Numerical analyses of bubble point tests used for membrane characterisation: model development and experimental validation

• Published in: Asia-Pacific journal of chemical engineering Vol. 6; no. 6; pp. 850 - 862
• Main Authors: Nassehi, V., Das, D. B., Shigidi, I. M. T. A., Wakeman, R. J.
• Format: Journal Article
• Language: English
• Published: Chichester, UK John Wiley & Sons, Ltd 01-11-2011
• Subjects: bubble point test; Bubbles; Calibration; Computer simulation; finite element method; Liquids; Mathematical models; membrane characterisation; Membranes; microhydrodynamics; penalty scheme; Porosity; Surface tension; volume of fluid method
• ISSN: 1932-2135; 1932-2143; 1932-2143
• DOI: 10.1002/apj.519

ABSTRACT
A model for simulating the microhydrodynamics of liquid/air flow inside membrane pores was developed, using a continuous penalty finite element scheme. The developed scheme combines the features of two‐phase flow systems with flexibility and accuracy. The volume of fluid (VOF) method was applied to track the motion of the gas–liquid interfacial boundaries as an approach to monitor the repulsion of the wetting liquid from the pores to detect their bubble pressures. To resolve the complexities arising from the inclusion of the surface tension at the liquid–gas interface as an unknown dynamic condition, it is treated as a resistance force in the equations of motion. The effects of the surface tension and other forces such as buoyancy are then determined by a model calibration factor with respect to a set of experimental data. To obtain the experimental data (e.g. pore size distribution, bubble pressures) for model validation, the bubble point test was conducted on different Nuclepore track‐etched membrane samples, which also provided insights into the mechanisms underlying the test and the interpretation of the pore size distribution. The experimental data are used to calibrate the numerical model. The calibrated model was, in turn, used to predict the outcome of bubble point tests for a range of inlet boundary conditions (e.g. pressures). The results obtained from our simulations are in good agreement with the experimental data, indicating the ability of the developed model to accurately predict the bubble point pressure. Copyright © 2010 Curtin University of Technology and John Wiley & Sons, Ltd.