CFD-PBM Invstigation Of The Liquid-Liquid Dispersions In Stirred Tanks

Liquid-liquid stirred tanks were used extensively in chemical engineering,pharmacy industry, they can be used to improve mixing, heat transfer and they are important applications in different industries. CFD can be used to investigate the liquid-liquid dispersions. As of the phase coupling is very complex and the breakage and coalescence is very complicated, CFD model can not predict accurate results. Hence, combining CFD and population balance model, in this work, the investigation of liquid-liquid dispersions in stirred tanks were launched. The results can be employed for large stirred tanks design and optimization.Firstly, the two fluid model, population balance model, quadrature method of moments and extended quadrature method of moments were discussed. The governing equations were also re-formulated. The boundedness and relizable of the variables were also tested. Then the population balance model was coupled with CFD and implemented into OpenFOAM to simulate liquid-liquid dispersions in stirred tanks.The surfactant has an effect on the droplets breakage and coalescence. In this work, QMOM coupled with CFD was used to solve the population balance model. The results show that the droplets coalescence was hindered.MB and Lehr breakage kernels can not predict accurate results. The results predicted by the CT, Alopaeus, and the improved MF breakage kernel is consistent with experimental data.For low dispersed phase fraction liquid-liquid dispersions, as of the droplets can hardly coalesce, eight different test cases from different literature were chosen to investigate the viscosity of dispersed phase and turbulent intermittency on the mean Sauter diameter. CT breakage kernel assumes that droplet breakage is highly related with the turbulent energy dissipation rate.MF breakage kernel assumes that droplet breakage is related with the turbulent energy dissipation rate, viscosity of dispersed phase and others. The results show that CT breakage kernel can not capture the turbulent intermittency’s effect on the mean Sauter diameter. MF breakage kernel predict better results.When the dispersed phase’s fraction increases, the droplet breakage and coalescence exist together. The dispersed phase’s fraction on the mean Sauter diameter can be explained by the turbulent dampening factor. In this work,different liquid-liquid dispersions with middle dispersed phase fraction were chosen to investigate the droplets breakage and coalescence processes. CT,MF kernels were used to solve population balance model. The results show that the log-normal EQMOM which can provide a continual number density function behaves better than QMOM, CT kernels predict fast breakage and coalescence process. However, MF kernels predict slower process which fitted with the experimental data. When the dispersed phase’s fraction is higher, the use of turbulent dampening factor makes the results better.