The effect of agitation on the flotation of platinum ores

doi:10.1016/j.mineng.2005.01.024

Minerals Engineering

Volume 18, Issue 8, July 2005, Pages 839-844

https://doi.org/10.1016/j.mineng.2005.01.024 Get rights and content

Abstract

Flotation is routinely used for the beneficiation of platinum ores for which the bulk of world production is from South Africa. Most South African platinum concentrators use mechanically agitated flotation cells which operate with power intensities higher than the typical industrial range of 1.0–2.0 kW/m³. This is due to the general philosophy in the platinum industry that increasing power intensity is beneficial to the flotation of finer particles. This paper investigates the influence of agitation on flotation with reference to results from tests on two pilot-scale mechanical flotation cells on South African platinum concentrators. Flotation tests were conducted on a 60 l forced-air flotation cell and a bank of four 150 l induced-air flotation cells over a range of impeller speeds, air flow rates and feed types. Flotation results demonstrate that increasing the level of agitation generally has a beneficial effect on the rate of flotation of platinum ores but that this is accompanied by significant decreases in concentrate grade. These decreases may be due to increases in entrainment or in the rate of flotation of poorly liberated (low grade) particles or floatable gangue. However, since platinum ores are measured in parts per million, relatively small amounts of additional gangue reporting to concentrate will impact significantly on concentrate grade.

Introduction

Froth flotation is a separation method used for the beneficiation of a considerable portion of the world’s mineral ores. Inefficiencies in flotation translate into an enormous loss of revenue and an unnecessary waste of these reserves, particularly in the finer particle size ranges where flotation efficiency is poor. Flotation is routinely used for the beneficiation of platinum ores for which the bulk of world production is from South Africa. Most South African platinum concentrators use mechanically agitated flotation cells. These generally operate with power intensities higher than the typical industrial range of 1.0–2.0 kW/m³ and intensities of up to 10 kW/m³ are in production use. This is due to the general philosophy in the platinum industry that increasing power intensity increases the rate of flotation through improved particle–bubble contacting, which is especially beneficial for fine particles. This paper investigates the influence of agitation on flotation with reference to results from two pilot-scale mechanical flotation cells tested on South African platinum ores.

Section snippets

Mechanically agitated flotation cells

Mechanical flotation cells are the work-horses of the flotation industry and, despite competition from a large variety of alternative flotation technologies, are still responsible for the bulk of world flotation. A mechanical flotation cell consists of square or round tank up to 250 m³ in volume and agitated by an impeller incorporated in a rotor-stator assembly situated near the bottom of the cell. Air is introduced into the rotor-stator assembly, either by induced air suction (induced-air) or

Experimental

The effect of agitation on flotation performance was studied on two pilot-scale mechanical flotation cells. The first study was on a single 60 l forced-air mechanical flotation cell fed by a rougher feed stream (Merensky ore). The second study was on a bank of four 150 l induced-air mechanical flotation cells fed by either a rougher feed or cleaner tails streams (UG2 ore). The UG2 ore rougher feed and cleaner tails grades were both in the vicinity of 3–5 ppm. However, as is typical, the rougher

Results and discussion

Results from the study are given in the three sections ‘Recovery’, ‘Rate constant’ and ‘Grade’. Here the sections ‘Recovery’ and ‘Rate constant’ show the effect of agitation on rates of flotation while the section ‘Grade’ shows the effect on flotation selectivity. For the 60 l cell results are experimental results based on mass balanced data while for the bank of 150 l cell results are from the regression analysis (i.e. response curves). However, in the section ‘Rate constant’ experimental

Recovery

Fig. 1(a) and (b) are graphs of recovery versus impeller speed for 60 and 150 l cells respectively. Here, Fig. 1(a) shows results for two different test conditions while 1(b) shows rougher feed and cleaner tails results at average test conditions (air flow rate, feed grade).

Recovery increases steeply with increasing impeller speed for the 60 l cell and the cleaner tails in the 150 l cells but goes through an optimum for the rougher feed in the 150 l cells and then starts to decrease. The ‘optimum’

Rate constant

Fig. 2(a) and (b) are graphs of flotation rate constant versus impeller speed for 60 l cell and flotation rate constant versus power intensity for the 150 litre cells respectively. Here, Fig. 2(a) again shows results for two different test conditions while 2(b) shows rougher feed and cleaner tails experimental results for the third 150 l cell in the bank of cells. Power intensity is used for representing the level of agitation in the 150 l cells as the results presented in Fig. 2(b) are for the

Grade

Fig. 3(a) and (b) are graphs of the concentrate grade versus the impeller speed for 60 and 150 l cells respectively. Here, Fig. 3(a) shows results for two different test conditions while 3(b) shows rougher feed and cleaner tails results at average test conditions (air flow rate, feed grade).

These figures both show that concentrate grade decreases significantly with increasing impeller speed for all conditions. Decreases in concentrate grade with increasing impeller speed are to be expected when

Conclusions

This paper investigated the influence of agitation on the flotation of platinum ores with reference to results from tests on two pilot-scale mechanical flotation cells on South African platinum concentrators. Flotation tests were conducted on a 60 l forced-air flotation cell and a bank of four 150 l induced-air flotation cells over a range of impeller speeds, air flow rates and feed types. From the review of theoretical and experimental findings from the flotation literature one may conclude that

Acknowledgements

The data used in this paper was obtained from industrial research conducted as part of an Australian Mineral Industries Research Association (AMIRA) P9 project. The authors would like to thank both AMIRA and the South African Platinum producers from whose operations the data was obtained.