Assessment of Drop Coalescence and Breakup for Stirred Extraction Columns

Abstract

This work is a model assessment study considering two major hydrodynamic parameters: the holdup of the dispersed phase and the Sauter diameter. This is done for two different types of extraction columns, namely, the rotating disc contactor (RDC) and the Kühni column, using different drop breakup and coalescence models in a droplet population balance model. Based on the film drainage models for undeformable (spherical drops) and deformable drops with partially mobile interfaces, different simulations have been performed and the results compared with experimental values. The agreement between the experimental observation and the simulation is encouraging and the used models have proven to be suitable to predict holdup and Sauter diameter profiles for the system toluene/water. IntroductionExtraction columns are widely used in petroleum, pharmaceutical and chemical industries, mainly with physical extraction systems [1,2]. Their advantage is their high efficiency and low cost in respect of number of stages, solvent inventory, settler area, site area, maintenance, etc. [3, 4]. The choice of the most appropriate model for calculating a given type of a liquid-liquid extraction column is not straightforward. Most of the recent models [5,6] aim to study the influence of breakup and coalescence rates in turbulent dispersions on hydrodynamics and mass transfer. Drop coalescence and breakage are key parameters for a fundamental understanding of the behavior of liquid-liquid dispersions in extraction columns. These two processes are dynamic in nature and, in a turbulent flow field, depend on the drop sizes, the 552

A Genetic Algorithm Based Approach to Coalescence Parameters: Estimation in Liquid‐Liquid Extraction Columns

“…ðexpÞ À Q k;out 3 ðsimÞ h i 2 (15) According to this, the simulated outlet cumulative distribution, Q 3 , is correlated to the experimental one by varying the coalescence parameters. This procedure is continued as long as the least square differences are moving towards a minimum.…”

“…In the work of Chesters [11], the collisions are differentiated as viscous and inertial, when deriving the following formula, for the coalescence efficiency (see also [15]):…”

Parameter optimisation and validation for droplet population balances

“…c 1 and c 2 values were compared to Jildeh et al published results (see Table ). Jildeh et al simulations used the Schmidt breakage model with critical rotor speed correlation taken from Hasseine et al Therefore c 1 value was sensitive to rotational speed. Also, the breakage model used was not appropriate to account for the changes in rotor speed that affects the energy dissipation.…”

Drop sizes and population balance model for a Karr column

“…Once solving it, the model will predict the drop size distribution along the column. The PBM has been applied in various types of columns in a stagewise or a differential form, for example, the rotating disc column, 15,16 pulsed sieve plate column, 17–20 Kühni column, 21,22 and some newly designed columns 23,24 . In these studies, terms of drop behaviors usually come from two ways: from the previous literature or from the single drop study 25 .…”