Turbulent dispersion coefficients in cyclone flow: An empirical approach

Among the various theories available to predict cyclone collection efficiency, the finite diffusivity theory of Mothes and Loffler (1988) has been shown to give the best fit of the observed grade-efficiency curves. However, lack of knowledge on the dependence of the particles' turbulent dispersion coefficient with cyclone geometry, operating conditions and particle size has so far hindered the application of this theory for predictive purposes and for improved cyclone design. In this work, this theory is applied for predictive purposes, through the use of an empirical relation for the particles turbulent dispersion coefficient. The proposed relation is based on an analogy with turbulent dispersion in packed beds, and correlates the particle radial Peclet and Reynolds numbers. Laboratory-scale reverse-flow cyclones of previously unpublished geometries were built to test the applicability of the proposed relation. The Mothes and Loffler ( 1988) theory, when coupled with the proposed estimates of turbulent dispersion coefficients, is a powerful tool for predicting cyclone collection efficiency, short of using computational fluid dynamics tools.