Minerals Engineering 34 (2012) 1–10

Contents lists available at SciVerse ScienceDirect

Minerals Engineering
journal homepage: www.elsevier.com/locate/mineng

Flash flotation. . . and the plight of the coarse particle

Bianca Newcombe ⇑, D. Bradshaw, E. Wightman
JKMRC, The University of Queensland, 40 Isles Rd., Indooroopilly, QLD 4068, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Coarse particles require distinctly different conditions to their fine and intermediate counterparts for suc-
Received 16 June 2011 cessful flotation and recovery to the concentrate launder. These range from simple operational require-
Accepted 22 March 2012 ments such as shallow froth depth, reduced impeller speed and higher collector dosage to those that must
Available online 15 May 2012
be optimised specifically for the coarser size fractions such as air addition rate and bubble size, as well as
the chemical environment (pH). This paper is the first of a series of publications on the topic of flash flo-
Keywords: tation and reviews many of the factors that affect coarse particle flotation with a view to how they impact
Flotation machines
the flash flotation process. A review of the current state of knowledge of the flash flotation process is pre-
Froth flotation
Particle size
sented and raises a number of issues in regard to both current operational knowledge and modelling
Sulphide ores practices.
Classification Ó 2012 Elsevier Ltd. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2. Flash flotation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
3. Process description. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
4. Factors affecting flash flotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
4.1. Particle size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
4.2. The role of the chemical environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
4.3. The role of detachment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
4.4. Coarse particles and the froth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
4.4.1. Froth stability and structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
4.4.2. Froth depth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
4.4.3. Bubble size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
4.4.4. Aeration rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
4.5. Hydrodynamics and slurry conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
4.6. Final remarks on coarse particle flotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
5. Predicting flash flotation performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
6. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

1. Introduction
From the perspective of a plant metallurgist or operator, the
role of a flash flotation cell in a concentrator is to remove any
liberated valuable mineral and/or any coarse composite particles
⇑ Corresponding author.
E-mail addresses: b.newcombe@uq.edu.au (B. Newcombe), d.bradshaw@uq.e-
du.au (D. Bradshaw), e.wightman@uq.edu.au (E. Wightman).
rich in the target valuable mineral; and this has also been the long
held belief of the authors. The single factor that is believed to dis-
tinguish a flash flotation cell from any other type of flotation device
is the size of the particles it recovers (i.e. coarse particles), and this
is repeatedly cited in the literature (Lynch et al., 2010; Lamberg
and Bernal, 2009; Yan et al., 2005; Mackinnon et al., 2003;
Sandström and Jönsson, 1988).
The role of particle size in the flotation response of an ore has
long been recognised as a key parameter affecting overall

