See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/283119643

Integrating Bulk Ore Sorting into a Mining Operation to Maximise Profitability

Conference Paper · September 2015

CITATIONS READS

15 8,810

4 authors:

Kristy-Ann Duffy Walter Valery

Hatch Ltd. Brisbane The University of Queensland
29 PUBLICATIONS 88 CITATIONS 111 PUBLICATIONS 879 CITATIONS

SEE PROFILE SEE PROFILE

Alex Jankovic Peter Holtham

Hatch Hatch
78 PUBLICATIONS 884 CITATIONS 90 PUBLICATIONS 1,345 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Pre-concentration View project

Throughput Forecast and Geometallurgical Modelling View project

All content following this page was uploaded by Kristy-Ann Duffy on 27 November 2016.

The user has requested enhancement of the downloaded file.

 HOME

273

Integrating Bulk Ore Sorting into a Mining Operation to

Maximise Profitability
K Duffy1, W Valery2, A Jankovic3 and P Holtham4
1. Manager – Mineral Processing, Metso Process Technology & Innovation,
PO Box 221, Kenmore 4069, [email protected]
2. Senior Vice President - Global, Metso Process Technology & Innovation, PO Box 221,
Kenmore 4069, [email protected]
3. General Manager - Technology & Innovation, Metso Process Technology & Innovation,
PO Box 221, Kenmore 4069, [email protected]
4. Engineering Consultant, Metso Process Technology & Innovation, PO Box 221,
Kenmore 4069, [email protected]

ABSTRACT
As high grade ore deposits are depleted, remaining deposits are generally lower grade and require the
mining and processing of larger volumes of material per tonne of product. These low grade deposits
often contain a large proportion of barren gangue. Pre-concentration aims to remove this barren
material at as coarse a particle size and as early in the process as possible. This has the potential to
reduce processing costs, energy and water consumption, increase production and improve resource
utilisation.

There are several technologies that may be applicable for pre-concentration. This paper looks
specifically at bulk ore sorting and how it can be integrated into existing or new mining operations to
improve profitability and reduce environmental impact. Bulk ore sorting is the separation of barren
gangue from a fully loaded conveyor belt based on the grade as measured or inferred from a sensor
measurement. It may be used to increase the production rate through fixed plant capacity, or increase
production and/or reduce the size of downstream processing equipment in Greenfield operations.
Marginal ore sub-economic reserves may be upgraded, making them economic to treat and improving
the resource utilization. Bulk ore sorting could also be used to reduce dilution and ore loss in mining
operations by improving grade control, or to separate ore types to treat via different process routes.
Various applications of bulk ore sorting are discussed in this paper, and three cases are investigated
considering the impact on the overall operation and the associated benefits.

INTRODUCTION
Due to the depletion of high grade deposits, remaining deposits are generally lower grade and more
difficult to extract, requiring the mining, movement and processing of larger volumes of material per
tonne of valuable mineral. The transportation, comminution and processing of these large volumes is
expensive and energy intensive. Low grade ore bodies generally contain a large proportion of
liberated barren gangue which may be removed at coarse sizes. This rejection of gangue from the feed
material, termed pre-concentration, reduces the total mass and increases the grade of ore proceeding
to the next stage of processing (Pokrajcic and Lewis-Gray, 2010). The rejection of material below the
defined cut-off grade avoids feeding the plant with material that will cost more to process than the
respective value of the contained valuable mineral.

The aim of pre-concentration is to remove barren material at as coarse a particle size and as early in
the process as possible. Gangue is usually high in silicates and typically harder and more competent
than the valuable minerals. Removal of this hard and barren material prior to comminution stages has
the potential to significantly reduce energy consumption and processing costs, and may also reduce
ore transport requirements. Pre-concentration can significantly improve resource efficiency by
upgrading uneconomic or marginal material and/or increasing production rates. It may be possible to
reduce the cut-off grade depending on the net balance of reduced processing costs (due to gangue
 274

rejection) and new costs associated with pre-concentration and mining additional material. More
valuable metal may be extracted from the resource while the processing plant treats less tonnes at
higher feed grade.

For effective pre-concentration, it is only necessary to have liberated gangue that can be removed; the
valuable minerals do not need to be liberated. There are several technologies that may be applicable
for pre-concentration including: gravity processes, magnetic separation, sensor based ore sorting and
screening. The suitability in each case depends on the ore properties. This paper looks specifically at
sensor based ore sorting; however, other alternatives which may be cheaper and/or simpler such as
gravity concentration and pre-screening should always be evaluated.

Pre-concentration with an ore sorter has the added advantage that it can complement mining
selectivity at the bench or stope by determining grade. It is an efficient way to deal with uncertainties
of grade, particularly where the complexity of mine geology makes the estimation of grade difficult.
This helps the mining operation to achieve the planned cut-off grade and optimise extraction of the
resource.

