Prediction of gas cavities size and structure and their effect on the power consumption in a gas-liquid stirred tank by means of a two-fluid RANS model

Aerated cavities behind the impeller blades in stirred tanks affect the power transferred to the liquid that in turns affects heat and mass transfer, thus the development of fully predictive simulation methods to detect the formation of cavities, their size and structures is of paramount importance for an effective simulation of aerated reactors and bioreactors. In this work, operating conditions corresponding to different cavity structures are investigated by means of a Reynolds averaged two-fluid model without adjustable parameters. Based on the comparison with previous experiments and correlations, the method proved to be reliable in the prediction of the transition between vortex-clinging and small '3-3′ cavities, cavity size and power drawn reduction. For the first time, small '3-3′ cavities with volume fractions close to unity are obtained with a steady approach. The power reduction mechanism is observed and a novel interpretation of the formation of the asymmetrical cavities is proposed.