April 2016

Future mineral processing challenges

  • By N W Johnson FAusIMM(CP), Senior Principal Consulting Engineer, Mineralurgy Pty Ltd and Adjunct Professor, JKMRC, SMI, University of Queensland

This article is based on the 2015 G D Delprat Distinguished Lecture delivered by NW Johnson

This article presents a view of future challenges in mineral processing. In addition, the future needs of a key related topic, a process analysis and improvement technique, are covered briefly. This technique is important in increasing the performance of existing concentrators and in improving the performance of a new ore during process development (Cameron et al, 1971).

These process analysis and improvement techniques focus on weaknesses in the separation of the liberated particles containing the minerals and/or increasing the level of liberation of the valuable mineral(s) (Johnson, 2010). These weaknesses are detected and quantified by examination of mineral separation data at the mineral recovery-size level and at the more powerful mineral recovery-size -liberation level. From such analyses of the flotation process, three groups of size fractions are typically identified for the minerals: the intermediate size fractions displaying the most efficient separation of the valuable mineral, and the more troublesome fine and coarse size fractions.

For conventional flotation cells, the troublesome combined fine fractions (-10 µm or perhaps -15 µm liberated, dispersed valuable minerals) have different flotation requirements from the troublesome coarse fractions (coarse low quality composite particles containing valuable mineral); the former require turbulent conditions maximising the number of collisions with fine bubbles while the latter require just sufficient turbulence for their suspension to avoid their detachment from the necessary somewhat larger bubbles. Independent setting of the collector/reagent addition rates for a troublesome group of size fractions, without causing adverse effects in other size fractions, is desirable for some systems.

An important topic, the improvement of feed quality early in the mining/processing sequence, is not included in this review. A recent overview is available in Pease et al (2015).

Future direction for concentrate sales contracts

The content of existing concentrate sales contracts is out of date for various reasons; for example, some contracts were written on the basis of downstream processes that are not in use currently and do not reflect current realities of hygiene/environmental issues (Munro and Munro, 2015). Revised future sales contracts will provide increased incentives for the production of higher quality concentrates for sale.

Examination of the quality of the common base metal concentrates on the basis of their valuable mineral content reveals that the zinc concentrates are, in general, the highest quality containing approximately 90 per cent of the valuable mineral by weight; these concentrates demonstrate the concentrate quality achievable in mineral separation circuits. The other common base metal concentrates contain a lower percentage of the valuable mineral (<70 per cent to 85 per cent) and are candidates for improvement in their quality. To raise the concentrate grade, one option is movement along the plant grade-recovery curve by sacrificing recovery of the valuable mineral; however, it would be preferable to employ process improvement techniques to allow operation at the required concentrate grade on a new grade-recovery curve with an improved position.

Future directions for circuit structures with lowered energy requirements

An alternative application of liberation data

Studies of ore samples ground to a wide range of 80 per cent passing sizing values (P80) and with the appropriate collection of liberation data can be used as a cost-effective method for recognition of different design directions. For those ores with suitable properties, these design directions may be radically different from current normal practice. An example with Mount Isa zinc-lead ore is summarised in Pokrajcic (2010). Five samples of the ore were ground to P80 values of 17, 76, 261, 400 and 1607 µm and the liberation levels of the minerals in each size fraction from the five cases were measured; the resulting liberation values for each sample were calculated from weighted summation of the data from the size fractions.

The liberation data were calculated also for the following four entities, where the term ‘recoverable’ refers to particles containing greater than 15 per cent by volume of the following minerals or mineral groupings and therefore potentially ‘recoverable’ in an industrial rougher scavenger separation:

  1. per cent galena recoverable
  2.  per cent sphalerite recoverable
  3.  per cent galena + sphalerite recoverable
  4.  per cent sulfides recoverable ie galena + sphalerite + iron sulfides.

The magnitudes for 1 to 4 at the five P80 values are shown in Figure 1 where the values for grouping 4 remained greater than 90 per cent for the five P80 values; this type of finding for an ore is important because it indicates that non-conventional circuit structures that are more energy efficient can be considered. Hydrophobicity of the galena, sphalerite and iron sulfides is a requirement for initial recovery of the particles in group 4. In contrast, the values on the Y axis of Figure 1 declined from greater than 90 per cent for a P80 of 17 µm to less than 20 per cent for a P80 of 1607 µm for groupings 1 to 3 and such opportunities are not available.

