liquid decreases in bubble column E-E simulation
 

Hi, I have a problem in bubble column simulation. the column is 0.1 m in diameter and 2m in height. Initially quiescent water is filled at the lower half. the gas is introduced at the bottom. The top surface is set as pressure outlet. k-eps model(mixture version) and QUICK scheme is used.
Question is: at heigher Ug(>0.1 m/s), the liquid will decrease dramatically, but not because entrainment since I have monitored the flux on outlet.
I guess some convegence problem causes this problem. Possibly the gas cap above the liquid bulk region, the backflow settings about the pressure outlet BC may be potential sticking points.

would u help me? Welcome constructive discussions.

ps: I know the degassing BC and it seems not practical for the simulation with gas cap. I have also extend the simulation domain but it doesn't help.

Did you set a minumum vf for the liquid phase? In that case your mass balance will not be satisfgied in regeions where vf_liq should go to zero