0892-6875/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.mineng.2012.03.023
2 B. Newcombe et al. / Minerals Engineering 34 (2012) 1–10
performance. Considerable research has gone into this area, but
until very recently has often been limited to the fine and interme-
diate size classes, with work frequently performed using single
minerals or fabricated mineral mixtures in idealised laboratory
settings. The study of unliberated coarse particles in an industrial
setting, as would be experienced in the feed to a flash flotation cir-
cuit, has been somewhat neglected, as the nature of the particles
presents one of the most difficult areas of study in this field. In this
series of papers, the authors investigate both the nature and
behaviour of the particles within a flash flotation system and also
the processes occurring within an industrial flash flotation cell.
This paper has been written to provide a comprehensive review
of the flash flotation process and the available information on the
coarse (mostly unliberated) particles that are its typical feed mate-
rial. Subsequent papers will present a comparison of the results of
laboratory tests with the actual measured plant flash flotation cell
performance of the same refractory gold ore, and demonstrate how
batch flotation tests can be used in conjunction with mineralogical
analysis to predict the amenability of an ore to the flash flotation
process. This will be followed by an investigation into the sub-pro-
cesses that are occurring within an operating flash flotation cell
with a view to develop a flash flotation specific model.
2. Flash flotation
The concept of floating coarse, potentially unliberated mineral
particles within the grinding circuit is not modern in inception,
but takes its origins from preliminary investigations conducted in
the 1930s. The consequent development of ‘Maxwell’ or ‘Denver’
‘unit cells’ was the precursor to what we now know as ‘flash’ flota-
tion (Lynch et al., 2010).
The first modern flash flotation cell installed in an industrial
concentrator occurred in June 1982 at the Hammaslahti concentra-
tor in Finland (Bourke, 1995). Since then the use of this unit oper-
ation has increased dramatically, both in greenfields plants and as
retrofits to existing concentrators. Consideration of flash flotation
in any new flotation plant is readily accepted by the industry, with
cells capable of processing 1800 tph being installed in new opera-
tions (Wade, 2006). Yet despite this widespread use for almost
30 years, there is very little technical information on its perfor-
mance and no accurate flotation model specifically for flash flota-
tion available in the literature.
Prior to the success of the flash flotation circuit, the most com-
mon methods for accommodating the different processing require-
ments of different size classes involved either split conditioning of
the flotation feed (where the size classes are separated, fed into
two separate conditioning vessels then recombined for flotation)
or split flotation (where the size classes are separated and fed into
two separate flotation circuits) (Alford and Clarke, 2007; Senior
et al., 1994; Crawford and Ralston, 1988; Jameson, 1984; Trahar,
1976). Both of these methods involve considerable capital outlay
and increased operating costs, and the complexity of the conse-
quent flotation circuits makes them challenging to operate effec-
tively. The use of a flash flotation cell reduced the cost and
complexity considerably not only by having a small footprint in
the plant but also by being capable of treating large volumes of
material in a single cell.
The primary purpose of a flash flotation cell is to remove valu-
able minerals from the grinding circuit, preventing their over-
grinding or sliming. Due to the higher SG of sulphide and precious
metals, they tend to accumulate in the cyclone underflow stream
and are liable to be ground to the very fine sizes required to make
them light enough to report to the cyclone overflow stream. This
would potentially incur recovery losses in the plant as fine parti-
cles are notoriously difficult to recover by conventional flotation
(Mulleneers et al., 2002; Yoon, 2000; McIvor and Finch, 1991;
Sandström and Jönsson, 1988; Trahar and Warren, 1976). The
treatment of the cyclone underflow stream via flash flotation can
remove these valuable particles before they become too fine. Treat-
ing the cyclone underflow stream also has the benefit that the fine
particle sizes have been removed, allowing a much higher concen-
trate grade to be achieved by minimising the entrainment of un-
wanted gangue fines (Sandström and Jönsson, 1988; Kallioinen
and Nitti, 1985).
McIvor and Finch (1991) suggested that flash flotation cells may
be an appropriate addition to circuits where different target miner-
als (e.g. galena and sphalerite) can be separated where size (liber-
ation characteristics) is a distinguishing factor; i.e. where the
liberation size of galena for example is much coarser than that of
sphalerite, the galena could be floated in the grinding circuit,
allowing the sphalerite to undergo further grinding and subse-
quent flotation in a conventional circuit.
The contribution of a flash flotation cell to the overall perfor-
mance of the plant has not been extensively studied, however
the work of Sandström and Jönsson (1988) provided an excellent
example of operating data from a number of different plants, with
and without a flash flotation cell; illustrating that the cell is capa-
ble of producing a saleable concentrate in a single step and in-
creased the plants overall recovery (Cu–Au ore). Their work also
showed that the size range of particles recovered was extended
from a top size of 125 lm to greater than 250 lm when the flash
flotation cell was in use.
3. Process description
A schematic depiction of a flash flotation cell and where it is lo-
cated relative to both grinding and conventional flotation opera-
tions is given in Fig. 1. Slurry feed to the cell is from the cyclone
underflow stream and consists of a mixture of both small rocks
and coarse sand-like material with water; typically this stream
would be between 60% and 80% solids. In order to allow for effec-
tive flotation in this environment water is added to the feed
stream, and reagents are employed. Reagent addition is typically
done in one of two ways: all reagents are added simultaneously
to the feed well of the cell; or either an activator and/or a collector
is added to the grinding circuit, with all remaining reagents being
added to the feed well. The method that is employed is specific to
the ore being treated.
Where the feed material is very coarse, the cell conducts sepa-
ration in two ways: firstly, as a classifier, allowing heavy coarse
material and rocks to flow straight down to the bottom discharge
point, effectively bypassing the flotation area of the cell; and sec-
ondly as a flotation cell, conducting a flotation process on the light-
er, finer material present in the feed. This is depicted in Fig. 2. Air is
added through the impeller shaft to form the bubbles that trans-
port the valuable hydrophobic material to the concentrate launder.
The effect of this dual action is that the bottom discharge point has
a much higher per cent solids and coarser size distribution than
material overflowing the concentrate launder. This allows the tail-
ings stream to be sent directly to a secondary mill, whilst the
concentrate stream is suitable for either further cleaning in subse-
quent flotation stages or may be of sufficient quality to be consid-
ered as final concentrate. Outotec flash flotation cells are designed
specifically to act as both a classifier and flotation machine (Cole-
man, 2011).
In terms of operability, it is the experience of the authors that
water management is the greatest challenge. A fine balance exists
between the requirements of the flotation cell and those of the
subsequent grinding operations; too little water impedes flotation
performance, while too much water can be detrimental to commi-
nution operations.
 B. Newcombe et al. / Minerals Engineering 34 (2012) 1–10 3

Cyclone Overflow
Conventional Flotation
Feed
Cyclone

Water Reagents
Underflow

Flash Feed
Concentrate Box
Flash
Cell

Flash
Primary Secondary Tail
Mill Feed Mill Mill

Optional
Reagents

Fig. 1. Location of flash flotation in a typical grinding/flotation plant.

recovery starts to decline, and this point is often system specific.

