0% found this document useful (0 votes)
237 views2 pages

The Frothbuster - A Solution To Frothing Thickeners Authors: Ian Arbuthnot - Outokumpu Technology Pty LTD Carla Kaboth - Normandy Mining LTD (Golden Grove)

Download as pdf or txt
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 2

The Frothbuster A solution to frothing thickeners

Authors: Ian Arbuthnot - Outokumpu Technology Pty Ltd


Carla Kaboth Normandy Mining Ltd (Golden Grove)
ABSTRACT
One of the most common problems in the thickening of mineral flotation concentrates is the flotation of solids
within the thickener as a result of the presence of fine air bubbles in the thickener feed. The result is a stubborn
froth on the thickener surface which causes significant carryover of solids into the overflow stream. This can
result in concentrate losses and/or expensive recovery procedures. In collaboration with Normandy Mining Ltd
(Golden Grove), Outokumpu Technology have developed a deaeration system which removes air bubbles from
the feed at minimal capital and operating cost, with the potential to produce crystal-clear overflow from concentrate
thickeners.
Frothbuster A solution to frothing thickeners
Motivation
It is recognised in mineral processing operations that the entrainment of finely divided air in thickener feeds can
cause significant problems in the operation of the thickener. This is particularly the case with thickening of
flotation concentrates, in which air bubbles are strongly attached to the mineral particles and are difficult to remove,
which results in the particles rising with the air bubbles to the thickener surface, and hence over the weir and into
the thickener overflow. High solids carryover can lead to a direct loss of mineral product, and if the product settles
in process water tanks and ponds, the recovery of this product by means of intermittent dredging operations is
usually expensive and unreliable.
Many concentrate thickeners are fitted with froth rings or baffles, skimmers and froth sprays, all of which play a
part in reducing the amount of solids carried over into the overflow. However, many froths are very stable and
persist despite these measures. In the worst cases, a thick froth layer forms which builds up to a depth greater
than the froth ring, and hence overflows or underflow it. This can also cause blockage of the overflow weir and
launder, and hence worsen the clarification performance by cutting off sections of the overflow handling system.
The formation of a froth layer on the thickener also makes it more difficult for the operator to monitor the
clarification performance.
Existing Methods of Deaeration
The use of stilling wells on the thickener feed has been partly successful in avoiding heavy frothing and has been
common in the coal industry. In paper pulp operations, both cyclone and vacuum deaeration has been used to
remove air from fibrous pulp which is critical in the papermaking process. However, we were not able to find a
device which could be used for this purpose for the deaeration of mineral slurries.
Development
Outokumpu commenced with the testing of a pilot-scale deaerator in mid 1997 on pyrite and zinc concentrate
feeds. Various configurations of deaerator were tested at three different minesites. Following these tests, it was
decided to scale the test unit up based on the test data. A unit capable of handling up to 40 m3/h of thickener feed
was built and installed in a Perth workshop, together with a concentrate storage tank and a variable speed pump.
Using this test rig, a series of tests was run both on aerated water and aerated concentrate, which allowed the
design of the internals of the deaerator to be refined to give very good results. On tests run on zinc concentrate
from Normandy Minings Golden Grove Operations, and copper concentrate from Newcrest Minings Cadia site,
fine air bubbles were injected into the concentrate slurry as it was pumped to the deaerator. Complete removal
of the air from the feed was achieved in each case, even when large quantities of finely divided air was present.
These tests verified the design of the deaeration unit and provided information on capacity and feed pressure
requirements. Following this test programme, which was completed in October 1999, the production-scale units
were designed.
Production-Scale Deaerators
Based on the successful testwork carried out on their concentrate, Normandy Mining committed to the installation
of a production scale deaeration system on the Zinc Concentrate thickener feed at Golden Grove. The capital cost
justification was based on the revenue loss resulting from carryover of zinc concentrate into the thickener overflow.
This overflow combines with smaller amounts of copper/lead concentrate and tailings material in the process water
pond. This combined material is difficult and costly to recover, due to its fineness, and too low grade to be marketed
as concentrate.
The concentrate system comprises a concentrate hopper which is fed by gravity from the flotation circuit and a froth
pump which delivers the concentrate to a thickener boil box where it mixes with the filtrate from the zinc concentrate
filters and the recycled concentrate from the clean-up sump in the flotation area. From the boil box, the concentrate
gravitates to the thickener feedwell. It was decided to locate the deaerator system on the thickener bridge, and so
the boil box has been bypassed by the main concentrate line, which now delivers directly into the feed manifold to
the deaerators. The concentrate pump hopper has been fitted with a dilution water line so that the concentrate can
be delivered at the desired pressure to the deaerators (i.e. 80-100 kPa).
Three deaerators were installed in a parallel bank, capable of handling 80 m3/h each of feed. Installation of the system
was completed in January 2000. Prior to starting up the system, the froth layer (about 100mm thick) on the thickener
surface was partially cleared so that the effect of the deaerators could be observed. With the deaerators in operation,
residual froth was quickly dispersed and showed no sign of rebuilding. At the recommended operating pressure of
80-100 kPa, the rising solution was observed to be essentially air-free. At times since startup, however, the feed
pressure to the deaerators has dropped to 60 kPa, and at this pressure some air entrainment still persists. This confirms
the test results, which showed consistently that the deaerator performance is pressure dependent. The other factor
which introduces air bubbles into the thickener is that the sump reclaim system and filtrates still bypass the deaerators,
and these streams do contain some entrained air. The plant management at Golden Grove have identified these factors
and are planning to eliminate them by re-routing the bypass streams to the concentrate hopper so that all thickener
feed is treated in the deaerators, and by improving the pump control system to ensure a constant pressure feed to the
deaerators.
Overflow clarity determinations to date with the deaerators on line indicate the 250-300 ppm is a realistic expectation.
This is somewhat better than the average clarity obtained during the pilot-scale testwork, which was 850 ppm; with
the deaerator off-line the pilot scale thickener produced an average overflow clarity of 5040 ppm this represented
an improvement of 83%.
Future Work
A deaerator installation of a similar size to the Normandy system has been installed at Newcrests Cadia operation
in N.S.W. This installation is designed to clean up the copper concentrate thickener, and allow the thickener to accept
future capacity increased without major losses of concentrate into the overflow. It will be commissioned around the
end of March 2000.
Outokumpu Technology are planning the testing of larger deaerators, to allow systems to be designed for larger
concentrate tonnages and also for other industries e.g. coal washing, where thickener feed rates are an order of
magnitude greater than those tested to date. Because the efficiency and capacity of the deaeration system has been
shown during the development work to be related to several physical parameters (including inlet velocity, unit depth
and unit cross-sectional area) in a complex way, it is felt necessary to run full-scale testwork on larger sized units,
to confirm the scale-up to these sizes before introducing them on the production scale.
Acknowledgements :
Normandy Mining
Newcrest Mining

You might also like