A numerical method for predicting the performance of a randomly packed distillation column

Abstract

A numerical method is proposed for modeling the distillation process in a randomly packed column. The proposed model is able to predict the axial and radial concentration distributions along the column without introducing the empirical turbulent Schmidt number or the experimentally measured turbulent mass transfer diffusivity. The present model involves the differential mass transfer equation set and the accompanied computational fluid dynamics (CFD) formulation with the conventional k − ε model. For the closure of mass transfer equation, the recently developed two-equation model is adopted, which consists of the equations for expressing the fluctuating concentration variance c 2 ¯ and its dissipation rate ε c [Z.M. Sun, B.T. Liu, X.G. Yuan, C.J. Liu, K.T. Yu, New turbulent model for computational mass transfer and its application to a commercial-scale distillation column, Ind. Eng. Chem. Res. 44 (12) (2005) 4427–4434]. The validity of the proposed model was testified by applying to a commercial scale randomly packed column of 1.22 m internal diameter and packed with 50.8 mm metal Pall rings in 3.66 m bed height for separating cyclohexane/ n-heptane mixture under total reflux and 165.5 kPa [A. Shariat, J.G. Kunesh, Packing efficiency testing on a commercial scale with good (and not so good) reflux distribution, Ind. Eng. Chem. Res. 34 (4) (1995) 1273–1279]. Satisfactory agreements were found between the model prediction and the published experimental measurement on the axial concentration distributions, the HETP and the turbulent mass transfer diffusivity along the radial and axial directions.