Computational
investigation of particle clustering in low Reynolds number stirred tanks

Weicheng Xu ^1,2, Xin Feng ^1,2*,
Steven Wang ³, Chao Yang ^1,2*

1 CAS Key Laboratory of Green Process and Engineering, Institute of
Process Engineering, Chinese Academy of
Sciences, Beijing
100190, China

2 School of Chemical Engineering, University of Chinese
Academy of Sciences, Beijing 100049, China

3 School of Chemical Engineering and Advanced
Materials, Newcastle University, NE1 7RU, UK

E-mail: xfeng@ipe.ac.cn; chaoyang@ipe.ac.cn

Abstract:
Usually the stirred tank is used as the mixing equipment in chemical
engineering, and working under highly turbulent condition. However, a low
Reynolds number condition is needed, when high-viscosity fluids or
shear-sensitive materials are mixed. Wang et al. [1] found that in a low
Reynolds number stirred tank the particles will cluster in a toroid region,
which can be developed as a separation technology for separating particles with
different properties. In order to have a better understanding of this
separation method, a full resolution simulation of a solid-liquid stirred tank
is carried out to investigate the particle motion at low Reynolds numbers,
which is based on the lattice Boltzmann method (LBM), immersed boundary (IB)
method and hard sphere method. The influences of initial particle location, particle
density and diameter on particle trajectory have been simulated. The results
show that the initial locations of particle have little influence on the particle
cluster. All the particles have the tendency of moving to their equilibrium
orbit. And the particle density has significant influence on the equilibrium
orbit. A heavy particle will move far away from the vortex center of liquid
phase, in contrast the equilibrium orbit of light particles is closer to the
vortex center. For the diameter range considered in our work, the particle
diameter has little effect on particle¡¯s trajectory. Besides, the influence of Reynolds
number of the stirred tank has been investigated. The results indicate that the
equilibrium area for particles will disappear while the fluid flow is turbulent
violently.

Figure
1 .The comparison of simulation and experiment.

Figure
2. The influence of particle density on its motion (Re_t=100).
The scatters are the development of particle location and its color indicates
the revolution number of impeller. The density ratio of particle to liquid is
(a) 0.9 and (b) 0.95.

Figure
3. The evolution of particle height with different sizes.

Figure
4. The classical result of particle trajectory in the Re_t=150
stirred tank.

Acknowledgements: The financial supports from the National Key
Research and Development Program (2016YFB0301701), the National Natural Science
Foundation of China (21776282, 21808221), CAS Key Research Program of Frontier
Sciences (QYZDJ-SSW-JSC030) and the Instrument Developing Project of Chinese
Academy of Sciences (YZ201641) are gratefully acknowledged.

Reference:

[1]
Wang, S., Metcalfe, G., Stewart,
R. L., Wu, J., Ohmura, N., Feng, X., Yang, C. Solid¨Cliquid separation by
particle-flow-instability. Energy
& Environmental Science, 2014, 7(12), 3982-3988.

[2] Wang, S., Stewart, R. L., Metcalfe, G. Visualization of
the trapping of inertial particles in a laminar mixing tank. Chemical
Engineering Science, 2016, 143, 99-104.