CURRENT ORE SORTING TECHNOLOGY

Ore sorting relies on measuring a property that is different in the valuable and waste components
using some form of sensor. A variety of sensors are available, and those commonly used in industrial
applications include photometric, electromagnetic, radiometric and x-ray. A summary of sensor types
is provided in Table 1.

Table 1 – Summary of sensors.

Sensor Type Separation Property Areas of application Current application / Speed Penetration
development status
Magnetic resonance Excitation and detection of Only applicable to selected Development for bulk Seconds Penetrating
(MR) specific radio frequency minerals (currently sorting
spectral lines chalcopyrite)
Radiometric (RM) Natural gamma radiation Uranium, precious metals Particle sorting Real-time Penetrating
Dual Energy X-ray Absorption x-rays (atomic Base and precious metals, Particle sorting Real-time Partly
Transmission density) industrial minerals, fuel, penetrating
(XRT) diamonds
X-ray Fluorescence The emission of secondary Diamonds Particle sorting Real-time Surface
(XRF) (fluorescent) X-rays
Colour (CCD Reflection, absorption, Base and precious metals, Particle sorting Real-time Surface
Colour Camera) transmission industrial minerals,
diamonds
Photometric (PM) Monochromatic Reflection Industrial minerals, Particle sorting Real-time Surface
/ Absorption diamonds
Electromagnetic Conductivity, permeability Base metals Particle sorting Real-time Penetrating
(EM)
PGNAA Prompt-gamma neutron Limestone, Fe, Al, Pb, Mn, Measurement only Minutes Penetrating
activation analysis Cu, Zn
PFTNA Pulsed fast and thermal Ni, Fe, Co, Mg, Si, Al, Mn, Measurement only Minutes Penetrating
neutron activation Cr, C, H, O
Near Infrared Reflection, Absorption Base metals, industrial Recycling applications Real-time Surface
Spectrometry (NIR) minerals
Infrared (IR) Heat conductivity, heat Base metals, industrial Development Real-time Surface
dissipation minerals
Laser-induced Atomic spectroscopy using Raw materials Raw material applications Real-time Surface
breakdown a highly energetic laser
spectroscopy pulse.
(LIBS)
Laser-induced A spectroscopic method, Raw materials Raw material applications Real-time Surface
fluorescence (LIF) measurement of photon
emissions
Eddy-current High-frequency changing Recycling Recycling applications Real-time Penetrating
magnetic fields create
strong eddy currents in the
non-ferrous metal parts
 275

Sensor based ore sorting is not new, and has been shown to be technically feasible. However, in the
minerals industry, it is currently only used in some niche applications such as industrial minerals (e.g.
calcite, rock salt or talc), diamonds and other gemstones. The current technology, based on the
measurement and separation of individual particles, is well documented in the literature (for example
Manouchehri, 2004; Wotruba, 2006; and Bergmann, 2009) and in publications by suppliers.

Current sorters are either belt or chute type sorters, and both require five process steps: material
conditioning, material presentation, detection, data processing and separation, as shown in Figure 1.
Careful feed preparation is required so that individual particles can be detected and measured, and
ejection of single particles is usually achieved by blasts of compressed air. Consequently, current
sorters have very low capacity (up to 300 tph for larger particles and much less for smaller particles).
They would not be economically viable for high tonnage pre-concentration. For this, ore sorting needs
to be applied to bulk quantities of ore, such as on a loaded truck tray or a fully loaded conveyor belt.

Belt Type Sorter Chute Type Sorter

Material conditioning - Sizing to provide feed in narrow size band and washing (scrubbing) if necessary
Material presentation - Present individual particles to sensors (belt or chute/freefall)
Detection – Measurement of material properties with a sensor or a combination of appropriate sensors
Data processing – Analysis of data from sensors to assign individual particles as accept or reject
Separation - Mechanical separation of accepts and rejects (usually by air ejection).

Figure 1 – Current ore sorting technology (for individual particles).

BULK ORE SORTING

The concept of bulk ore sorting involves the separation of a large volume of barren gangue from a
fully loaded conveyor belt based on the grade as measured or inferred from a sensor measurement.
Most mining deposits are heterogeneous which should allow the separation of such large volumes if
sensing is conducted early in the process, before excessive handling and mixing occurs. Highly
disseminated ores are unlikely to be amenable to bulk ore sorting.

A review of existing sensor technologies indicates that most are currently not suitable for bulk ore
sorting, as they are either not sufficiently penetrating or are too slow for effective separation, as can
been seen from the summary of sensors provided in Table 1. For example, laser-induced breakdown
spectroscopy (LIBS), laser-induced fluorescence (LIF) and photometric sensors are surface only
measures (not penetrating into the rock). X-ray fluorescence (XRF) has a beam size and penetration
of only a few millimetres. Therefore, these sensors cannot provide a representative measure for large
quantities of heterogeneous material such as required for bulk sorting.

