(444f) Using a Novel Approach to Model Drop Size Distribution: The Adaptive Multi-Size Group Method

Liquid-Liquid dispersions and emulsions occur in many applications in the chemical, food and cosmetics industry in equipment such as stirred vessels, rotor-stator devices and colloid mills. The behavior of such two phase systems is strongly governed by how one phase disperses in the continuous phase. Such behavior can understood only when a good description of the distribution of droplet size is available. The droplet size distribution depends on the coalescence of small droplets in to larger droplets and break-up of large droplets in daughter droplets. Coalescence and break-up is a complex phenomenon governed by the rheology of the fluids involved as well as the fluid dynamics effect such as turbulence and shear.Many researchers have investigate these behavior both experimentally and numerically [1].The mathematical models used to simulate such systems include simple correlation based models [1], moments based methods such as S-gamma model [2] and quadrature based models such as QMOM/DQMOM

The simpler models can increase the accuracy in dispersed flow but are not sufficient to capture all the details, especially if the size distributions do not follow log normal type of distributions. The quadrature based models generally give higher accuracy but at a disproportionately higher compute cost. This is due to them computing size bins which may be empty and/or cannot adapt to the changing distribution shape. In this paper we present the results from a recent model termed as AMuSiG or adaptive multi-size group model. This model is expected to give higher accuracy comparable to quadrature based models but at a much lower compute cost. A basic description of the model with validation examples will be provided.

[1] H. Steiner, R. Teppner, G. Brenn, N. Vankova, S. Tcholakova, N. Denkov, Numerical simulation and experimental study of emulsification in a narrow-gap homogenizer, Chemical Engineering Science 61, 2006, 5841 – 5855

[2] S. Lo, A .Splawski & B.J. Yun, “The importance of correct modeling of bubble size and condensation in predicting sub-cooled boiling flows”, JThe Journal of Computational Multiphase Flows pp299-308, 2012.

[3] A. Vichansky, A. Splawsky, Adaptive multiply size group method for CFD-population balance modelling of polydisperse flows, The Candadian Journal of Chemical Engineering, Volume 93, Issue 8 August 2015, Pages 1327–1334