Search

The control of coal fragmentation and fines generation during mining and processing is important in coal production. A method to characterise, model and simulate coal size degradation and fines generation based on lithotypes has been developed. This method was refined to cover both high energy single impact to mimic blasting and crushing and low energy incremental breakage to mimic coal handling, transiting, stockpiling and processing. The JKRBT was utilised to characterise high energy single impact breakage and drop shatter tests were used to characterise low energy incremental breakage. X-ray Computed Tomography (XCT) scanning was used as an undisruptive technique to estimate size distributions of drill cores in the drop shatter tests. The JK size-dependent breakage model was applied for breakage characterisation, size degradation modelling and fines generation simulation. The results indicate that coal lithotype has a significant influence on coal degradation and fines generation. This paper has demonstrated that the adaption of two distinct breakage characterisation tests and linkage via the one model is a significant advance in quantifying coal degradation and fines generation during coal production.

Conventional coal washability analysis is achieved by conducting Float and Sink (FaS) test, which is sampling dependent, destructive and uses environmentally hazardous chemicals. With the improvement of modern 2D and 3D imaging technologies, image analysis methods of digital CT scanned images are being developed and trialled to predict coal washability in a more flexible, efficient and safe way. To support the validity of this approach, this study illustrates the fundamental differences in breakage and liberation behaviour of coal lithotypes at the same rank from a single seam. Nine lithotype samples were repeatedly CT scanned before and after a series of 2 m drops to emulate fragmentation during the drop shatter test. The 3D watershed segmentation method was applied to scans to monitor changes in fragment size distribution and then assign a density to each fragment to obtain coal washability. Comparison between the processed image analysis results and FaS tests conducted after 31 drops show a very good match both in particle size distribution and coal washability for all the lithotypes. Results also show that dull coal or stone illustrate better agreement between CT image analysis and FaS test due to their very low percentage of small particles which may lead to underestimation when using CT image analysis. Possible ways of improving small particles estimation would be to use an XY plane H-maxima method and low merging coefficient in the image analysis process.

Flotation is one of the most important primary separation processes in the minerals industry. As far as the mechanism of flotation is concerned, turbulence is one of the key parameters determining flotation performance because it affects three main processes: suspension of particles, air dispersion and particle-bubble collision, attachment and detachment. To study turbulence in industrial flotation cells, both numerical simulation and experimental measurement can be performed. Development of turbulence models and validation of Computational Fluid Dynamic (CFD) numerical simulation need experimental data obtained from turbulence measurement techniques that can be used in the three phase abrasive opaque environment present in a flotation cell. In this paper, the different techniques which have been used to characterise turbulence in the literature are reviewed in terms of their basic principles, system structure, range of application and limitations. Laser Doppler Anemometry (LDA), Particle Image Velocimetry (PIV), Constant Temperature Anemometer (CTA) and the Aeroprobe are all techniques that have been widely used to characterise the turbulence created in flotation machines operating with only fluid (or fluid and air). They cannot however be used when the concentration of solids is high as commonly occurs in a flotation machine. Techniques that have been identified that have the potential to be used to produce accurate measurements in three phase flows include Positron Emission Particle Tracking (PEPT), Piezoelectric Vibration Sensor (PVS) and Electrical Resistance Tomography (ERT). It is envisaged that applications of PEPT in three phase flotation cells will mostly be confined to studies at the laboratory scale. ERT has been tested in flotation cells filled with water and air but needs more development before it can be applied confidently in industrial scale flotation units. PVS, on the other hand, has been validated at laboratory scale and has been applied successfully for measuring turbulence in large scale operating flotation machines.