Prompt-gamma neutron activation analysis (PGNAA) and pulsed fast and thermal neutron activation
(PFTNA) sensors measure elements and can penetrate the full cross section on a loaded conveyor belt.
However, currently the measurement speeds are too slow for effective bulk ore sorting, in the order of
minutes rather than seconds. It may be possible adapt these sensors for bulk sorting; the trade off
would be some reduction in accuracy and increased cost.
 276

The CSIRO is developing a sensor using magnetic resonance (MR) that has the ability to rapidly
measure batches of ore on large primary production conveyors (Miljak, 2011). The MR sensor is well
suited to a bulk ore sorting application as it is penetrative and can measure large throughputs on fully
loaded conveyor belts. In addition, the measurement response time is rapid, thus allowing diversion
of different grade streams in an ore sorting application. However, the MR sensor measures an
individual mineral (not element) and may have limitations measuring ores with complex mineralogy.
The sensor is currently developed for chalcopyrite, a dominant copper mineral, and with further
development could potentially be applied to other minerals (Heselev, 2012).

Bulk ore sorting may incorporate more than one type of sensor to overcome the limitations of the
different sensor types. It also requires a control system, to interpret the data from the sensor or sensors
and make an accept or reject decision, and a diversion system such as a diverter gate to separate the
valuable “batches” of ore from waste, as shown in Figure 2. Metso has developed conceptual designs
for implementation on plant feed or in-pit conveyors treating up to 3,600 tph and belt speeds of up to
5 m/s. Ore sorting benefits from the natural heterogeneity of deposits, and should be implemented as
early in the process as possible where the variability is greatest and to maximise the benefits.
Therefore, either in-pit or plant feed conveyors provide the best opportunities for bulk ore sorting.

Figure 2 – Metso bulk ore sorting concept.

AMENABILITY OF ORE TO BULK ORE SORTING

To be amenable to separation by bulk ore sorting the material presented to the sorter needs to have
sufficient grade variability occurring in large enough batches of material for effective separation. For
example, separation of a couple tonnes or more of material may be possible for large tonnage, fast
moving conveyor belts (say 3600 tph and 5 m/s). In this case, effective pre-concentration could be
achieved with a bulk ore sorter if the material presented to the sorter contains batches of below cut-off
grade material that are at least one to two tonnes or more.

Very little is currently known about the scale and frequency of grade variability in Run-of-Mine
(ROM) or plant feed material, as it is difficult to measure. It depends on the in-situ variability of the
ore deposit and also the mining methods, ore handling, crushing and blending prior to delivery to the
plant. Understanding the grade variation is important to determine the required response time for
measurement and separation (diversion), and also the potential grade uplift which will determine if
there is economic value in bulk sorting for each case.
 277

Most deposits are heterogeneous; which would make them potentially amenable to upgrading with
bulk ore sorting. However, current mining practices are generally designed to blend out the variation
and provide a consistent, stable feed to the processing plant.

The mining industry typically takes a ‘one size fits all’ approach, and endeavours to treat all the ore
from a deposit through one extraction process. The process is designed for the average or typical ore,
and a consistent feed grade is requested from the mine so the process can be stabilised and optimised.
Therefore, mining operations generally have blending strategies to deliver, as far as possible with the
ore available, a stable feed grade to the plant. It is not unusual to have a number of stockpiles on the
ROM pad with different grade classifications which can be fed to the process in the required
proportions to achieve a desired feed grade.

Underground mines can have hundreds of draw points, and be extracting from many of these at the
same time. The active draw points are generally dictated by mining requirements, but then may be
selected from based on draw point grade control samples to meet plant feed grade requirements.
Significant mixing occurs through ore passes, underground crushing and materials handling prior to
delivery to the plant. The Northparkes block cave mines illustrate this point. The schematic layout
for E26 Lift 2 mine is shown in Figure 3, and Figure 4 shows the E48 Lift 1 mine materials handling
system. These demonstrate the number of draw points and materials handling steps (ore passes, bins,
crushers, conveyors, feeders).

Figure 3 – Schematic layout of the E26 Lift 2 block cave mine (after Butcher et al, 2013).
 278

(1) Extraction from draw points by 4 – 5 continuously operating Load Haul Dump Units (12 t bucket)
(2) Ore is fed to a coarse ore bin that feeds the primary crusher (800 t live capacity)
(3) Plate feeder (1000 tph)
(4) Gyratory crusher (1000 tph)
(5) Crusher ore pass / bin (400 t live capacity)
(6) Vibratory feeder (1050 tph)
(7) Ore is transferred by high-speed inclined conveyors to the underground loading station
(8) Underground loading station - three ore passes feeding the hoisting system (two 16 t payload skips)
(9) Hoisted ore is transferred via a short conveyor to a secondary crushing station
(10) Ore from the secondary crusher to then conveyed to the concentrator by a curved high speed conveyor

Figure 4 – E48 Lift 1 mine materials handling system (after Butcher et al, 2013).

