Reducing Parasitic Flows in Phase-Field Method with Height Function Surface Tension Model

[Yannick Peng]

Dear All,

I am currently working on two-phase flow simulations using the phase-field method with a continuous surface tension model in Basilisk. I am encountering a significant issue with parasitic flows, which are larger than desired for my research requirements.

To provide some context, I have also tested the Volume of Fluid (VOF) method along with its surface tension model, which employs a height function approach as implemented in the files tension.h, curvature.h, heights.h, and iforce.h. The VOF method has proven to produce minimal parasitic flows that are suitable for my needs.

Given this, I am curious about how to integrate the phase-field method, which utilizes a diffuse interface, with the height function approach for surface tension calculations. Specifically, I am wondering if the heights.h file can still be applicable in the context of a diffuse interface or if there are alternative methods to directly reduce parasitic flows in the phase-field method.

Thank you in advance for your time and assistance. I

Best regards,

Yannick Peng

[Stephane Zaleski]

Dear Yannick Peng

   There is some work by the group of Alfredo Soldati combining the CSF method with the phase field. 

    Best regards

    Hope it helps

Stéphane

[Victor Boniou]

Dear Yannick Peng,

There is an article which should be of interest for your question from the team of Olivier Desjardins :

 "A discontinuous Galerkin conservative level set scheme for interface capturing in multiphase flows"  https://doi.org/10.1016/j.jcp.2013.04.036 

In this article, the height function is applied to the accurate conservative level-set (which is another type of diffuse interface with a reinitialization step to control the interface thickness). They noticed that for the HF to work with a diffuse interface, they needed to expend the stencil (11x3x3) while 9x3x3 is enough to handle all situations in classic VOF and 7x3x3 is usually chosen as a balance between accuracy and stencil length. 

Best,

Victor