A typical value for coarse particles is often +75 lm, however, as
pointed out by Trahar (1981) it may be as high as +150 lm (for a
pyritic ore). An example of the differing size-recovery relationships
Concentrate of different mineral systems is given in Fig. 3, which illustrates that
the point at which recovery starts to decline is different for the var-
ious minerals studied, and will be dictated by both the chemical
Flotation Zone
within cell
and physical conditions present within a given system. In most
flotation plants particles above 150 lm are often considered too
coarse for conventional flotation, and it is a widely held belief
Very coarse
within the industry that the use of a flash flotation cell will extend
particles and
rocks by-pass the the size range of recoverable material. Whilst there is very little
flotation zone evidence available in the literature to support this, the work of
Van der Spuy and Ross (1991) demonstrated that sulphide flotation
can indeed be achieved at the industrial level with particles up to
500 lm, albeit with a specialised flotation cell, purpose built for
the specific task (Deister Flotaire cell). The significance of this
Fig. 2. Schematic of flash flotation cell operation. finding is that very coarse particles may be being recovered/or

There are a number of distinct advantages to operating a flash

flotation circuit. The concentrate produced by this unit will have
a distinctly coarser size distribution, which is known to be benefi-
cial to downstream process such as thickening and filtration
(Sandström and Jönsson, 1988). The removal of coarse liberated
or rich composite valuable particles at this stage reduces the po-
tential for over-grinding and sliming of these particles, which,
due to their higher SG often accumulate in the cyclone underflow
stream. If they are not removed as early as possible there is poten-
tial that they will continue to be ground to the point where they
are so fine that they will not be recovered effectively in a conven-
tional flotation circuit. Therefore, the removal of valuable particles
as early as possible has the potential to improve the overall recov-
ery in flotation circuits (Lamberg and Bernal, 2009; Yan et al.,
2005; Mackinnon et al., 2003; Laplante and Dunne, 2002).