Every time the ore is rehandled, transferred, crushed, blended etc. the degree of mixing increases;
reducing the variability and thus the potential for effective separation of batches of barren gangue
(waste) from ore. To maximise the value of bulk ore sorting a shift in mining practices will be
required in order to exploit the natural variability in the deposit rather than blend it out.

IN-PIT OR NOT IN-PIT – THE QUESTION OF SORTER LOCATION

A bulk ore sorter for pre-concentration could be located anywhere from the mining face to the
grinding circuit feed. The best location depends on the particular operation and will be affected by
factors such as:
 279

• mine type and mining methods

• location of waste dumps and processing plant
• operating costs of the mine, sorter and downstream processes
• impact on productivity in the mine and plant
• changes to layout
• space required
• environmental impact
• Greenfield operation or retrofit to an existing operation

In general, as early as possible in the process is best. There is less mixing, preserving the natural
heterogeneity of the ore which is required for effective separation. Additionally, extra costs and lost
capacity associated with treating non-economic material are avoided. The capacity, quality, energy,
and cost benefits of discarding barren material are carried through all the downstream process steps.

In almost all cases, bulk ore sorting would need to be implemented after primary crushing to present
material at a size that can be handled by the sorter (conveyor, sensor and diverter). However, each
additional crushing or material transfer stage incorporated prior to sorting increases mixing, costs and
unnecessary energy consumption, so should be avoided if possible.

Conducting bulk ore sorting at the mining face, either in-pit or underground, allows ore and waste to
be directed to the appropriate destination (waste dump or process) immediately. Thus, minimising
rehandling, mixing and potentially also ore transportation costs depending on the relative positions of
waste dump and process plant. Implementation at the mine face will depend on having suitable space
available for the sorter as well as the mining method employed and potential impact on mining
productivity.

Underground pre-concentration, prior to haulage or hoisting, produces a solid waste underground

which can be combined with tailings and cement from the surface as required and disposed of as fill
(Bamber, 2008). In open pit mines, bulk ore sorting could potentially fit well with in-pit crushing and
conveying systems. The sensor(s) would be located on a conveyor leaving the pit, and a flop gate
used to separate waste and ore onto their respective conveyors. A system equiped with multiple flop
gates (like that shown in Figure 5) could be used to separate different ore types onto separate
stockpiles which could be either blended according to the downstream process requirements or sent to
different process routes.

Sensor(s) WASTE
Flop gates

ROM
Ore Ore 3
Ore
1
2

ORE BLEND
to the process plant

Tunnel feeders

Figure 5 – Sensor and multiple flop gate system for bulk ore sorting.
 280

BENEFITS OF BULK ORE SORTING

By removing coarse barren material, pre-concentration has the potential to significantly reduce the
amount of material that requires downstream processing. If conducted close to the mining face, it can
potentially reduce ore transport requirements by rejecting barren gangue and transporting less ore to
the processing plant. Pre-concentration effectively upgrades the plant feed; less tonnes of ore are
treated in the processing plant per tonne of product, thus reducing the costs, energy and water
consumption per tonne of product.

In existing operations with fixed plant capacity the production rate can be increased after sorting due
to the increase in feed grade. In Greenfields operations the size of downstream processing equipment
can be reduced (reducing the capital and operating costs), or the production rate can be increased.

Pre-concentration can upgrade previously uneconomic material to increase resource utilisation, and
should be factored into the cut-off grade decision. It may be possible to reduce the cut-off grade
depending on the net balance of reduced processing costs (due to gangue rejection) and extra costs
associated with pre-concentration and mining additional material. Additional mining resources (drill
and blast, load and haul, waste handling) may be required to maintain the plant feed rate due to the
rejection of material (previously considered ore) by the sorter. However, this may be compensated for
by the increase in plant feed grade and the subsequent reduction in downstream processing costs,
increased production and greater extraction of valuable mineral from the resource.

Additionally, bulk ore sorting can reduce dilution and ore loss in mining operations by improving
grade control. In some cases, mining costs may be reduced, with the bulk ore sorter providing
selectivity thus allowing less selective mining processes.

The environmental footprint of the mine is also reduced due to lower energy consumption, greenhouse
gas emissions and water losses per tonne of product. Less fine wet tailings are produced requiring a
smaller tailings storage facility and minimising the surface impact. The waste dump area may
increase; however, in some cases the dry coarse waste from the sorter could be useful as aggregate or
for other fill purposes.

