Estudo da mistura e segregação de material granular em tambor rotatório com suspensores empregando a abordagem numérica Lagrangeana

Abstract

Rotary drums are frequently used when processing granular materials. The study of particle flow inside a rotating drum is imperative to improve the effectiveness of this equipment. One of the processes it is used is in the mixing of granular materials. However, particles with distinct properties such as shape, density and size, have a tendency to segregate when rotating inside a drum. Smaller and heavier particles percolate through larger and lighter particles in the active layer and concentrate at the center of the bed. Multiple variables affect the mixing efficiency such as filling degree, drum geometry, particle properties and the use of baffles, which disturb the particle flow. In the present study, the Discrete Element Method (DEM) was utilized to investigate the influence of adding baffles to the drum geometry on the mixing rate and the degree of segregation for a binary mixture. The simulations were performed with LIGGGHTS®-PUBLIC open source software. The static and rolling friction coefficients - DEM model parameters - were analyzed through a series of simulations based on a Central Composite Design. The simulation results for the repose angle and the mixing rate constant were compared to experimental findings and a high degree of agreement was observed for six conditions. The one with DEM parameters closest in value to those reported by Lima (2017) was used for subsequent simulations. A 3³ experimental design was performed to evaluate the effects of filling degree, size and number of lifters on the mixing rate constant and degree of segregation (Lacey index). Segregation was analyzed for both size-driven segregation and density-driven segregation. Regarding segregation control, the use of baffles was responsible to increase Lacey index from 0,51 to 0,94 at 50% filling degree using 6 baffles and baffle length to drum radius ratio (L/R) of 0,75. Baffles also showed significant effect in improving the mixing velocity, raising the mixing rate constant at 25% filling degree from 0,194 s-1 without baffles up to 0,377 s-1 using 6 baffles and L/R of 0,50. The particle size was also investigated but had no effect on mixing velocity.