Bubble in superheated liquid collapses after slight expansion when surface tension is high #62
Description
I would like to report a possible issue in the TwoPhaseFlow solver related to the interaction between surface tension and phase change modeling.
In simulations of a vapor bubble in a superheated liquid with high surface tension (e.g., 72e-3 N/m, similar to water), we observed that the bubble initially expands slightly—as expected due to evaporation—but then unexpectedly collapses (shrink). This behavior is unphysical, since the liquid is superheated and no condensation should be present; the vapor phase should continue to grow, instead of vanishing.
The issue becomes more pronounced as the surface tension increases. In contrast, when surface tension is set to zero (e.g., as in the scrivenWedge tutorial), this collapse does not occur. Moreover, when surface tension is high but phase change is disabled, the bubble is transported correctly and behaves stably. This strongly suggests that the problem lies in the coupling between surface tension and phase change models.
Interestingly, at the moment of collapse, the pressure inside the bubble becomes lower than the outside, indicating a reversed pressure gradient. This may point to a numerical or physical inconsistency in the implementation.
To further illustrate the issue, I created a new simulation where a vapor bubble (air) is placed in a superheated liquid (water) inside a horizontal channel. The setup differs from the scrivenWedge tutorial, but the bubble still shrinks unexpectedly, despite the surrounding liquid being clearly superheated.
Could you confirm whether this is a known issue or offer insight into the possible cause? I would be happy to provide a simplified case file for reproduction. Thank you very much for your continued work on this excellent solver.