4. Factors affecting flash flotation

4.1. Particle size

The exact size at which the intermediate size class ends and the
coarse classification starts is often defined as the size at which Fig. 3. Size-recovery response of different mineral systems (after Jowett, 1979).
4 B. Newcombe et al. / Minerals Engineering 34 (2012) 1–10
may have the potential to be recovered in flash flotation circuits
and work encompassing an extended size range beyond the typical
upper limit of 150 lm is warranted. A further limitation within the
current body of literature is that most studies have generally been
conducted on either single mineral or synthetic mineral mixtures
in laboratory systems, far removed from the industrial plant envi-
ronment. An example of this is the work of Drzymala (1994); in
which flotation of pyrite particles as coarse as 1.2 mm is achieved
in the quiescent conditions of the monobubble Halimond tube;
demonstrating that sulphides can be recovered at very coarse sizes,
albeit under highly controlled conditions. One study of significance
was that of Sutherland (1989), who looked at the behaviour of
chalcopyrite particles of varying size and liberation in a laboratory
system. This study found that coarse liberated material performed
in a similar way to intermediate composite particles and that a
coarse particle of the same composition as an intermediate particle
would float more slowly.
The liberation characteristics of particles play a significant role
in their treatment requirements for flotation and this has been an
area of considerable research input and is covered extensively else-
where (Runge, 2009; Suthers et al., 2004; Vianna, 2004; Bazin et al.,
1994; Sutherland, 1989). A key question in terms of a flash flota-
tion circuit is the extent to which liberation influences perfor-
mance. Does a flash flotation cell recover coarse composite
material, as is the widely held belief, or is it limited to recovering
well liberated fast floating material due to operational factors such
as low residence time and mass-pulling capacity. Further study is
warranted in order to determine the both the nature and range
of particles that are recovered by the flash flotation process, and
this will be covered in a subsequent paper.
4.2. The role of the chemical environment
In the majority of flash flotation systems there is very little con-
ditioning time available before particles reach the ‘flotation zone’
of the cell, with reagents typically being added into the cells feed
well or feed pipe. This means that the conditions present in the
grinding stage prior to flash flotation will have a significant impact
on performance and consequently the reagents used in flash flota-
tion must be carefully selected to reflect this minimal contact time
before flotation occurs.
Adjustment of flotation conditions, such as reagent addition
rate and pH have a much greater impact on coarse particle flotation
than any other size range (Trahar, 1981) and specialty reagents are
often required to extend the upper size limit of flotation (Yoon
et al., 2002). An example which highlights the impact of pH on
the coarse fractions is given by Senior et al. (1994), who studied
the size response of a synthetic mixture of pentlandite and quartz
to changes in pH levels. The results are presented in Fig. 4 and
clearly show that the coarsest fractions are most affected by the
change in pH conditions. Whilst the study of size based response
of a mineral to changing pH has been limited, the general finding
that there exists an optimum pH for flotation within a given sys-
tem has been found elsewhere (Vieira and Peres, 2007; Akerman
et al., 1987; Pritzker et al., 1985). In this particular example, an in-
crease in pH clearly causes substantial degradation of flotation
response, most significantly in the coarsest fractions, and whilst
there are many factors within a system that will be affected by
pH (froth properties, chemical solution properties) it is a change
in the surface chemistry of the mineral particle (hydrophobicity)
that may be of most significance to the coarse particles.
Collector concentration (hydrophobicity) has been shown to
impact on the performance of coarse particles within a system,
with numerous authors pointing to the higher collector addition
requirements of coarse particles (Gontijo et al., 2007; Bravo et al.,
2005; Vianna, 2004; Bazin and Proulx, 2001; Senior et al., 1994;
Fig. 4. Recovery by size of pentlandite with varying pH conditions (after Senior
et al., 1994).
Trahar, 1976). Crawford and Ralston (1988) found that there is a
threshold minimum level of hydrophobicity for each size class, be-
low which flotation will not occur and that this level is highest for
the coarsest particles (121 lm). Increasing the hydrophobicity be-
yond this threshold value incites a strong flotation response in
both the intermediate and coarse fractions, however the fines are
relatively unaffected. A similar result was obtained by Senior
et al. (1994).
A detailed study on the surface chemistry of various sized par-
ticles in a Pb/Zn/Ag ore flotation system was conducted by Vianna
(2004) which showed that coarse galena particles (+150 lm) re-
quire much larger surface coverage of collector than the intermedi-
ate size fraction. Increasing the reagent addition found that a
minimum addition rate of collector existed which must be reached
or exceeded to initiate meaningful levels of coarse particle flota-
tion. A similar relationship with surface hydrophobicity and parti-
cle size was also found by Gontijo et al. (2007) for quartz particles.
Bazin and Proulx (2001) found that utilising staged reagent addi-
tion was of particular benefit to the coarse particles within a
system.
One drawback of using increased concentrations of collector to
improve the flotation of coarse valuables was highlighted by Tra-
har (1976); in the process of improving the coarse particle recovery
of galena, the increased collector addition also caused an unwanted
increase in the recovery of other minerals (sphalerite) in the inter-
mediate size fractions. This work highlighted the fact that ‘the
selectivity between two minerals is greatly influenced by the relative
floatabilities of the coarse fractions of the more floatable mineral
and of the intermediate fractions of the less floatable mineral’ (Trahar,
1976). A similar result was obtained by Senior et al. (1994) who
improved the coarse particle recovery of the target mineral (pent-
landite) with both an increased collector addition and again with
an increased activator addition. However in both cases, this was
at the expense of selectivity, with the recovery of the intermediate
fractions of the unwanted mineral (pyrrhotite) also being signifi-
cantly improved.
In most flotation systems where coarse, intermediate and fine
particles are all mixed together, the effectiveness of higher collec-
tor addition to increase coarse particle recovery has the additional