Mineral liberation is a key step in many mineral processing flow sheets and is achieved by breaking large pieces of ore into smaller particles which are suitable for the subsequent separation process. The ore being broken consists of mineral grains which may exhibit a range of properties affecting how they break. In the literature on mineral liberation the breakage of ore is frequently described as random or non-random, with several types of non-random breakage identified. With numerous researchers investigating this topic over the years, a variety of definitions of random and non-random breakage in mineral liberation have been presented in the literature. This paper examines the published work in this area and provides a comprehensive review of random and non-random breakage published in the literature on mineral liberation. It does not aim to revise the definition of these terms but to review the wide range of descriptors used by researchers in this field and identify common approaches to defining random breakage. The definition could be summarised as random liberation being the independence of breakage from both ore properties and mechanical properties during comminution.

A novel ore pre-concentration technique using high voltage pulses is proposed in this study. The technique utilises metalliferous grain-induced selective breakage, under a controlled pulse energy loading, and size-based screening to separate the feed ore into body breakage and surface breakage products for splitting of ores by grade. Four copper ore samples were tested to demonstrate the viability of this technique. This study consists of two parts: Part 1 presents the principle, the validation and the major findings; Part 2 discusses the new opportunities and challenges for the mining and mineral industry to take up this technique.

With the unprecedented rate of mining and mineral processing over the past decade, there has been an increasing level of concern worldwide about the fate of tailings and the irreversible consequences of contaminant release through dust, tailings dam seepage, dam wall failure, or direct disposal of tailings into waterways. It is likely that in the future the mining of higher tonnages of lower grade ore and the possibility of more extreme weather events associated with global climatic change, may increase the risks associated with mine tailings management. In this paper we review alternative tailings management methods that aim to improve environmental, social and economic outcomes. Three approaches in particular are highlighted: paste and thickened tailings; tailings reuse, recycling and reprocessing; and proactive management (e.g. the integration of sulphide flotation with cemented paste backfill). The aim of the present study is to scope the opportunities that could be gained by developing an integrated research framework that traces tailings properties back to the orebody. In the future models could be constructed to optimise environmental, social and economic outcomes of tailings management across the value chain through integrated orebody characterisation, mine planning, processing, disposal, re-processing, recycling and reuse.

Comparative comminution between high voltage pulses and conventional grinding, at the same specific energy levels, shows that the electrical comminution generates a coarser product with significantly less fines than the mechanical breakage. However, minerals of interest in the electrical comminution product are better liberated than in the conventional comminution with an over 95% statistical significance. There is a potential to use less energy in the electrical comminution to generate the similar degree of mineral liberation as in the mechanical comminution. Distribution of the liberated minerals demonstrates that, in the electrical comminution product, a large percentage of the liberated minerals appear in size fractions coarser than 53 μm; while in the mechanical comminution product, the liberated minerals are accumulated in fine and very fine size fractions. Therefore there may be potential benefits in recovering the coarse liberated minerals in the electrical comminution product, prior to further grinding.

The observed behaviours of mineral particles in mineral processing operations have been exploited in the past to model comminution and concentration processes. In this work this concept has been taken a step further, exploiting the mineralogical characteristics of particles to link comminution, concentration and smelting. This approach is demonstrated using a laboratory-based case study of a Ni-Cu sulphide ore. The case study focused on the effect of shifting energy between the comminution and smelting stages on the overall energy consumption for the metal production process. To model this effect the mineral composition of the particles was linked to the behaviour of the ore particles in the primary grinding, regrinding and flotation stages. This application of process mineralogy provides a methodology to minimise energy use across mineral concentration and smelting processes, an important aspect of sustainable processing. © 2011.

Using a mineral liberation analyser (MLA), the shape properties of an iron-oxide hosted copper–gold ore following comminution in either a hammer mill or a piston–die compression unit were examined. It was found that particle angularity distributions were able to be fitted to the beta distribution, providing a convenient way to quantify changes in particle shape properties as a function of the breakage method employed. Particles discharged from the hammer mill were less angular than particles discharged from the piston–die compression unit. This was attributed to attrition breakage mechanisms, whereby topographical features are chipped off parent particle surfaces. The results suggest that the quantification of shape properties is a convenient method for obtaining insight into the nature of breakage events that take place inside comminution devices.