Current commercial conventional flotation cells for sulfide systems would be able to exploit this finding only partially; the current cells may be able to exploit the finding in the region of 261 µm and possibly in the region of 400 µm but exploitation in the region of 1607 µm would be impossible at present. Hence, for full exploitation, a different basis of separation with existing equipment would be required. Existing gravity based technologies are candidates; further, a gravity based separator with some superimposed flotation process characteristics would also be a candidate (eg the Hydrofloat technology of Eriez).

In addition to having suitable coarse particle separators, the following points are relevant:

  1. For flotation based separations, all three sulfide minerals (galena, sphalerite and iron sulfides) must be hydrophobic to confer sufficient hydrophobicity to such composite particles for their initial recovery in roughing.
  2. Regrinding of the relatively low flow of rougher concentrate is much more energy efficient than grinding to the same fineness a much higher flow of rougher feed, the practice usually adopted at present. It can be noted that each halving of the rougher feed P80 value requires 42 per cent additional energy. Hence, for ores with the relevant liberation properties, circuits with significantly lower capital and operating costs can be expected in comparison to the normal type of circuit.

It can also be noted that developments in regrinding technology in the last two-and-a-half decades have resulted in new regrinding mills that are more energy efficient than for the conventional mills that have been used for primary and secondary grinding, ie for producing relatively fine size distributions as the feed for roughing. Hence, an additional source of energy efficiency is noted.

Future directions for equipment in each bank in a circuit

The concepts summarised in this section exist in the literature but are rarely used in industry (Heyes and Phelan 1988). The key development happening currently is that flotation cell manufacturers are seeking to differentiate themselves by matching details of their cell designs to the size of the processed particles. The fine and the coarse fractions are typically more troublesome than the intermediate fractions, as noted earlier.

The following options exist for the flotation of the fine, intermediate and coarse size fractions of valuable minerals in conventional cells:

  1. one conventional cell type in a bank (all size fractions present) implying necessarily tuning for the coarse or fine fractions, or for a ‘compromise’ tuning; no flexibility exist
  2. two conventional cell types in a bank (all size fractions present) allowing tuning one cell type for the coarse fractions and the other cell type for the fine fractions; considerable flexibility exists
  3. separation of the bank feed into, for example, two size fractions and separate flotation of the streams in banks containing cells that match the properties of each size fraction; a high level of flexibility exists, noting that the reagent conditioning before each bank and the staged addition of reagents along each bank are also fully independent in this system.

Further, separation of the bank feed into, for example, two size fractions for independent reagent conditioning is also possible, followed by joining of the two streams for combined flotation (cases 1 or 2 could be used).

The future selection of flotation cells or separation technologies not based on flotation within the described options for bank equipment is now discussed:

  1. for conventional circuit structures, the evolving conventional cells can be expected to provide the necessary capability in each bank, with an increased likelihood of specification of improved conventional or alternative new designs for the scavenging of coarse low quality composites containing valuable minerals.
  2. for non-conventional circuit structures with coarser rougher flotation feeds, proven non-conventional cells or proven radical cell designs are needed.
  3. for very coarse streams, alternative separation equipment (eg gravity based) may be required.

Other benefits from coarser tailing

There are many benefits that follow from circuit designs that provide a coarser tailing sizing as a result of roughing at a significantly coarser sizing than in current normal circuit structures: 

  1. The solid-liquid separation on the rougher tailing becomes less onerous, requiring the installation of less capacity, lowering capital costs. Further, the use
    of filtration of the tailing becomes more accessible, because the required filtration capacity would be lowered (being considered for some current advanced designs around the world).  
  2. The low water content of the filtered tailing allows its stable and safe ‘dry stacking’ on the surface, ie a tailings dam would not be required precluding the possibility of tailing dam failure (typically 1 to 2 failures yearly in the mineral industry world-wide).
  3. The removal of an increased quantity of water from the tailing stream by filtration or other method increases the extent of water recycling, a desirable outcome.

Future directions for water systems and for other topics

Increasing the extent of water recycling leads to minimisation of the externally supplied water for an operation or in planning a new operation, a desirable outcome for the following reasons:

  1. assistance in obtaining permits in arid and semi-arid regions
  2. minimisation of capital and operating costs where the externally supplied water has to be pumped long distances or to high altitudes 
  3. assistance in obtaining approvals in a region with already extensive current water consumption for towns or farming.