drawback in that most of the extra collector will be consumed by
the fine particles due to their higher specific surface area (Vieira
and Peres, 2007). As the fines in the feed to a flash flotation cell
have been removed by the cyclones, this situation is avoided and
 B. Newcombe et al. / Minerals Engineering 34 (2012) 1–10
Fig. 5. The influence of concentration and stage addition on the rate (K) and over-all recovery (R) of a copper sulphide ore (after Klimpel and Isherwood, 1991).
µ
Fig. 6. Froth recovery of attached galena by size fraction down a rougher bank
(after Savassi, 1998).
targeting reagents to the coarse particles can be more effectively
achieved.
Another good example for showing the different reagent
requirements of coarse particles is to consider frothers. Many con-
centrators utilise a different frother in their flash flotation circuit,
to that used in the conventional part of the plant. Both the addition
rate and structure of the frother molecule affect the size range
recoverable by flotation. Research by Klimpel and Isherwood
(1991) provides an excellent example that shows as addition rate
(of an alcohol frother) increases, the effective size range recover-
able becomes coarser, and furthermore, if 2 stage addition is em-
ployed an even greater size range of effectiveness is observed,
this is illustrated in Fig. 5. Branching of the carbon chain in alcohol
based frothers has been shown to reduce the recoverable particle
size range, whereas if the length of a polyglycols chain is increased,
the upper limit of the particle size recoverable is increased (Klim-
pel and Isherwood, 1991).
4.3. The role of detachment
The factors that influence the detachment of a particle from a
bubble (turbulence in the collection zone, longer induction time
5
Where: Rpfi – Pulp-froth interface recovery
CP – Chalcopyrite
Gal – Galena
Sphal – Sphalerite
Py - Pyrite
NSG – Non-sulphide gangue
Fig. 7. Size by mineral recovery at the pulp-froth interface for a Cu/Pb/Zn ore in a
3 m3 cell (after Seaman et al., 2006).
or decreased buoyancy of the particle bubble aggregate) may be
of most significance to particles in the coarse size fractions (Yoon
et al., 2002; Bazin and Proulx, 2001; Yoon, 2000; Feng and Aldrich,
1999; Jameson, 1984; Schulze, 1977). Both Crawford and Ralston
(1988) and Gontijo et al. (2007) found that there is a kinetic limit
to the size of particles which can be successfully floated in a given
system and detachment of particles will occur when the particle
kinetic energy is equal to or greater than a critical size, determined
by system parameters including turbulence, bubble size and bub-
ble loading. The detachment energy is determined by a balance
of the forces acting upon the particle and incorporates gravita-
tional, acceleration and buoyancy forces as well as hydrostatic
and capillary pressures (Crawford and Ralston, 1988; Schulze,
1977).
If detachment occurs in the froth phase, the detached particle
will either drain back into the pulp, become re-attached to another
bubble or become entrapped within the froth. If the particle is
coarse in size and/or dense it is more likely to flow back to the
6 B. Newcombe et al. / Minerals Engineering 34 (2012) 1–10
pulp, making particle drainage from the froth a selective process
(Seaman et al., 2006; Ross, 1991b). This theory is demonstrated
by Savassi (1998) in an industrial rougher bank treating a galena
ore. Fig. 6 shows a clear decrease in froth recovery with increasing
particle size; similar findings have been reported elsewhere
(Seaman et al., 2006; Vera et al., 1999).
The flow of coarse particles from the froth into the concentrate
launder can be hindered by the thinning of the films surrounding
the bubbles; when the condition is met that the films are the same
thickness as the particles, flow back to the pulp will cease. This in
turn may increase the percentage solids within the froth and re-
duce froth mobility. Contrary to this, the flow back of fine particles
into the pulp is seemingly unaffected by the thickness of the bub-
ble films (Ross, 1991c).
Seaman et al. (2006) also showed that the recovery of particles
at the pulp-froth interface decreases with increasing particle size,
which may be a function of poorer attachment due to lower hydro-
phobicity (possibly as a function of particle liberation/mineralogi-
cal effects). The pulp-froth interface recovery of minerals with
respect to size of the system studied in their work is presented
in Fig. 7. This behaviour is expected to be system specific.
4.4. Coarse particles and the froth
4.4.1. Froth stability and structure
Factors such as reagent chemistry, cell hydrodynamics, particle
size and composition as well as basic operational variables such as
air flow rate, froth depth and pulp density all contribute to the con-
sequent flotation froth (Tao et al., 2000). In the industrial environ-
ment the stability and structure of the froth is often an excellent
performance indicator; where a froth is visibly unstable both grade
and recovery are generally suffering (Farrokhpay, 2011; Barbian
et al., 2006, 2003; Tsatouhas et al., 2006; Klimpel and Isherwood,
1991).
In an industrial concentrator it is typically the case that both the
concentrate grade and unit recovery will decrease down the bank;
and this apparent decrease in performance may be due to a change
in froth stability (Tsatouhas et al., 2006). This may be the result of
the progressive removal of fast floating material, leaving the
slower floating less hydrophobic composite particles, as well as a
lower solids loading in the froth due to the continual removal of
floatable material (Tsatouhas et al., 2006).
Froth stability has been shown to be significantly influenced by
particle size, with an optimum size range existing for both stability
and instability under various conditions (Tao et al., 2000; Feng and
Aldrich, 1999; Johansson and Pugh, 1992; Subrahmanyam and
Forssberg, 1988; Cutting, 1989; Dippenaar, 1981a, 1981b). The
complexities of understanding the role of particles in the froth
has been highlighted by Ross (1991c) and Dippenaar (1981a) in
that under one set of conditions particles of a particular size may
stabilise the froth, while under a different set of conditions they
can cause destabilisation of the froth. Hunter et al. (2008) observed
that the coarser, non-uniform, hydrophobic particles encountered
in flotation systems are excellent foam breakers. For a particle to
rupture the liquid film between bubbles (hence causing detach-
ment and instability), the film must thin to a point related to the
size of that particle; the larger the particle, the quicker that point
will be reached (Dippenaar, 1981b). Jameson (1984) commented
that the effect of a particle on froth stability appears to be a func-
tion of size, with fine particles significantly influencing froth stabil-
ity, and while this may be the case in most situations, the stable
