data can, in principle, employ numerical simulations or experimental techniques based on Lagrangian particle tracking to provide the trajectories of fluid elements or solid particles. We use computational fluid dynamics (CFD) simulations based on an Eulerian-Lagrangian approach in which particles are tracked through the continuous phase in transient mode, to predict the turbulent 3D flow of a particle-liquid suspension in a mechanically agitated vessel. Various computational codes are developed for performing Lagrangian statistical data analysis, Lagrangian-to-Eulerian data conversion, local pointwise multiphase mixing evaluation and detailed Eulerian plane by plane investigations inside the 3D flow. The CFD simulations are validated by comparing with experimental Lagrangian data obtained from a unique positron emission particle tracking (PEPT) technique which is able to accurately track different components of a multiphase flow in 3D space and time, and determine their individual long-term trajectories. The Eulerian velocity distributions inferred from both methods are in excellent agreement, confirming that the Eulerian-Lagrangian simulations are able to accurately predict the key features of the flow, as shown in Figures 1-6. These simulations are extended to study the flow of more complex multiphase particle-liquid suspensions relevant to industrial applications. The effects of various flow parameters are investigated including particle size, density and concentration, and carrier fluid properties. The aim of this work is to evaluate the capability of CFD to predict such complex flows and thus facilitate their modelling for research and industrial design purposes.