The role of particle shape properties in determining flotation kinetics was investigated in this study. -150/+75 μm chalcopyrite was floated in a microflotation cell, and shape properties of the timed concentrates were quantified using a mineral liberation analyser - a mineralogical characterisation system based on automated scanning electron microscopy. It was found that, when floated in the absence of collector, particles with angular shape features reported to the concentrate faster than round particles. X-ray photoelectron spectroscopy (XPS) and time-of-flight secondary ion mass spectroscopy (ToF-SIMS) showed that angular particles also exhibited surface chemistry characteristics favourable for flotation. When particle floatabilities were high (i.e. when floated in the presence of potassium amyl xanthate), shape properties were not found to contribute significantly to the rates at which particles floated. © 2010 Elsevier Inc. All rights reserved.

A new comminution method has been developed by applying high voltage pulses at specific energy 1– 3 kWh/t to pre-weaken mineral particles, leading to reduction in energy consumption in the downstream grinding process. Four ore samples were tested using high voltage pulses and conventional crushing in parallel for comparison. Evidence of cracks and microcracks measured with X-ray tomography and mercury porosimetry supported the principle of high voltage pulses induced damage on rocks in the electrocomminution process, which resulted in energy saving up to 24% found in this study. Ore surface texture and mineral properties affected the efficiency of high voltage pulse breakage. The feasibility of the electro-comminution and its benefits need to be investigated case by case.

The effect of particle-bed breakage mechanisms on the liberation properties of ores remains unclear. Surprisingly few studies have been published in this area and limitations in liberation measurement techniques previously used have prevented definitive conclusions from being reached regarding whether particle-bed breakage enhances the liberation properties of mineral ores relative to conventional grinding mechanisms. In this study, two sulphide ores of differing textures were comminuted to various size distributions using impact and particle-bed breakage mechanisms in a hammer mill and a piston–die compression unit respectively. The liberation properties of the various discharge samples were then characterised using a mineral liberation analyser – a mineralogical characterisation system based on automated scanning electron microscopy. It was found that the size-by-size liberation properties of both valuable and gangue mineral phases were independent of both the method used to comminute the samples, as well as the particle size distribution of the final products. These effects are discussed in terms of how they may be exploited in liberation modelling and characterising comminution circuit performance.

Hydrophobicity of ZnS particles in the 150–212 μm size fraction was examined using film flotation. It was revealed that hydrophobicity of studied ZnS particles was affected by slime coatings as well as prolonged conditioning in water. Conditioning in concentrated methanol solution resulted in increased hydrophobicity of ZnS particles, similar to the activation with CuSO 4 . Copper activation with CuSO 4 is a common procedure to increase hydrophobicity of ZnS and to facilitate xanthate adsorption to form stable chemical compound. Conditioning ZnS with concentrated methanol was shown to have beneficial effect on the increase of hydrophobicty of ZnS particles. The atomic force microscopy (AFM) with tapping mode imaging technique, was used to examine ZnS surface before and after the methanol treatment. The nanobubbles were found on the ZnS surface after the methanol treatment and this led to an increased in hydrophobic force and an increase in the range of attraction between examined hydrophobic surfaces. The slime effects on hydrophobicity of ZnS surface were investigated using contact angle (captive bubble and sessile drop). The slime coatings before and after methanol treatment were observed with AFM. The methanol conditioning was shown to result in aggregation of slimes and eventually cleaning the surface of the mineral.

