according to open literature, there should be a phase split phenomena when the two-phase flow is introduced into a T-junction. In most cases, the air phase is more likely to enter the side arm due to the existence of a thin water film on the T-junction wall, and it is just the mass ratio of air and water entering the side arm are of interest to me.

1.My problem considerd is wate/air two phase flow in T-junction

2.the flow pattern is annular.

i used multiphase(mixture and eulerian) first, but there is none liquid film, and it seems that the phase donot separate at all. (velocity inlet, outflow, outflow)

then i turn to DPM, and i turn on the "water-film" option in the wall BC, after calculation, there is still no water film, and the splitting ratio does not seem to be good

where is the problem? Is there anyone have ever performed such a research like mine? can you help me?

Many mang thanks

In your problem is there a distinct interface between the water and air at the inlet? ie. the phases are not mixed at the inlet.

If this is the case you should use the VOF model.

to：CDE first, i would like to express my appreciate for your answer.

In my problem, there is not a distinct phase interface, cause that the air and water are premixed, and it is injected in to the domain in a atomize way. I assume that the averaged diameter of the liquid is about 5um, so vof model could not be used

thanks