CFD Simulation of the Separation Zone of an Annular Centrifugal Contactor

Abstract

This paper presents computational fluid dynamics (CFD) simulations of the flow inside the rotor of an annular centrifugal contactor. The model geometry was based on the rotor of a commercially available contactor unit with a closed upper weir. Simulations were performed at various flow rates and it was found that the narrow flow area above the upper weir seals with water at high flow rates resulting in the formation of a siphon. A method for predicting the zero-point flow rate from CFD was also developed and simulations were performed which demonstrate a zero-point elevation due to this siphon.

Keywords:

• Computational fluid dynamics (CFD)
• solvent extraction
• centrifugal contactors
• liquid-liquid separation
• reprocessing

ACKNOWLEDGEMENTS

The authors would like to thank Mark Anderson for assistance with the experimental setup for the measurements. Thanks also to Ralph Leonard of Argonne National Laboratory for many helpful discussions and for contributing his vast contactor experience to provide encouragement and support for the simulation results. This research was performed under appointment to the U.S. Department of Energy Nuclear Engineering and Health Physics Fellowship Program sponsored by the U.S. Department of Energy's Office of Nuclear Energy, Science, and Technology. This work was partially supported by the National Center for Supercomputing Applications under TG-ECS070009 and utilized the Tungsten Cluster.

Notes

∗The purpose of this upper weir cap which effectively closes the upper weir is to hold the removable weir plate in place. For a permanent upper weir plate, the weir cap should be unnecessary.

†The weir cap is machined such that it fits comfortably around the rotor shaft although there seems to be some variability between units in the tightness of this cap around the rotor shaft. For the experimental measurements of the zero-point reported here a layer of seal tape on the rotor shaft was used to ensure air-tightness and consistency of measurements and provide the best comparison with the model geometry.

‡Because the free surface position does not vary significantly in the rotor for most cases, originally steady-state calculations were attempted. These were found to be unsolvable likely due to the interface discontinuities above the upper weir and therefore were abandoned. It may be possible to obtain a converged time-invariant solution for high flow rates such shown in Fig. 9 although this was not attempted.