Plant-based Investigations of Hydrodynamic Behaviours in Large Coal Flotation Cells

The performance of flotation cells can be impaired by the occurrence of biased feed distribution and unfavourable hydrodynamic behaviours, such as short-circuiting, pulp circulation flows, dead volumes or excessively high shear stress.

The hydrodynamics of a flotation cell is mainly determined by the cell design, but also influenced by operating conditions. Residence time distributions (RTDs) of phases are the easiest-to-obtain indications of the hydrodynamics of a flotation cell. The most important outcome from RTD experiment is the measured mean residence time. If the experimental mean residence time is different from the expected value, it indicates that there are unfavourable hydrodynamic behaviours such as dead volumes, short-circuiting, pulp recirculation or another anomaly. The cause of the difference can usually be determined by comparing the shapes of the RTD curves for the ideal case and the case in question.

Since the 1980s, fine coal cleaning in Australia has been dominated by Microcel and Jameson cell technologies. However, the hydrodynamic behaviours of these cells have not been experimentally characterised in the context of fine coal flotation. Therefore, industrial trends towards an increasingly larger capacity of new and existing installations and the need for improving combustibles recovery require a better understanding of hydrodynamic behaviours in industrial coal flotation cells.

The objective of this project was to characterise the hydrodynamic behaviours of Microcel columns and Jameson cells by carrying out plant-based experiments to determine the RTDs of liquid and solids phases and measure gas holdup distributions in the pulp phase.

This project has successfully developed effective approaches for characterising hydrodynamic behaviours in plant-scale flotation cells, based on residence time distribution (RTD) tests combined with electrical impedance spectroscopy (EIS) measurements. Plant-based tests have demonstrated that potassium chloride solutions and fine magnetite particles coated with sodium oleate can be conveniently used as liquid and floatable solids tracers in RTD tests in practical conditions. It has also been demonstrated that EIS is a sensitive tool for identifying abnormal distributions of froth characteristics (froth stability, solids loading and water content) and gas hold-up in the pulp phase.

Plant-based investigations of hydrodynamic behaviours in two Microcel columns were carried out at two Bowen Basin coal preparation plants. The following major unfavourable hydrodynamic behaviours were observed in the Microcel columns:

· Inappropriate positioning of four feed inlets caused the biased transfer of floatable solids towards the inner launder and the solids loading in the froth phase within the inner launder was almost twice as high as that in the froth exiting from the outer launder of the Microcel column, resulting in local overloading of the area surrounded by the inner launder.

· Sub-optimal injection locations of air spargers generated strong circulation flows along the axial direction, leading to large stagnant volumes of up to 70% of the total cell volume and short circuiting in the Microcel column with a height of 16 m and 25% dead volume in the column with a height of 9 m.

· Uneven distribution of the feed inlets caused froth solids loading to vary along the outer launder.

· The flow patterns in the pulp phase were either perfectly mixed flows or contained a major component of a perfectly mixed flow.

· Significant portions (63% for the 16 m column and 25% for the 9 m column) of feed reported to the tailing exit in less than 5 mins, resulting in low combustibles recoveries.

RTD tests and EIS measurements for solids loading of the froth phase were carried out on a circular and a rectangular Jameson cell at a Bowen Basin coal preparation plant. The following major hydrodynamic behaviours have been identified from these investigations:

· The multi-port feed distributor on the top of the circular Jameson cell produced a significant bias in the slurry feed distribution to the 24 downcomers, as evidenced by different RTD curves and solids loadings of froth phase.

· The pipe manifold feed distributor on the rectangular Jameson cell subdivided the slurry feed into branch streams with different solids content to each downcomer, leading to the solids content in the downstream branches being higher than those in upstream branches.

· The plunge of multi-phase mixture from the downcomer into the pulp phase generated high shear stress which could disrupt weak particle-bubble aggregates.

· The plunge of multi-phase mixture caused a fast channel flow between the downcomer exit and the tailings exit.

The improved understanding of hydrodynamic characteristics in Microcel columns and Jameson cells has allowed low-cost modifications of the internal structure of the flotation cells to be developed to optimise flow patterns and improved flotation plant performance in terms of capacity and product yield.