Bulk sorting may also be used to separate different ore types to treat separately. High grade material
could be processed through a more costly but higher recovery process due to the high contained value,
while the low grade material is processed through a less costly but lower recovery process. For
example, some operations may benefit from treating high grade material with flotation and low grade
material by heap leach. This means a mine may have a variety of cut-off grades, depending on the
cost of the processing route options ahead. It may also be possible to recover valuable components
from waste dumps and low grade stockpiles (marginal reserves) that would be uneconomic to treat
without upgrading.

The applications and potential benefits of bulk ore sorting are illustrated in Figure 6. Despite the
apparent benefits, uptake has been slow. This is possibly due to perceptions of unacceptable metal
losses, insufficient understanding of ore characteristics, and a lack of understanding of the systemic
impact. Limitations in sensor capabilities, and most importantly the low throughput of the existing
individual particle sorters is also a significant deterrent to application. However, metal losses in pre-
concentration are offset, if not completely compensated for, by the increased recovery in downstream
processes due to the higher feed grade, and the overall economics of the project benefit from reduced
costs and/or increased production. A bulk ore sorting system would be much simpler, less expensive
(in both capital and operating costs) and require far less footprint than the current individual particle
sorting technology.
 281

Improve resource utilisation

The cut-off grade in the mine may be
decreased, and mineral reserves Reduce mining costs
increased. Using selection of ore by bulk ore sorter may require
less selective mining processes reducing costs.
Recover marginal material
Upgrade low grade or waste stockpiles Improve grade control
making them economic to treat Sorter measurement decreases misclassification of
ore and waste, reducing dilution and ore losses.

Produce dry coarse waste Bulk Ore Sorter

Reduce ore transport requirements
Dry coarse waste from the sorter
Reject barren waste and transport less
may be useful as aggregate or
tonnes to the process plant
other fill purposes

Optimise process route

Reduce comminution Different ore types can be separated and
energy Process A Process B processed differently
Gangue minerals are
usually hard. Removing Increase feed grade
these can reduce the
By upgrading the plant feed, less tonnes
hardness of ore to be Power are treated per tonne of product.
treated, increasing the
Thus:
capacity and reducing the
energy consumption of Reduce costs
comminution circuits Water Lower operating costs
Lower capital costs in new plants
Improve Recovery
Reduce environmental footprint Higher feed grade improves grade-
Less tonnes of ore are treated per tonne of recovery performance of some
product reducing the energy consumption, processes
greenhouse
1 gas emissions, water losses and Increase production
fine wet tailings per tonne of product. More product per tonne of ore treated
Tailings Storage Facility
Product

Figure 6 – Applications and potential benefits of bulk ore sorting.

The overall impact on project economics will of course depend on the ore deposit and characteristics,
mining and processing methods, site conditions, local economic climate, etc, and needs to be
evaluated on a case-by-case basis. To demonstrate the potential impact of bulk ore sorting on project
economics, three different cases have been considered, these are:
• Reducing cut-off grade for a Greenfield operation
• Increasing production at an existing operation with fixed plant capacity
• Reducing ore losses and dilution through improved grade control
Each case is discussed in the following sections.

Case 1 – Reducing cut-off grade for a Greenfield operation

Bulk sorting increases the resource utilisation by reducing the cut-off grade. More valuable metal is
extracted while the processing plant treats less tonnes at higher feed grade.

A hypothetical copper deposit is considered with a grade tonnage curve as shown in Figure 7. Mining
and processing cost assumptions were made based on operations at similar copper deposits and used
to conduct some preliminary break-even cut-off grade calculations, as shown in Table 2.

The break-even cut-off grade is the grade at which it becomes economically feasible to mine a block
of material. It is the grade at which enough revenue is generated to pay for the extraction, processing
and selling costs (Hall, 2014). Cut-off grade is used to decide a course of action; i.e. to treat the
material as ore or waste. In this example, the ore-waste break-even (where the cost per tonne of waste
equals the value if treated as ore) has been used to define rock as ore or waste. By only treating
material that is above the cut-off grade, only material that adds value is processed.
 282

Grade x Tonnage Curve

400 0.8
350 0.7
Cumulative Million Tonnes

300 0.6

Average Grade (%)

250 0.5
200 Cumulative million tonnes 0.4
150 0.3
Average grade
100 0.2
50 0.1
0 0.0
0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65 0.70
Cut off Grade (%)

Figure 7 – Grade tonnage curve for a hypothetical copper deposit.

Table 2 – Ore-waste break-even cut-off grade calculations.

