Loading…

On suitability of phase-field and algebraic volume-of-fluid OpenFOAM® solvers for gas–liquid microfluidic applications

Microfluidic devices often contain several phases. Their design can be supported by interface-resolving numerical simulations, requiring accurate methods and validated computer codes. Especially challenging are submillimetre air bubbles in water due to their large density contrast and dominance of s...

Full description

Saved in:
Bibliographic Details
Published in:Computer physics communications 2019-03, Vol.236, p.72-85
Main Authors: Jamshidi, F., Heimel, H., Hasert, M., Cai, X., Deutschmann, O., Marschall, H., Wörner, M.
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Microfluidic devices often contain several phases. Their design can be supported by interface-resolving numerical simulations, requiring accurate methods and validated computer codes. Especially challenging are submillimetre air bubbles in water due to their large density contrast and dominance of surface tension. Here, we evaluate two numerical methods implemented in OpenFOAM®, namely the standard solver interFoam with an algebraic volume-of-fluid method relying on a sharp interface representation and phaseFieldFoam relying on the phase-field method with diffuse interface representation. For a circular bubble in static equilibrium, we explore the impacts of uniform grid resolution and bubble size on bubble shape, mass conservation, pressure jump and spurious currents. While the standard interFoam solver exhibits excellent mass conservation with errors below 0.1% on fine grids, it lacks the accuracy to predict reasonable physics for a bubble in microfluidic systems. At higher resolution, large spurious currents significantly displace and deform the bubble, which is oscillating with resolution dependent mode and frequency. Furthermore, the pressure jump is consistently underestimated by more than 10%. The solver phaseFieldFoam suffers from much larger mass losses of up to 2%, which decrease as the ratio between interface thickness and bubble diameter decreases provided the diffuse interface region is adequately resolved. Spurious currents are very low and the bubble remains circular preserving its initial position with an error in pressure jump below 1%.
ISSN:0010-4655
1879-2944
DOI:10.1016/j.cpc.2018.10.015