This paper presents and interprets results of experimental measurements of the spatial gas hold-up distribution in a 3 (3) glass rectangular flotation cell at the JKMRC using two different techniques. The gas hold-up device with the capturing technique was developed at the JKMRC and has been used widely in the P9 project(1) while the one with conductivity technique was developed at the CSIRO Thermal and Fluids Engineering laboratory at Highett, Victoria, Australia. Measurements were conducted at more than 64 locations in the cell to determine the local gas hold-up distribution in the cell. Since the measurements using the two techniques were conducted at the same locations, the results may be compared with each other. The results indicate that the gas hold-up varies widely inside the flotation cell. The gas hold-up distributions measured by the two techniques are relatively similar except in some locations which can be reasonably explained. (c) 2006 Elsevier Ltd. All rights reserved.

It has been observed in several Jameson cell installation where the source for flotation feed is deslime screens, that the recovery of coal particles greater than 0.5 mm is not as great as that of finer material. Consequently, a research project was undertaken at a CHPP in the Bowen Basin Queensland to assess the possibility of increasing the recovery of coarser particles (+0.5 mm) within the downcomer of the Jameson cell. The effect of decreasing turbulence and agitation in a commercial-scale downcomer was investigated to assess the effect on the recovery of both coarse and fine coal particles. This paper details the findings of the test work, summarising the results relating to differences in the operating parameters within the downcomer.

A technique for determining the recovery of attached particles across the froth phase in flotation that relies on measuring the rate at which bubble-particle aggregates enter the froth is used to investigate the selectivity of attached particles across the froth phase. Combining these measurements with those of other techniques for determining the froth recovery of attached particles provides an insight into the different sub-processes of particle rejection in the froth phase. The results of experiments conducted in a 3 m(3) Outokumpu tank cell show that the detachment of particles from aggregates in the froth phase occurs largely at the pulp-froth interface. In particular it is shown that the pulp-froth interface selectively detaches particles from aggregates according to their physical attributes. (C) 2005 Elsevier Ltd. All rights reserved.

Solids concentration and particle size distribution gradually change in the vertical dimension of industrial flotation cells, subject primarily to the flotation cell size and design and the cell operating conditions. As entrainment is a two-step process and involves only the suspended solids in the top pulp region near the pulp-froth interface, the solids suspension characteristics have a significant impact on the overall entrainment. In this paper, a classification function is proposed to describe the state of solids suspension in flotation cells, similar to the definition of degree of entrainment for classification in the froth phase found in the literature. A mathematical model for solids suspension is also developed, in which the classification function is expressed as an exponential function of the particle size. Experimental data collected from three different Outokumpu tank flotation cells in three different concentrators are well fitted by the proposed exponential model. Under the prevailing experimental conditions, it was found that the solids content in the top region was relatively independent of cell operating conditions such as froth height and air rate but dependent on the cell size. Moreover, the results obtained from the total solids tend to be similar to those from a particular gangue mineral and hence may be applied to all minerals in entrainment calculation. (C) 2004 Elsevier Ltd. All rights reserved.

A new device has been developed to directly measure the bubble loading of particle–bubble aggregates in industrial flotation machines, both mechanical flotation cells as well as flotation column cells. The bubble loading of aggregates allows for in-depth analysis of the operating performance of a flotation machine in terms of both pulp/collection zone and froth zone performance. This paper presents the methodology along with an example showing the excellent reproducibility of the device and an analysis of different operating conditions of the device itself.

Modelling of froth transportation, as part of modelling of froth recovery, provides a scale-up procedure for flotation cell design. It can also assist in improving control of flotation operation. Mathematical models of froth velocity on the surface and froth residence time distribution in a cylindrical tank flotation cell are proposed, based on mass balance principle of the air entering the froth. The models take into account factors such as cell size, concentrate launder configuration, use of a froth crowder, cell operating conditions including froth height and air rate, and bubble bursting on the surface. (C) 2004 Elsevier Ltd. All rights reserved.