froths achieved in flash flotation cells (where the fines have been
removed by cycloning) are evidence that significant proportions
of fines are not a requirement of a stable froth. Flash flotation
froths are observed to be visually similar to those of conventional
rougher cells.
Johansson and Pugh (1992) used a system consisting of fine par-
ticles to demonstrate that if a certain level of particle hydrophobic-
ity was exceeded, the particles would penetrate the liquid air
barrier, and the resultant film rupture would cause the froth to col-
lapse; however when coarser size fractions were used the same
trends were not observed. Subrahmanyam and Forssberg (1988)
found that a much greater mass of coarse particles are required
to cause froth instability than fine particles of the same type, while
Tao et al. (2000) showed that coarse coal particles destabilise the
froth at lower pulp densities, but stabilise the froth at higher pulp
densities. Feng and Aldrich (1999) found that coarse particles re-
sulted in a decrease in froth loading and a corresponding decrease
in froth stability, while Van Deventer et al. (2002) used very coarse
hydrophobic particles +2.8 3.35 mm as foam breakers in a re-
verse flotation process.
Ross (1990, 1991b) demonstrated that the degree of wetness in
the froth can also impact on the recovery of particles of different
sizes. In drier froths (such as cleaner froths) coarse particles
( 150 lm) behave independently of water and tend not to drain
freely, whilst in wet froths (such as rougher froths) coarse particles
behave similarly to the flow of water within the froth. If the froth is
very dry, coarser particles may form ‘rafts’ on the froth surface,
which (if not quickly transported to the concentrate launder) will
cause a localised collapse of the froth (Ross, 1997; Dippenaar,
1981a).
4.4.2. Froth depth
Ross (1990) used data from a pyrite flotation plant to analyse
the behaviour of particles of various sizes in the froth. The model
developed can be used to predict the concentration of the various
size fractions (up to +212 lm) at various heights within the froth
phase. This work showed that the recovery of the coarsest
(+212 lm, +150/ 212 lm) fractions decreased with increasing
froth height, and that finer particles were more easily recovered
because they were less influenced by bubble breakage, this is de-
picted in Fig. 8. This work also demonstrated that froth height
can have a significant effect on the recovery of the coarser size frac-
tions, as the froth height in a rougher cell increased, the bubble
films became thinner (due to drainage) and consequently unable
to support the larger particles, whose concentration decreased pro-
portionally with particle size (+212 lm particles were lost at 7 cm
froth height, +150 lm particles were lost at 12 cm froth height,
etc.). Kallioinen and Heiskanen (1993) also found that coarse
particles require a shallower froth to be effectively recovered. In
Fig. 8. Variations in concentration of pyrite by particle size with froth height (after
Ross, 1990).
 B. Newcombe et al. / Minerals Engineering 34 (2012) 1–10
Fig. 9. Zones of best flotation performance for particles of different sizes (after
Johnson, 2006).
a later paper Ross (1991a) postulated that the probability of de-
tached coarse particles becoming entrapped within the froth with
increasing froth height will increase as a result of the thinning bub-
ble films. In the experience of the author, industrial flash flotation
cells are typically run with a shallow froth. From an operating
viewpoint, this is done to maximise unit recovery, but is also in line
with these findings from the literature: that a shallow froth is more
beneficial for coarser particle recovery.
4.4.3. Bubble size
Bubble size must be considered with numerous authors point-
ing to a larger bubble size requirement for coarser particles, this
is illustrated in Fig. 9 (Johnson, 2006; Feng and Aldrich, 1999;
Ahmed and Jameson, 1985; Trahar and Warren, 1976). This
appears to be simply due to the fact that coarser particles require
larger bubbles (greater buoyancy) to lift them (Jameson, 1984).
Schulze (1977) demonstrated that the buoyancy of the particle-
bubble aggregate plays a significant role in determining whether
recovery to the concentrate launder is achieved, and that in some
systems this may be the decisive factor. Contrary to this, a study
by Gontijo et al. (2007) postulated that smaller bubbles could im-
prove coarse particle flotation due to their reduced kinetic energy.
Yoon (2000) demonstrated mathematically that the probability of
a particle attaching (Pa) to a bubble is a function of both bubble size
and particle size, as the particle size decreases Pa increases, and as
the bubble size decreases Pa increases to a point at which the bub-
bles become too small for flotation.
Yoon (2000) showed that the level of agitation also relates to
the bubble size, with strong agitation being detrimental where fine
bubbles are used and more beneficial where larger bubbles are
used. In the flash flotation environment, where the level of agita-
tion must be high enough to keep coarse solids in suspension,
the general rule of thumb still remains the best guide; that larger
bubbles are required for coarse particles, and small bubbles should
be used for finer particles.
4.4.4. Aeration rate
The aeration rate (or air addition rate) to cells also has a signif-
icant effect on recovery (both in the pulp and froth phases). As the
air addition rate increases so does the gas hold-up within the cell.
Ross (1997) showed that coarser chalcopyrite particles have a
much lower recovery than intermediates or fines at an equivalent
gas holdup. This indicates that higher aeration rates would be re-
quired to achieve meaningful recoveries in the coarser fractions;
however Kallioinen and Heiskanen (1993), with a pentlandite
ore, showed that whilst increasing the air addition rate will in-
crease coarse particle recovery, a maximum will be reached
7
beyond which recovery will fall. Fine particles are seemingly unaf-
fected by excessive air addition rates (Kallioinen and Heiskanen,
1993). A similar observation can be made from the work of Aktas
et al. (2008), where the maximum froth height obtained with a
mixed sulphide ore (PGM) increased with increasing airflow rate
for finer particles, whilst with the coarser particles, a peak in
maximum froth height was observed at an intermediate air rate.
4.5. Hydrodynamics and slurry conditions
The cell hydrodynamic conditions for flotation of coarse parti-
cles are significantly different to those of finer sizes. Schulze
(1977) demonstrated that the levels of turbulence within the cell
maybe the primary deterministic factor for the successful recovery
of a particle to the concentrate launder. As such, operational factors
such as impeller speed must be altered to meet the specific
