Application of Computational Fluid Dynamics Modelling to Coal Flotation

The work is an extension of a previous ACARP project C17038 in using the latest advances in computational fluid dynamics (CFD) to improve understanding of the hydrodynamics in coal flotation, and to improve the design and operation of the Microcel columns and Jameson Cells in existing installations.

A total of 98 multi-phase flow simulations (for liquid, air and solids) have been performed for both types of flotation cell. Six design modifications for each type of cell, including the standard designs have been investigated with different operating conditions of solid density, feed rate and particle size. CFD simulation results consisted of velocity vectors, gas hold-up, turbulent energy dissipation rate, and particle-bubble attachment and detachment rates. Averaged results have also been obtained for the whole cell and at the pulp-froth interface for analysis.

For the Microcel column, the CFD results showed that turbulent energy dissipation rates are greatest near the sparger exits and near the pulp-froth interface because of the feed inlet flows and the presence of feed pipes and disperser plates. Modifications investigated for the Microcel were focused on the design of the feed points. From CFD results, the attachment rates were found to be generally high at the centre of the Microcel column as well as near the feed and sparger exits, while high detachment rates occurred mainly near the sparger exits where the turbulent energy dissipation rates are high.

For optimising flotation performance, the average attachment rates and froth detachment rates in six modifications of the Microcel have been analysed to find the modification that produces the maximum attachment rate and the minimum detachment rate. From this analysis, the Microcel modification with a larger feed pipe and a larger disperser disk with standard gap has been found to give the best compromise between attachment and detachment rates. The Microcel modification with reverse feed pipes performs badly due to the high turbulence produced by the reverse flow.

Hydrodynamic results for the Jameson Cell indicated that the general pattern of flow from the downcomer exit to the pulp-froth interface is around the diffuser plate with a small portion of flow through the diffuser plate. The problem of high gas hold-up beneath the inner launder in the standard Jameson Cell which was identified in previous work has not been observed in any of the six modifications investigated in this project.

The average attachment rates and froth detachment rates for six modifications of the Jameson Cell have been analysed to find the modification that produces the maximum attachment rate and the minimum detachment rate. From this analysis, Modification 2 which has a short baffle with standard downcomer configuration, and Modification 5 which has the same short baffle with a lower diffuser plate perform well having good compromise between attachment and detachment rates.

This work has demonstrated that CFD modelling is a cost effective means of developing an understanding of particle-bubble attachment and detachment, and can be used to identify and test potential cell or process modifications.