Without With
Cut-off grade calculations bulk bulk
sorting sorting
Waste
Mining cost US$/t 1.2 1.2
Processing Cost US$/t 0.3 0.3
Overhead Cost US$/t 0.3 0.3
Total Cost of Waste US$/t 1.8 1.8
Ore
Mining cost US$/t 1.2 1.25 Increase for reject handling
Processing cost US$/t 11.5 8.05 Reduce by the mass rejected from sorter
Overhead Cost US$/t 0.5 0.5
Sorting cost US$/t 0 0.4 Sorter operating costs
Total Cost of ore US$/t 13.2 10.2
Value per block
Cu price US$/t 6000 6000 LME 3 Month futures at 14/1/2015
Smelter cost US$/t 700 700
Sort Recovery % 100 90 Assume 90% recovery in 70% mass
Plant Recovery % 90 90
Smelter recovery % 96 96
Overall recovery % 86.4 77.8
Cut-off grade % 0.25 0.20
Sorting reduces cut-off grade by % 0.05

The base case scenario (without bulk ore sorting) has a cut-off grade of 0.25% copper delivering
110 million tonnes of ore to the processing plant with an average grade of 0.39% copper. With the
inclusion of bulk ore sorting, the cut-off grade is reduced to 0.20%, changing the delineation of the
ore body and pit shell and increasing the amount of above cut-off grade material, as illustrated in
Figure 8. The resulting average grade of mined ore is reduced to 0.35% copper. However, bulk ore
sorting upgrades this to deliver 102 million tonnes of ore with an average feed grade of 0.45% copper
to the processing plant (using conservative estimates of sorter performance in which 90% of the
copper is recovered in 70% of the mass). Overall, almost 8% (8 million tonnes) less ore is treated by
the processing plant, but at a higher feed grade, and with 5% more contained metal than the base case
scenario without bulk ore sorting.
 283

Without Bulk Ore Sorting With Bulk Ore Sorting

Cut-off grade 0.25 % Cut-off grade 0.20 %
Average head grade 0.39 % Average head grade to sorter 0.35 %
Average head grade to plant 0.45 %
Total ore to plant 110 million tonnes Total ore to plant 102 million tonnes
Total Cu to plant 433 thousand tonnes Total Cu to plant 452 thousand tonnes

0.00 0.00 0.10 0.15 0.20 0.30 0.30 0.50 0.20 0.10 0.00 0.00 0.10 0.15 0.20 0.30 0.30 0.50 0.20 0.10
0.00 0.10 0.15 0.20 0.25 0.40 0.70 0.40 0.25 0.10 0.00 0.10 0.15 0.20 0.25 0.40 0.70 0.40 0.25 0.10
0.10 0.10 0.15 0.25 0.40 0.60 0.70 0.30 0.15 0.10 0.10 0.10 0.15 0.25 0.40 0.60 0.70 0.30 0.15 0.10
0.10 0.10 0.20 0.30 0.50 0.60 0.40 0.20 0.10 0.00 0.10 0.10 0.20 0.30 0.50 0.60 0.40 0.20 0.10 0.00
0.00 0.15 0.25 0.30 0.40 0.30 0.20 0.10 0.00 0.00 0.00 0.15 0.25 0.30 0.40 0.30 0.20 0.10 0.00 0.00
0.00 0.10 0.15 0.25 0.20 0.15 0.10 0.00 0.00 0.00 0.00 0.10 0.15 0.25 0.20 0.15 0.10 0.00 0.00 0.00
0.00 0.00 0.10 0.15 0.15 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.10 0.15 0.15 0.00 0.00 0.00 0.00 0.00
Green line indicates the pit shell
Red shading indicates above cut-off grade material

Figure 8 – Impact of bulk ore sorting on reporting pit and material extracted.

The sorter performance and also the quantity and distribution of marginal grade ore within the deposit
would have a significant impact on the results. In practice, the ore body delineation, mine scheduling,
processing options and capacity should be optimised together concurrently with ore sorting options
and cut-off grade selection as part of full mine strategy optimisation.

This simple high level example is only intended to illustrate how bulk ore sorting could be
implemented to reduce cut-off grade and increase resource utilisation based on a hypothetical deposit.
A detailed study is being conducted for a known resource that is currently being mined with
conventional practices and will be reported at a later date.

Case 2 – Increasing production at an existing operation with fixed plant

capacity
Bulk sorting upgrades the plant feed, increasing the production rate. Production and profit is
brought forward in the schedule, and the amounts of ore treated (and associated costs and energy)
are reduced per tonne of product.

This case considers a 200 million tonne copper deposit with a fixed plant capacity of 40,000 tpd,
0.52% copper feed grade and 15 year mine life (based on data from a feasibility study). The inclusion
of bulk ore sorting upgrades the plant feed and thus increases the production rate. A higher mining
rate is required to meet the plant capacity following rejection of waste by the ore sorter, as shown in
the diagram provided in Figure 9.

High level Net Present Value (NPV) analysis suggests that incorporating a bulk ore sorting plant to
upgrade the feed grade to 0.62% copper for this case may improve the NPV over the life of mine, see
Figure 5. The NPV analysis assumes the plant capacity is fixed and the capital and operating costs of
the existing plant remains constant. Additional capital and operating costs are included for the ore
sorting plant and associated equipment (conveyors, sensor, control system, diverter) and mining
equipment for the increased mining rate and additional material handling associated with sorting.