requirements of coarser particles, with some authors stating that
a slower speed (or lower power input to the slurry) is recom-
mended for coarse particle flotation (whilst still exceeding the
minimum level of agitation for suspension of solids) (Grano,
2006; Schubert, 2008; Deglon, 2005; Kallioinen and Heiskanen,
1993; Ahmed and Jameson, 1985) and others recommending a fas-
ter impeller speed be employed to target coarse particles and avoid
sedimentation, this is depicted in Fig. 9 (Johnson, 2006; Trahar and
Warren, 1976). As pointed out by Kallioinen and Heiskanen (1993)
even the design of the rotor-stator system can significantly impact
on the size range that can be effectively recovered.
Most concentrators treat a distribution of particles that encom-
pass both coarse and fine sizes and consequently it will be impos-
sible to utilise an impeller speed for optimum recovery of all size
classes; ultimately the speed used will be the optimum balance
for a given system.
Pulp density will affect the size of particles able to be recovered,
with recovery decreasing in coarser sizes as density is increased
(Lins and Adamain, 1993). The pulp density effect is of particular
importance in flash flotation, where the cyclone underflow stream
is fed directly into the cell. The densities in this stream can be as
high as 80% solids and close management of water addition is vital
not only to the success of flotation, but also on the milling opera-
tions that are directly linked to this unit. Within the flash cell itself
there exists a profile of densities, the larger heavier particles and
rocks will short-circuit to the cell discharge point under gravity,
making the tailings stream very coarse and viscous, while the light-
er particles and water will tend to the upper part of the cell, reduc-
ing the slurry density significantly in this part of the cell and
allowing for successful flotation. The implications of this type of
dual operation will be discussed further in a subsequent paper.
4.6. Final remarks on coarse particle flotation
The way coarse particles behave with respect to many of the
key process variables has been briefly discussed in order to high-
light the different requirements for successful recovery of these
particles, as opposed to particles in the finer size classes. What this
discussion has highlighted is that while there has been some re-
search into the area of coarse particle flotation, there is a lot of con-
flicting recommendations for the process operator and there is still
a significant amount that is yet to be established. Table 1 has been
adapted and expanded from the work of Kallioinen and Heiskanen
(1993) and provides a useful summary of the findings of the liter-
ature on size based flotation. This indicates that a flash flotation
cell should be operated at low froth depth, with large bubbles
and an impeller speed that is above the settling threshold to max-
imise unit recovery. Whether this reflects the ‘typical’ operating
conditions within industry is currently unconfirmed and will be
discussed in detail in a subsequent paper.
8 B. Newcombe et al. / Minerals Engineering 34 (2012) 1–10
Table 1
Relative differences of coarse and fine particles in flotation systems (Kallioinen and
Heiskanen, 1993, ⁄added by author).
Fine
Suspension in slurry Easy
Rate of collision Low
Induction time Long
Rate of bubble-particle aggregate formation Low
Rate of bubble-particle aggregate loss in transport Low
Rate of recovery into froth (true flotation) High
Rate of entrainment High
Rate of drop-back from froth Low
⁄
Air addition N/A
⁄
Froth depth N/A
⁄
Bubble size requirements Small
⁄
Impeller speed requirements Higher
5. Predicting flash flotation performance
Whilst numerous methods exist for modelling conventional and
column flotation, there are very limited examples of the applica-
tion of models to flash flotation circuits, and very little research
that considers the coarse particle sizes specifically. In industrial
plants, coarse particles (above 75 lm for example) are often unlib-
erated and are therefore considerably harder to categorise in the
available modelling framework. A brief discussion on the relevant
aspects of flotation models and the prediction of flotation perfor-
mance will be provided here; a more rigorous review of flotation
modelling in this respect is a substantial topic in itself and has
been provided elsewhere (Runge, 2009; Vianna, 2004; Savassi,
1998).
There are many modelling methods described in the literature
for predicting the flotation response of an ore; however, these
models often require accurate plant survey data to establish per-
formance parameters. Unfortunately, for the majority of flash flota-
tion installations, both the way the cells are installed within the
grinding circuit and the nature of the streams (high volume, very
coarse, often with small rocks in the feed and tails) prevent de-
tailed, accurate sampling from occurring. When this is the case
simulation programs such as JKSimFloat treat the flash flotation cell
as a ‘splitter’ and do not apply any flotation model to calculate per-
formance (Schwarz, 2010). Where samples are able to be taken, the
same conventional flotation models that are used for fine and
intermediate particle streams are applied. Whilst considerable ef-
fort has gone into predicting flotation performance on the basis
of particle size and mineralogy, none of the flotation models stud-
ied specifically included flash flotation circuits in their predictive
capabilities, and none linked feed size-by-mineral attributes to
resultant actual (measured) plant performance. Where batch flota-
tion test results and ore properties have been used to predict plant
performance, no relevant plant data is available to evaluate the
accuracy of the predictions made (Amini et al., 2009; Bulled and
Connolly, 2009; Runge et al., 2007; Dobby and Savassi, 2005;
Hay, 2005; Pietroben et al., 2004; Hay and Rule, 2003; Bazin
et al., 1994; Kallioinen and Nitti, 1985).
Of the limited studies that have been performed on flash flota-
tion systems, some are focussed on the comminution circuit, or
provide qualitative information on circuit performance at best,
whilst others are focused solely on the precious metal behaviour
and compare the results of the gravity gold recovery tests to flash
flotation performance (Yan et al., 2005; Mackinnon et al., 2003;
Laplante and Dunne, 2002). Batch flotation tests were used in sev-
eral cases to try and predict plant performance with various mea-
sures of success; however, none produced a rigorous model that is
capable of predicting actual performance for a given ore sample
(Yan et al., 2005; Mackinnon et al., 2003; Laplante and Dunne,
2002; Kallioinen and Nitti, 1985).
The work of Mackinnon et al. (2003) provides a useful insight
into the difficulties encountered when dealing with flash flotation.
Although this work was directed at improving grinding perfor-