Considerable recycling of water at operating sites is already the practice. It is known that recycling water usually increases the concentration of various species (frother, collector, inorganic and other organic species, and microbiological species); from the literature, the effects of recycling water on the flotation process are generally detrimental, with only occasional examples of beneficial effects. Formal methods for evaluation of the magnitude of the detrimental effects are needed, both for existing plants and during circuit development for new operations.

Increased recycling of water by retrieval of additional water from the tailing stream will tend to increase the magnitude of these mainly detrimental effects and increase the incentive for installation of a water treatment plant for a portion of the recycled water to improve the water quality used in the plant, for the following reasons:

  1. improvement in metallurgical performance
  2. lessening of the recycled water as a source of plant disturbances.

In some instances, the externally supplied water may be of sufficiently low quality that a water treatment plant will be justified through the resulting improved metallurgical performance of the plant.

Additional technical challenges also exist:

  1. utilisation of froth region recovery measurements at selected locations in plant surveys directed at process improvement 
  2. development of a standard procedure for determination of the state of dispersion of a pulp.

Further, it is necessary to evaluate the stability over time of data produced by systems containing both multiple sulfide mineral electrodes and the traditionally used electrodes giving a pulp potential measurement
(Woods 2010):

  1. matching of various online cleaning methods for electrodes to the various electrode types to ensure integrity of the data over time in a plant 
  2. application of the proven equipment from item 1 to seeking increased separation efficiency between the valuable sulfide and the gangue sulfide minerals in the operating circuit.

Some models for automated scanning electron microscopy technology exist that have been ‘ruggedised’ for transportation to and use at processing sites. An important future challenge is to provide genuine assistance to around the clock operation by providing rapid offline measurements (eg modal or liberation data) on plant ‘control’ samples (turn-round of a few hours). This task includes providing rapid methods for preparation and presentation of the ‘control’ samples
to the instrument.

Future directions – process analysis

The industry makes very limited use of liberation data for valuable and other minerals to establish the contribution of each grinding and regrinding step in the circuit to the total liberation level achieved (Johnson 2010). Increasing the utilisation of such process analysis data (ie the extent of its generation and interpretation in the industry) is a very important challenge for the future.

The number of size fractions generated routinely in the region 0 to 10 µm is too low for adequate process analysis from plant surveys for two reasons:

  1. insufficient data points exist in this region in recovery size-relationships
    for all ore types  
  2. liberation values continue to increase in this region for very difficult to liberate ores (eg McArthur River and Century).

Final remarks

The existence of an urgent need was a precursor for successful development of the industrial flotation process at Broken Hill. G D Delprat, in whose honour this lecture series is delivered for the AusIMM, was very instrumental in the urgent development of the industrial flotation process.

Future developments will also depend on the perceived ongoing urgency to improve the metallurgical and energy efficiency of mineral separation circuits. An important theme has been the importance for the industry of making greater use of liberation data in routine and non-routine applications:

  1. understanding the contribution of each grinding and regrinding step in the circuit to the total liberation level for each mineral.
  2. recognition of different design directions with lowered energy requirements for those ores with suitable liberation properties from appropriate tests.

When it is possible to conduct roughing at coarser feed sizings, other important implications arise as described in the text for water recycling and safe tailing storage methods.  

REFERENCES

Cameron A, Kelsall D, Restarick C and Stewart P, 1971. ‘A detailed assessment of concentrator performance at Broken Hill South Limited’, Proceedings of the AusIMM, no. 240, pp. 53-67.

Heyes G and Phelan J, 1988. ‘The application of separate conditioning to improve zinc metallurgy at Woodlawn Mines’, Proceedings of the Third AusIMM Mill Operators’ Conference, Cobar, pp. 85-89.

Johnson N W, 2010, ‘Existing methods for process analysis’, in Flotation Plant Optimisation, ed C J Greet, AusIMM, pp. 35-63.

Munro P and Munro S, 2015. ‘Base metals concentrate sales contracts – change Pavlov and the dog’, Proceedings of MetPlant 2015, Perth, pp. 46-60.

Pease J, Walters S, Raassina M, Keeney L and Shapland G, 2015. ‘A step change in mining productivity’, AusIMM Bulletin, April, pp. 52-55.

Pokrajcic Z, 2010. A methodology for the design of energy efficient comminution circuits, Ph.D thesis, JKMRC/SMI/Univ. of Q’ld.

Woods R, 2010. ‘Electrochemical aspects of sulfide mineral flotation’, in Flotation Plant Optimisation,
ed C J Greet, AusIMM, pp. 123-135.

Share This Article