Over recent years a number of techniques have been developed as part of the AMIRA 2 P9 project to represent mathematically or measure directly sub-processes occurring in flotation cells. These include the determination of: • a machine parameter (bubble surface area flux, Sb); • a froth parameter (froth recovery, Rf); • entrainment and drainage parameters (ξ and δ); and • an ore floatability parameter (Pi). A methodology for estimating froth recovery in large scale plant flotation cells is proposed. The primary aim of the technique is to estimate the froth recovery in a way that is simple, inexpensive and non-intrusive. Plant engineers can use the estimate of froth recovery in day to day trouble-shooting and circuit optimisation. The current definition of froth recovery is given, along with a detailed description of the proposed methodology, equipment required, assumptions and calculations. Advantages and limitations of the methodology are described and possible applications of the froth recovery estimate in current flotation circuit modeling are presented.

This paper proposes an integrated methodology for modelling froth zone performance in batch and continuously operated laboratory flotation cells. The methodology is based on a semi-empirical approach which relates the overall flotation rate constant to the froth depth (FD) in the flotation cell; from this relationship, a froth zone recovery (R,) can be extracted. Froth zone recovery, in turn, may be related to the froth retention time (FRT), defined as the ratio of froth volume to the volumetric flow rate of concentrate from the cell. An expansion of this relationship to account for particles recovered both by true flotation and entrainment provides a simple model that may be used to predict the froth performance in continuous tests from the results of laboratory batch experiments. Crown Copyright (C) 2002 Published by Elsevier Science B.V. All rights reserved.

The mineralogical characteristics of enargite-bearing copper ores from the Tampakan deposit have been investigated as the means to understanding the separation of enargite from other copper sulphides (chalcopyrite and bornite). It was found that a separation of these minerals was possible after grinding the ore samples to a P80 of 90 microns; enargite tends to be ground more readily than the other copper sulphide minerals present in ore samples tested, and consequently, it mostly reported to the fine size fractions.

The impact of fluorine in copper flotation was relatively unknown until the introduction of skarn ores in the Ok Tedi concentrator. Fluorine in the copper concentrates reports to the gas phase during the smelting stage and forms a corrosive H2SO4-HCl-HF acid brine mixture which must be neutralised. This work was aimed at studying the mineralogy of the fluorosilicate minerals contained in the various oretypes present in the Ok Tedi porphyry copper deposit. The electron microprobe was used to analyse for fluorine and hence identify the fluorosilicate minerals in each oretype. This study revealed talc, phlogopite, biotite, clays, amphiboles, fluoroapatite and titanite to be the sources of fluorine in the orebody. Laboratory and plant investigations were conducted to study the flotation response of these minerals. Chemical assaying of the products of these tests was done to determine the bulk assay of fluorine, Using Rietveld analysis, quantitative estimates of the fluorosilicate minerals in these products were generated. Marrying of the bulk assay with the respective mineralogical assay enabled the understanding of the flotation behavior of fluorine and it's associated mineralogy. Talc and phlogopite were found to be the causes of the fluorine problem at Ok Tedi. (C) 2001 Elsevier Science Ltd. All rights reserved.

A study was undertaken to determine the effects of five industrial frothers on the relationship between entrainment recovery and water recovery. The experiments were conducted using a Batequip rectangular 60-L steel flotation cell that was run in parallel with the first cell in a copper/silver prefloat rougher circuit. The concentration of the frothers ranged from 2.5 to 8.4 g/t. The experiments were repeated at three different froth depths. A power function was found to fit this relationship with no discontinuity in the relationship caused by the changes in frother type and concentration and froth depth.

A flotation model developed at the Julius Kruttschnitt Mineral Research Centre was used to predict the performance of copper rougher and scavenger circuits at the copper concentrator at Mount Isa Mines. The model describes the kinetic response of an ore in a flotation cell in terms of ore floatability, cell characteristics and froth recovery. The float-ability parameters of the rougher feed ore were obtained directly from a detailed plant survey. Flotation analysis was carried out for twenty different size-by-liberation classes to account for the effects of particle size and liberation on flotation. This paper investigates the capability of the model to predict the flotation behavior of 20 different size-by-liberation classes in the rougher and the scavenger banks.

