Scale-Up of Coal Flotation Using Picobubbles

The use of cavitation to enhance the flotation of minerals and coal has been studied over a number of years. Nicol et al. (1986) reported that the use of a superimposed acoustic field nucleated precipitation of extremely small air bubbles (picobubbles) on the surface of the low energy hydrophobic coal particles. Cavitation, in this instance, was produced by the rapid changes in pressure caused by the passage of acoustic waves through the flotation pulp. This resulted in an increase in the effective hydrophobicity of the particle, as does the coating of a hydrocarbon collector on the coal surface.

The work by Attalla et al. (2000) showed that, at least for in situ picobubble formation, cavitation produced in an acoustic field would not only improve product recovery, but also reduce collector use requirements.

Based upon the results of the previous ACARP project (ACARP Project No 9048), it was decided to investigate the concept at full scale with a unit that was retrofittable to a full scale coal flotation plant. The work program incorporated a plant assessment phase, screening possible coal preparation plants (CPP) in order to select the most prospective operation. The next phase was to design and fabricate a system that would allow the selected plant to operate in two modes

• Normal operation
• Cavitated feed operation

The design of the full scale cavitation unit was a scaled up version of the venturi unit used in the pilot scale project.

The venturi was chosen as the best option, because the gradual reduction and the subsequent gradual expansion reduced the possibility of blockage. The venturi also has very low head loss characteristics which minimises the impact on the pump capacity in a plant. The permanent head loss of the other options could be as high as 3-4 times that of the venturi. The ability to vary the throat length in the venturi was also an advantage. This means that the critical pressure can be maintained to allow the expansion of gas nuclei in the fluid before a region of high pressure is reached, where the bubbles collapse.

Initial testing was carried out where 200 litres of Macquarie CPP flotation feed was processed through the pilot scale unit. Two tests were operated, one with cavitation the other without. In each test, one sample of feed, product and tails was taken. The sample used for the test contained the flotation feed dosed with the plant's flotation reagent, a combined reagent, Fuchs Centifroth. It was taken back to the CSIRO Newcastle test facility and processed immediately.

Table A. Results of Runs 1 and 2 Pilot Scale Tests

Subsequently, a larger quantity of flotation feed slurry was processed, this time the flotation reagent was added during the pilot scale test work, undertaken at CSIRO Newcastle. Run 3 was a non cavitated run while Run 4 was cavitated.

The results showed that processing this coal with cavitation should have a significant effect. Although the yields are low they are nevertheless consistent with the operation at the plant at that time. On the day of sampling the plant was producing 10%ash product at a yield of 22.7% from Flotation Bank No. 2, according to CSIRO sampling and analysis. It was decided that the tests be undertaken at Macquarie Coal Preparation Plant, even though the use of a combined reagent may be an issue.

The flotation feed is composed of nominally - 0.5 mm coal at approximately 10 wt% solids. The plant uses a combined collector and frother, Fuchs Centifroth. It is added in two stages, firstly at the feed sump and then into Cell 3. The flotation feed is pumped from two sumps to two banks of Denver Flotation Cells. Each bank comprises a 40.1 m3, 5 cell unit. The test work was carried out on Flotation Bank No 2.

The main focus of the project was to compare flotation response with and without cavitation, a second issue was to assess if cavitation could maintain good flotation response when operating at lower chemical addition rates. With these issues in mind the approach taken to conduct a test program was to simply operate on a direct comparison (on/off) basis, sampling the conventional flotation process for up to 1.5 hrs and then switching the feed over so as to pass through the cavitation unit and then sampling over a 1.5 hr period, assuming that the flotation feed has not changed significantly compared to the non cavitated test period. This plant has a stacker/reclaimer stockpile system feeding the plant, which should provide a reasonably homogeneous feed to the plant.

The results from pilot scale tests on the flotation feed from Macquarie CPP showed that cavitation could increase the product yield of the flotation product. However, the work carried out on the full scale in the Macquarie CPP froth flotation plant, while always producing a higher product yield for cavitated runs compared to non cavitated conventional flotation, has not shown as high a response as expected. The statistical analysis of the results of the full scale test work proved inconclusive due mainly to variability of the feed making comparative assessments of cavitated verses non-cavitated runs difficult.

As for the mechanism responsible for improving yields when cavitation is employed, the ash by size analysis gave some insights into this process. The results show that cavitation seems to allow the collection of the ultrafine coal particles. This is supported by the fact that the ashes for the sized fractions for both cavitated and non cavitated were almost the same and there is an increases in the amount of ultrafine material in the cavitated run compared to a noncavitated run. This suggests that the mechanism proposed by Zhou et al., that cavitation produces agglomeration of ultrafine particles, which in turn present as particles of apparent coarser sizes with higher probability of attachment to the large bubbles in a flotation chamber, is the probable mechanism operating within this system.