Simulation of Liquid-Liquid Dispersed Flow in Horizontal Pipe Using Computational Fluid Dynamics

Liquid-liquid dispersed flows are commonly encountered in many of the industrial applications such as polymerization, emulsification, batch and continuous stirred reactors and pipeline flows such as in petroleum industries. Liquid-liquid two phase flows are very complex due to the existence of sever...

Full description

Saved in:
Bibliographic Details
Main Author: Walvekar, Rashmi G.
Format: Thesis
Language:English
English
Published: 2007
Subjects:
Online Access:http://psasir.upm.edu.my/id/eprint/5183/1/FK_2007_17.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Liquid-liquid dispersed flows are commonly encountered in many of the industrial applications such as polymerization, emulsification, batch and continuous stirred reactors and pipeline flows such as in petroleum industries. Liquid-liquid two phase flows are very complex due to the existence of several flow patterns and mechanisms. Numerical approaches offer the flexibility to construct computational models which can adapt large variety of physical conditions without constructing large scale prototypes. The present work focuses on predicting the phase hold-up across a pipe cross-section and ambivalence range for phase inversion phenomena at different mixture velocity and range of input water fraction. The Computational Fluid Dynamics (CFD) computations were carried using FLUENT 6.2.16 while the geometry was created in pre-processor, GAMBIT 2.2.3. Dispersed phase dynamics and the turbulent continuous phase are modeled using an Eulerian-Eulerian approach and standard ε−k turbulence model. To check the reliability of the CFD code, the predicted results were validated with experimental results of previous work at different mixture velocities and phase fractions. CFD predicted the flow phenomenon such as phase transition from water-in-oil dispersion to oil-in-water dispersion and flow development along the length of the pipe. CFD code also predicted the phase hold-up distributions at pipe cross section. The pressure gradient trends similar to those observed in previous experimental results were obtained using CFD code. The phase inversion point obtained was within the ambivalence range suggested in literature. The numerical CFD simulations also confirmed that interphase drag, lift and turbulent dispersion forces has significant influence on spatial phase distribution. CFD simulations so obtained were subsequently compared with experimental results from previous researchers and correlation featuring range of mixture velocities and phase inputs. The predicted hold-up profiles were in good agreement with the previous experimental results for high mixture velocities and were in reasonable agreement with those of lower mixture velocity. Overall good qualitative agreement was achieved between physical model and simulated results.