This paper presents the results of an investigation in which the new JKMRC Flotation Cell was used to determine the collection zone rate constant and froth zone recovery of a copper rougher ore simultaneously. The determination of these two parameters has been based on the straight line relationship that exists between the overall flotation rate constant and the froth depth [1]. Experimental work was conducted using a copper rougher ore with a P 80 of 200 μm. Operating variables such as air flow rate, impeller speed, feed percent solids, collector and frother dose, and wash water flow rate were investigated. Analysis for copper and iron minerals (chalcopyrite and pyrite, respectively) was carried out. The results indicate that the collection zone rate constant of both copper and iron minerals increased with increasing air flow. Froth zone recovery, on the other hand, decreased as air flow was increased, possibly as a result of increased detachment of particles from bubbles in the froth. Increasing the impeller speed also increased collection zone rate constant and decreased the froth zone recovery of both minerals. Experiments at different wash water flow rates have showed that events occurring in the froth zone do not affect the kinetics of the pulp zone. Moreover, and interestingly, the froth recovery of attached particles (wash water reduced entrainment to a minimum) was non-selective. The froth recovery curves for chalcopyrite and pyrite followed each other very closely in every instance studied. The work has proved that it is possible to measure both the collection zone rate constant and froth zone recovery simultaneously and continuously in a mechanical flotation environment. The results obtained to date are interesting and the work is continuing.

The High Bubble Surface Area Flux Flotation Cell (HSbFC) is a 16-litre mechanical flotation cell with a bottom driven impeller, which is operated continuously. Bubble formation is carried out using an in-line mixer which enables the cell to achieve superficial gas velocities (J(g)) equivalent to those generated in industrial flotation cells (0.7-1.2 cm/s). At the same time, the cell produces considerably smaller bubble sizes, consequently, a high bubble surface area flux (Sb) Can be generated. (C) 1999 Published by Elsevier Science Ltd. All rights reserved.

A previous investigation carried out by the authors at the Hellyer concentrator, using a 3 m3 cell fitted with four different impellers treating plant zinc cleaner feed ore, suggested a linear correlation between flotation rate constant k and bubble surface area flux Sb. The relationship between k and Sb was found to be independent of impeller type. This paper describes an investigation at the Scuddles concentrator in Western Australia to validate the findings at Hellyer for a different ore. Unlike the Hellyer work for which only one froth depth was used, the Scuddles work was carried out at different froth depths using the same 3 m3 cell fitted in turn with three different impellers viz. Batequip, Dorr-Oliver and Outokumpu. The results confirmed the strong correlation between k and Sb at three different froth depths used for the study. Moreover, this relationship was found to be practically independent of impeller type. However, at shallow froth depth the kSb relationship was found to be linear, whereas at intermediate and deep froth depths the relationship was found to be non-linear with the froth playing an important role in the overall kinetics.

Previous studies by the authors have shown that there is a strong correlation between the overall flotation rate constant k and the bubble surface area flux, S b , in a flotation cell, and that this relationship is independent of impeller type. More recently it was found that results obtained from a 250 litre cell and a 3 m 3 cell treating the same feed could be represented by the same k-S b relationship, as long as the ratio of froth height to superficial gas rate in the cells was the same. This ratio is termed the froth residence time, τ fg . This paper examines the effect of τ fg on the kinetics of flotation, by re-evaluating and re-interpreting these results. It is found that k decreases exponentially with τ fg but the relationship is dependent on cell size and therefore not directly useful for scale-up. When τ fg is divided by a typical cell dimension to take into account the effect of froth transportation distance in cells of different sizes, the relationship between k and the “specific froth residence time” fits the data better and and is found to be independent of cell size. Finally, when the results are evaluated in terms of τ fg , the froth residence time calculated on the basis of the concentrate slurry flow rate, the fitting od the data is found to be very much better again. The implications of these relationships for flotation modelling and scale-up are discussed.

