Improving Coal Flotation Recovery Using Computational Fluid Dynamics

This work involves using the latest advances in computational fluid dynamics (CFD) to increase understanding of the hydrodynamics in coal flotation and to identify any opportunities to improve design and operation of both the Microcel column and Jameson cell. The CSIRO CFD model incorporates micro-processes from cell hydrodynamics that affect particle-bubble attachments and detachments. CFD simulation results include the liquid velocities, turbulent dissipation rates, gas hold-up, particle-bubble attachment rates and detachment rates.  

The issues facing coal flotation are two-fold: firstly, a drive to increase capacity of existing installations, and secondly, a need to increase the recovery of coarse coal. Both of these issues are influenced by the flow conditions inside the cell. The conditions that affect coal recovery have been identified through an analysis of the CFD results and residence time distributions. The effect of composite particles in coal flotation is considered in this work through the spread of specific gravities and particle sizes. An additional part of the work program involves modelling the froth zone to allow investigation of the behaviour of coarse coal particles in the froth zone and the effects of wash water on particle detachment from bubbles in froth.

The attachment and detachment rates per particle have been used as a comparative measure of cell performance. CFD results for both the Microcel column and Jameson cell indicate that the attachment and detachment rates per particle both increase with increasing solid density and particle size. The increase in the attachment rates from increasing the solid density is because there is less number of particles at higher densities, but the effect is small in comparison with the effect from particle size, with smaller particles having lower attachment rates.  

Under the conditions investigated, the Microcel column is well designed with fairly good performance throughout the column, but further improvements are still possible. In a detailed analysis of the Microcel column, the attachment rates were found to be higher at the pulp-froth interface than the effect in the bulk, but the detachment rates are lower at the interface than in the bulk. Although the detachment rates are lower at the interface, the question is whether they can be decreased further by minimising the regions with high turbulent dissipation rates, such as by using larger feed pipes or positioning the exits of feed pipes further away from the interface. The attachment and detachment rates per particle both decrease with increasing solid feed rate because there are more particles for the same number of bubbles. The performance in the attachment rate decreases by about 30% when the feed rate is doubled.

In the Jameson cell, the turbulent dissipation rates were found to be greatest near the inner launder where the flow is influenced by the air coming from beneath the sloping launder wall. Because of this effect, the turbulent dissipation rates at the pulp-froth interface are also higher near the inner launder. As the local turbulent dissipation rate directly influences the local detachment rate, it is therefore important to minimise these regions with high turbulent dissipation rates. A recommendation is to investigate a deeper inner launder which extends downward towards the region with low void fractions to trap less air beneath it.

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