Comparative study of mass-conserving interface capturing frameworks for two-phase flows with surface tension

• Comparison of mass-conserving interface capturing frameworks.
• Compressive VOF framework provides comparable accuracy as PLIC frameworks.
• Height-function technique estimates interface curvature most accurately.
• Compressive VOF framework and coupled VOF/level set framework able to capture complex interface characteristics.

A variety of frameworks to model two-phase flows with surface tension are available, each with its individual advantages and disadvantages. The understanding of the implications of the different frameworks is essential to conduct accurate and reliable two-phase flow simulations. In the presented study, three mass-conserving interface capturing frameworks are examined and compared. The frameworks can be distinguished by the method to capture and transport the interface, i.e. a compressive VOF method, a VOF-PLIC method and a coupled VOF/level-set method, as well as by the method to evaluate the interface curvature, namely a least-squares fit based on the VOF colour function, a height function technique and finite differencing. The interface frameworks are examined by means of three representative test cases, specifically chosen to assess the accuracy of the curvature evaluation, the prediction of capillary effects and the correct interaction between surface tension, viscous stresses and buoyancy. Most interestingly, the results demonstrate that advanced compressive VOF methods are able to transport evolving interfaces with an accuracy comparable to more complex and computationally expensive interface reconstruction methods, such as the applied VOF-PLIC method, and to predict surface-tension-dominated flows as accurate as coupled VOF/level-set methods. The results also show that, among the tested methods, the height function technique estimates the interface curvature most accurately, although the absolute differences in curvature error and parasitic currents between the methods are small.