The entrainment and drainage of particles in a flotation froth have long been recognised as important factors affecting both concentrate grade and recovery. In this paper, an empirical partition curve is proposed that describes the degree of entrainment within a conventional flotation cell. Two empirical parameters are defined: • • entrainment parameter (ζ)—the particle size for which the degree of entrainment is 20%. • • drainage parameter (δ)—related to the preferential drainage of coarse particles. The model is flexible enough to fit a broad data set published in the flotation literature. Results from a case study indicate that the model is sufficiently detailed to describe the entrainment and drainage mechanisms. This being so, the effect of process variables which affect the degree of entrainment can be properly quantified and then included in a simulation package, assisting in the optimisation of existing flotation circuits and the design of future ones.

The metallurgical performance of a 2.8m 3 portable industrial scale flotation cell was measured when treating zinc cleaner feed at Hellyer concentrator in Tasmania, Australia. The cell was fitted in turn with four different impeller-stator systems and operated over a wide range of air flow rates and impeller speeds. Bubble size, gas holdup and superficial gas velocity were measured at each of 64 different operating conditions along with the metallurgical performance of the cell. When metallurgical performance was expressed in terms of a kinetic constant, it was found that neither bubble size nor gas holdup nor superficial gas velocity could be related to flotation rate individually; but when taken together, they determine the bubble surface area flux in the cell, which could be related to flotation rate extremely well. A linear relationship between flotation rate and bubble surface area flux was found for all four impellers investigated: the slope of the line was independent of the type of impeller used. The linear relationship was verified for different size fractions of the ore: the slope of the straight line was different for different size fractions, values being greater for the smaller size particles. The relationship was also independently confirmed at another zinc cleaner operation. This finding has potential practical application in flotation plant modelling, design and optimisation.

Superficial gas velocity was measured at different locations in a 2.8 m 3 portable industrial scale sub-aeration flotation cell, treating zinc cleaner feed at Hellyer Concentrator in Tasmania, Australia. The cell was fitted in turn with four different impeller-stator systems, and operated over a range of air flow rates and impeller speeds. Superficial gas velocity values ranged from 0.29 cm/sec to 6.4 cm/sec at the air flow rates employed. The distribution of air in the cell was very dependent on the impeller type, as well as on the operating conditions in the cell. For a well dispersed condition, the value of superficial gas velocity was uniform at different locations, whereas for flooding conditions the superficial gas velocity was very high near to the impeller and much lower away from the impeller.

Gas holdup was measured at different locations in a 2.8 m 3 portable industrial scale subaeration flotation cell, treating zinc cleaner feed at Hellyer Concentrator in Tasmania, Australia. The cell was fitted in turn with four different impeller-stator systems, and operated over a range of air flow rates and impeller speeds. The gas holdup was found to increase with increase in impeller speed as well as with increase in air flow rate, the manner in which it increased depended on the impeller type. Values ranged from 2% to 33%, with the greatest values produced by the Outokumpu impeller.

Bubble size distributions were measured at different locations in a 2.8 m3 portable industrial scale sub-aeration flotation cell, treating zinc cleaner feed in the Hellyer Concentrator in Tasmania, Australia. The cell was fitted in turn with four different impeller-stator systems, and operated over a range of air flow rates and impeller speeds. The mean bubble size was found to increase with increase in air flow rate at different locations in the cell, for all four impellers, and to decrease with increase in impeller speed. The mean bubble size was largest close to the impeller shaft and smallest at the impeller discharge point, for all the impellers. The shape of the bubble size distribution also changed with location in the cell. The “global mean” bubble size calculated by simple arithmetic average of the values at six locations in the cell coincided remarkably well with the mean bubble size measured halfway between the impeller shaft and the side of the cell, at the top of the pulp. In general, the impellers produced “global mean” values of 1.0 mm or less at the manufacturer's recommended impeller speed.