No abstract

No abstract

No abstract

Bubble breakup plays an important role in gas–liquid flows, but detailed studies are still scarce. In this work, the breakup behavior of a single bubble in a stirred tank was experimentally studied with a high‐speed camera, focusing on the effect of gas density, liquid properties, agitation speed, and mother bubble size. The bubble breakup time, breakup probability, breakup rate, and daughter bubble size distribution were determined. The internal‐flow phenomenon that the gas flowed from the smaller part to the larger part of a deformed mother bubble was observed by a high‐speed camera. The results showed that with increasing gas density, agitation speed, mother bubble size and decreasing surface tension, the bubble breakup rate and probability of equal‐size distribution significantly increased. With increasing liquid viscosity, the bubble breakup rate decreased especially in the high viscosity range. An M‐shaped daughter bubble size distribution was observed, which was consistent with our previous bubble breakup model.

To account for the effect of liquid viscosity, the bubble breakup model considering turbulent eddy collision based on the inertial subrange turbulent spectrum was extended to the entire turbulent spectrum that included the energy-containing, inertial, and energy-dissipation subranges. The computational fluid dynamics-population balance model (CFD-PBM) coupled model was modified to include this extended bubble breakup model for simulations of a bubble column. The effect of turbulent energy spectrum on the bubble breakup and hydrodynamic behaviors was studied in a bubble column under different liquid viscosities. The results showed that when the liquid viscosity was < 80 mPas, the bubble breakup and hydrodynamics were almost independent on the turbulent spectrum. At liquid viscosity > 80 mPas, the bubble breakup rate and gas holdup were significantly under-predicted when the inertial turbulent spectrum was used, and when using the entire turbulent spectrum the predictions were more consistent with experimental data.