Gas dispersion plays important role in flotation – Part one

By Rob Coleman*

The recovery in a flotation cell is directly related to the amount of air added to the cell. Therefore there is a minimum air requirement for a given number of solid particles, below which efficient flotation cannot take place. The method by which the air is added to the flotation cell is also vitally important as it controls the size of the bubbles generated and the flow patterns in the cell. The floatation rotor and stator must be designed to provide sufficient turbulence for bubble-particle collisions to occur and be able to generate bubbles in a certain size range depending on the particle size to be floated. The correct flow patterns up the cell of particles and bubbles must then be formed so that the particles are carried up to the froth phase without significant dropback occurring. In other words, if the gas phase is not handled properly, chances are the flotation cell is not performing as well as it could be. So how good is the gas dispersion in your flotation cell and what can you do to optimise it?
There are actually a number of gas phase parameters that can be directly measured and used to optimise the performance of this phase. Typically the gas phase can be described by four parameters:
1. Gas hold-up
2. Bubble size and bubble size distribution
3. Superficial gas velocity
4. Bubble surface area flux

1. Gas Hold-up (eg)
Gas hold-up is the volume of the gas in the pulp zone of a flotation cell. The volume of gas reduces the pulp volume and therefore decreases the residence time available for flotation. The gas hold-up depends on the amount of air added to the flotation cell and is a strong function of pulp viscosity. Through the proven design of the flotation tank and mechanism, gas hold-up in Outotec flotation cells is typically limited to between 5 per cent and 15 per cent of the total pulp volume, thereby maximising the cell volume and residence time.

2. Bubble size and bubble size distribution (db)
Bubble size and its distribution in a float cell’s pulp zone directly affect the particle-bubble interactions and hence flotation performance. For optimal flotation performance, it is critical to generate bubbles of the correct diameter based on the size of particles to be floated. Smaller bubbles are generally required for fine particle flotation and larger bubbles for coarse particle flotation.
Let’s look at the following example, 1 m3 of air contains approximately 566 million bubbles of 1.5 mm diameter. At an aeration rate of 20 m3/min, 189 million bubbles/sec must be generated. Similarly, 1 ton of typical solids contain 1 billion (spherical particles) of 70 microns in size (after grinding). At a solids feed rate of 300 tph, 83 million particles/sec must be treated. This corresponds to 2.3 bubbles per particle. This may seem sufficient - however, due to issues such as poor liberation, incorrect reagent addition and pulp chemistry, and oxidation, flotation recoveries of 100 per cent are never achieved. If the bubble diameter was 2.0 mm, there would only be 80 million bubbles/sec, which would reduce the number of bubbles per particle to less than one. So how do you ensure the optimal bubble size and distribution range? The actual design of the cell mechanism plays a vital role, as does the means of introducing air into the cell. It is also vital to work with a flotation supplier who really understands the intricacies of this process, as poor advice will lead to a waste of time and money.
The bubble size and bubble size distribution can be measured in each flotation cell using a photographic Bubble Sizer. A sample of bubbles is photographed with a digital still camera and an automated image analysis procedure is used to size the collected bubbles from the digital images.
There are two main methods of calculating the average bubble diameter of a distribution. The first is to calculate the average of all bubble diameters in the distribution (known as the average bubble diameter d32). The second is to calculate the sum of all bubbles’ volume divided by the sum of all bubbles’ surface area (known as the Sauter mean bubble diameter d32). The Sauter mean bubble diameter is always larger than the average bubble diameter as it takes more account of large bubbles with large volumes; therefore it is a better measure of bubble size. The Sauter mean bubble diameter is also used to calculate the bubble surface area flux.
Outotec’s flotation mechanism is able to produce small bubbles with average bubble diameters between 1.0 mm and 1.5 mm and Sauter mean bubble diameters between 1.5 mm and 2.0 mm.

* Dr Rob Coleman is currently Technology Leader – Flotation for Outotec in Australia. He has a Chemical Engineering degree from the University of Witwatersrand in Johannesburg and a Doctorate in Minerals Processing from the JK Minerals Research Centre - University of Queensland. He has more than 14 years experience in the operation, design, modelling, simulation and optimisation of flotation circuits and has published and presented papers at many international conferences.

For more information contact tel: +61 (0)2 9984 2500 or email: laura.white@outotec.com

To read the final part of this report see next week’s AJM newswire