The largest impact on NPV is that the production and profit are brought forward in the schedule, with
mining completed within 12 rather than 15 years. Additionally, the operating cost of the sorter is less
than a quarter of the cost of treating the rejected waste through the processing plant.
 284

At this stage the analysis does not consider that gangue minerals are generally harder than valuable
minerals, and rejecting the gangue could reduce the hardness of the plant feed and further increase the
throughput and reduce energy consumption of the comminution circuits. Likewise, the increased
plant feed grade could deliver improved grade-recovery performance in flotation circuits, but this has
not been incorporated in the current analysis. Therefore, the benefits will potentially be even greater
than estimated in this study.

Base Case:

Plant Feed Process Plant

40,000 tpd ore Capacity
0.52 % Cu 40,000 tpd ore
207 tpd Cu

Plant capacity is fixed

Bulk ore sorting example: i.e. The same for both cases

Mined Ore Ore Sorter New Plant Feed Process Plant

53,333 tpd ore Recovery to Accept 40,000 tpd ore Capacity
0.52 % Cu 90 % Cu 0.62 % Cu 40,000 tpd ore
277 tpd Cu 75 % Mass 250 tpd Cu
higher mining rate
to provide plant feed rate 1.2 Upgrade factor
to match plant capacity Reject Plant feed is upgraded due
13,333 tpd ore to rejection from sorter
0.21 % Cu
28 tpd Cu
extra mining equipment
is required for reject handling

Assumptions / Inclusions:
Assume plant capacity is fixed and existing plant capex and opex constant
Additional opex is included for ore sorting plant equipment (conveyors, sensor, control, diverter)
Additional opex is included for reject handling (removal and stockpiling of rejected material)
Additional capex is included for the increased mining rate and reject handling
Additional capex is included for ore sorting plant including conveyors, sensor, control system, diverter

Figure 9 – Schematic of bulk sorting with fixed plant capacity.

Cumulative NPV
2000
1438
1500
1359
NPV (Million US$)

1000

500

-500

-1000
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Year
Base Case 15 year LOM With Sorting (nominal)

Figure 10 – Impact of bulk ore sorting on NPV over life of mine (fixed plant capacity).

Sensitivity analysis indicated the most significant factor affecting the NPV results was the sorter
upgrade factor, as shown in Figure 11. The upgrade achieved by the sensor is influenced by the
 285

variability in the ore as the sorter receives it and the efficiency of the sorter. Therefore, it is very
important to understand the ore characteristics, variability at the measurement point and sorter
performance.

Bulk Ore Sorting - Sensitivity Analysis

25%
Percent Change in NPV from No Sorting Scenario

1.3 Upgrade Factor

20%

15%
0.2 $/t ore
10% 0.6 $/t reject
0.6 $/t ore
$75 M USD
5% $49 M USD 1.0 $/t reject

Nominal
0%
Sorting Case
1.2 Upgrade
-5% negative impact $62 M USD
on project NPV 0.4 $/t ore
-10% 0.8 $/t reject

1.1 Upgrade Factor

-15%
-60% -40% -20% 0% 20% 40% 60%
Percent change in Variable (sorter upgrade, opex, capex)
Sorter Upgrade Opex Capex

Figure 11 – Sensitivity analysis of the impact of ore sorting on NPV.

Case 3 – Reducing ore losses and dilution through improved grade control
Measuring grade with a bulk ore sorter enables correct allocation of ore and waste, reducing dilution
and ore losses.

A mining block model and selection of material to be mined is based on grades determined using
kriging (interpolation), which introduces errors. The actual grade is not known, and is often not
measured before material is classified as ore or waste. This results in dilution and ore losses due to
mis-classification. Blast movement, internal waste mixed with ore, inaccuracies in ore and waste
boundaries, large scale (non-selective) mining equipment and logistical errors also contribute to
dilution and ore loss. Actions can be taken to reduce dilution and ore loss, but it cannot be completely
avoided (Ebrahimi, 2013), see Figure 12.

Dilution can have a very significant impact on the profitability of mining operations. Dilution
decreases the plant feed grade. Operating costs are increased, as a higher tonnage is treated by the
processing plant per tonne of product, and the processing plant capacity is effectively decreased,
prolonging the mine life. Lower plant feed grade also reduces plant recovery in many cases. Dilution
also increases the cut-off grade which reduces the overall resource utilisation. Ore losses occur when
ore is misclassified as waste and not processed resulting in lost revenue.
 286

Figure 12 – Schematic illustration of dilution and ore losses.

