Percolation Segregation Model for Similar and Differing Constituents

Abstract

Segregation is a ubiquitous and undesirable phenomenon that occurs nearly anywhere and anytime as particulate materials are stored, handled, processed, or conveyed. New percolation segregation mechanistic hypotheses were formulated to interpret the percolation and sieving mechanisms. The theoretical hypotheses consist of two attributes: hypothesis I, larger particle sizes have higher potential of segregation rate; hypothesis II, falling path become less tortuous with larger size ratio. A mechanistic theory–based segregation model (denoted MTB model) for binary G-g (coarse material glass beads, G, and fine material glass beads, g) and F-g (coarse material poultry feed, F, and fine material glass beads, g) combinations was developed using the principles of mechanics, dimensional analysis, and linear regression methods. The MTB model successfully correlated explicitly the effect of particle size and shape with the particle density effect implicit in segregation potential of binary mixtures in one quantitative equation. The verification results showed that the MTB model accurately (root mean square error, RMSE = 1.22) predicted the segregation potential for G-g and F-g combinations with size ratios of 4:1, 6:1, and 8:1 and absolute sizes of 710, 1,000, and 1,400 μm. The validation results showed that the MTB model produced RMSE = 1.69 for smaller size ratios such as 3:1 and 2:1.