451940 Limitations of Simple Mass Transfer Models in Polydisperse Liquid- Liquid Dispersions

Tuesday, November 15, 2016: 3:15 PM
Union Square 1 & 2 (Hilton San Francisco Union Square)
Jeremias De Bona1, Dongyue Li2, Antonio Buffo3, Wioletta Podgorska4, Marco Vanni5 and Daniele Marchisio1, (1)Institute of Chemical Engineering, Department of Applied Science and Technology, Politecnico di Torino, Torino, Italy, (2)State Key Laboratory of Chemical Resource Engineering, School of Chemical Engineering, Beijing University of Chemical Technology, Beijing, China, (3)Department of Biotechnology and Chemical Technology, School of Chemical Technology, Aalto University, Espoo, Finland, (4)Faculty of Chemical and Process Engineering, Warsaw University of Technology, Warsaw, Poland, (5)Institute of Chemical Engineering, Department of Applied Science and Technology, Politecnico di Torino, Torino, Italy, Politecnico di Torino, Torino, Italy

In this work turbulent liquid-liquid dispersions in stirred tanks are simulated by using computational fluid dynamics (CFD), population balance models (PBM) and different simplified models. The evolution of the droplet distribution is evaluated by means of the PBM in all simulations; different strategies are instead adopted for the description of the flow field: zero-dimensional (0D) models, in which perfect mixing of the phases is assumed, and three-dimensional (3D) models, where the spatial inhomogeneities are considered through the numerical solution of mass and momentum balances. 3D simulations are performed with our implementation of the quadrature method of moments (QMOM) into the Eulerian-Eulerian two-fluid solver of the open-source CFD code OpenFOAM, while 0D models use different Quadrature-based moment methods (QBMM) for the solution of the PBM. Different coalescence and breakage kernels are considered, based on the classical homogeneous and multifractal turbulence theories. Model predictions are eventually validated against experimental data for different test cases, consisting of different stirred tank geometries and different continuous and disperse phases. The comparison of the different (0D and 3D) models clearly shows that the 0D models can be used under dilute conditions, whereas in the dense regime the 3D models should be used instead. The comparison allows also to derive a correct formulation of the 0D models, that are instead often wrongly formulated.

Extended Abstract: File Not Uploaded
See more of this Session: The Use of CFD in Simulation of Multiphase Mixing Processes II
See more of this Group/Topical: North American Mixing Forum