The amount of dilution and ore losses is typically not well understood as it is difficult to quantify. It
is likely that it is underestimated at many operations. At Somincor, an underground copper and zinc
mine in Southern Portugal, the SmartTagTM ore tracking system is used to track ore and waste. It was
found that around 20% of waste was treated as ore and a similar amount of ore reported as waste prior
to improved grade control practices using the SmartTagTM system (Wortley, 2013; personal
communication).

Reducing dilution and ore losses represents a significant opportunity to increase profitability. For
example, Prati (2014, personal communication) estimated US$29 million is lost per annum due to
dilution of feed grade for an 8 Mt/y copper mine. The lost value in ore misclassified as waste is
potentially much greater, as was demonstrated by the Somincor example.

A bulk ore sorter can reduce both dilution and ore loss by improving grade control. Measuring grade
with the sorter will allow correct allocation of ore and waste. Only above cut-off grade material is
sent to processing, improving the execution of the theoretical cut-off grade. Likewise, above cut-off
grade material (and revenue) is not lost to waste dumps. Ore sorting may also reduce (but not
eliminate) the quality control requirements at bench level. This has potential to increase mine
productivity and reduce costs, particularly in deposits with complex mine geology and when mining
near ore-waste boundaries.

CONCLUSIONS
The purpose of bulk ore sorting is to remove coarse barren material before energy and cost intensive
downstream processing and handling. Fewer tonnes of ore are treated per tonne of product, reducing
the costs, energy and water consumption per tonne of product. Thus, bulk ore sorting has the
potential to improve the profitability and reduce the environmental impact of a mining operation.

Bulk ore sorting may be used to increase the production rate through fixed plant capacity, or reduce
the required size of downstream processing equipment. Uneconomic or marginal reserves in the
deposit, waste or low grade stockpiles may be upgraded making them economic to treat and
improving the resource utilisation. Uneconomic deposits may become viable due to the upgrade and
increased productivity following bulk ore sorting. Additionally, bulk ore sorting can reduce dilution
and ore loss in mining operations by improving grade control, and could also be used to separate
different ore types to treat separately. An efficient ore sorter allows selectivity on the basis of a
smaller volume; for example, meters of material on a conveyor belt rather than by truck load.

The key driver affecting the economics of bulk ore sorting is the upgrade that can be achieved by the
sorter (the recovery and mass rejection). Essentially, how effectively the sorter can remove the barren
material while recovering the valuable components. This is a function of the ore characteristics,
 287

distribution of grade throughout the deposit, and the efficiency of the sorter, and requires extensive
testing and evaluation on a case by case basis. The impact on the overall operating capital and
operating costs for the operation also needs to be considered. It is likely that, at least in some cases,
the additional costs associated with bulk ore sorting will be outweighed by the reduction in
downstream processing costs and/or increased production.

REFERENCES
Bamber, A, 2008. Integrated mining, pre-concentration and waste disposal systems for the increased
sustainability of hard rock metal mining, PhD thesis, University of British Columbia, Vancouver.

Bergmann, C, 2009. Developments in ore sorting technology [online], Council for Mineral
Technology. http://www.mintek.co.za/Mintek75/Proceedings/M02-Bergmann.pdf [Accessed: 19th
January 2015].

Hall, B, 2014. Cut-off Grades and Optimising the Strategic Mine Plan, 311 p (The Australian Institute
of Mining and Metallurgy: Melbourne).

Manouchehri, H, R, 2004. Sorting: possibilities, limitations and future

[online]. http://pure.ltu.se/portal/files/299975/article [Accessed: 19th January 2015].

Miljak, D. 2011, Ore sorting provides sustainable future for mining, CSIRO.

Pokrajcic, Z, Lewis-Gray, E, 2010. Advanced comminution circuit design – essential for industry,
AusIMM Bulletin, August 2010, pp 38 – 42.

Ebrahimi, A, 2013. The importance of dilution factor for open pit mining projects [online]. Available
from: http://www.srk.com.au/files/File/papers/dilution_factor_openpit_a_ebrahimi.pdf [Accessed: 6th
January 2015].

Heselev, T, 2012. Ore sorting: Sound technology making waves [online], Resourceful. Available
from: http://www.csiro.au/Portals/Publications/Magazines/resourceful/Issue-2/2-sound-
technology.aspx [Accessed: 13th January 2015].

Wotruba, H, 2006. Sensor sorting technology – is the minerals industry missing a chance? in
Proceedings of XXIII International Mineral Processing Congress (IMPC), Istanbul, Turkey, 3-8
September 2006.

Butcher, A, Cunningham, R, Edwards, K, Lye, A, Simmons, J, Stegman, C, Wyllie, A, 2013.

Northparkes Mines, Rio Tinto, in Monograph 28 – Australasian Mining and Metallurgical Operating
Practices – Third Edition (ed: W J Rankin), pp 861 – 878 (The Australasian Institute of Mining and
Metallurgy: Melbourne).

View publication stats