Coarse mance, the intimate relationship between the two parts of the cir-
Difficult cuit meant that the development of a flash flotation model was
High necessary in order to predict milling response to circuit changes.
Short Both plant surveys and batch flotation tests were used to evaluate
High
High
behaviour on a size-by-size basis (mineralogical content was as-
Low sumed, with sulphur assays taken as being indicative of pyrite);
Very Low plant data was used to apply an internal size split to categorise par-
High ticles as either floatable or non-floatable, while the batch test data
Optimise
was used to establish ‘recovery factors’ for each size fraction in the
Low
Large floatable category. Scale-up factors had to be introduced to try and
Lower match the batch test response to that observed in the plant. The
resultant model contained significant errors in its predictive capa-
bilities, with large ‘adjustments’ being made, particularly to sul-
phur grade. A key finding of this work was that the plant flash
flotation cell achieved better performance than the laboratory
batch cell with particles coarser than 300 lm, although it appears
that only four batch tests were performed at different aeration
rates. A more rigorous test work program is required if an appro-
priate set of laboratory operating conditions is to be used in such
a comparison.
Yan et al. (2005) extended on the work of Mackinnon et al.
(2003) to incorporate the conventional flotation circuit of the plant
under consideration. This qualitatively illustrated that the flash flo-
tation circuit plays a significant role in determining both the over-
all grade and recovery of the operation. Unfortunately the model
predictions on the impact of changing the proportion of material
going to the flash cell are not confirmed by plant data, making it
difficult to quantify the accuracy of the model developed.
SGS MinnovEx uses batch flotation tests at two different grind
sizes to estimate the anticipated flash flotation response of an
ore. The coarser grind size is the estimated size in the top portion
of the flash flotation unit, whilst the finer grind size represents the
percent passing size (p80) delivered to the rest of the plant in the
cyclone overflow (Lozano, 2010). No information is given as to
whether the feed to the coarser float is de-slimed prior to the flo-
tation test being conducted (as it would be in a plant).
Most recently, work into the ‘modelling and simulation’ of a
flash flotation circuit has been conducted by Lamberg and Bernal
(2009). This work was conducted on a new copper/gold ore body,
using both laboratory batch flotation tests and pilot scale work
to establish kinetic rate constants on the basis of particle size
and assay data. This information was then used for flowsheet
development for the new ore body; HSC Chemistry was the soft-
ware platform used. Whilst their paper provides an example of
the use of conventional methods to predict the applicability of
flash flotation to a new ore, there is no detailed information given
as to the nature of the batch tests performed, no comparison with
actual measured mineralogical data, and (as this is a new ore body)
no comparison with actual plant performance. It can only be as-
sumed that conventional flotation models were used in this work
as no specific detail on the modelling method was given in the pa-
per. What this highlights is that there is still very little information
available in the literature that is directly applicable to the flash flo-
tation process.
With the current drive towards combining comminution and
flotation models into a single all-encompassing physical property
based model, a focus on the one section of the plant that provides
this physical link would be pertinent. Of the many types of models
available, the ones based on size-by-mineral data appear to be the
most attractive for this particular application (Bazin et al., 1994;
Runge et al., 2007; Vianna, 2004). However, as pointed out by
many authors, a hybrid approach to modelling that incorporates
both empirically derived parameters as well as the measureable
 B. Newcombe et al. / Minerals Engineering 34 (2012) 1–10
physical properties of an ore may provide the most robust solution
to producing an accurate and reliable prediction of its flotation
performance (Runge, 2009; Vianna, 2004; Savassi, 2005).
6. Conclusions
Coarse particles behave in a distinctly different way to interme-
diate or fine particles, and have distinctly different processing
requirements, and whilst it is a widely held belief throughout the
flotation industry that flash flotation cells are used to recover both
fast floating and coarse composite particles, work needs to be
undertaken to determine if this is actually the case.
The key question that must be asked is whether the current flota-
tion models can be used to reflect the behaviour within a flash flota-
tion cell, or if a different/modified model needs to be applied above a
certain size class. In order to predict the amenability of an ore to the
flash flotation process, can a laboratory batch flotation test focussed
on coarse particles, coupled with mineralogical knowledge, provide
sufficient meaningful information to estimate the performance of a
flash flotation cell at an industrial level? As plants tend to deviate
from their design capacities in order to treat increasing tonnes, thus
making the flotation feed coarser, and new greenfields ores are
becoming more complex in their mineralogy, considerable study is
warranted to understanding the behaviour of coarse particle
flotation and the modelling of the flash flotation unit.
Acknowledgements
The author wishes to thank Barrick for the on-going sponsor-
ship of this project and acknowledges the support of the AMIRA
P90 Project and its sponsors. Special thanks to Sergio Vianna for
his help and guidance in preparing this document for publication.
References
Ahmed, N., Jameson, G.J., 1985. The effect of bubble size on the rate of flotation of
fine particles. Int. J. Miner. Process. 14, 195–215.
Akerman, P.K., Harris, G.H., Klimpel, R.R., Aplan, F.F., 1987. Evaluation of flotation
collectors for copper sulphides and pyrite. I. Common sulfhydryl collectors. Int.
J. Miner. Process. 21, 105–127.
Aktas, Z., Cilliers, J.J., Banford, A.W., 2008. Dynamic froth stability: particle size,
airflow rate and conditioning time effects. Int. J. Miner. Process. 87, 65–71.
Alford, R., Clarke, G., 2007. A review of processing at the Cannington Ag/Pb/Zn mine.
In: Proceedings Ninth Mill Operators Conference, pp. 195–200. (AusIMM).
Amini, E., Alexander, D.J., Wightman, E., 2009. Using laboratory scale flotation
testing to predict pilot scale flotation performance. In: Proceedings 41st
Canadian Mineral Processors, Ottawa.
Barbian, N., Hadler, K., Cilliers, J.J., 2006. The froth stability column: measuring froth
stability at an industrial scale. Miner. Eng. 19, 713–718.
Barbian, N., Ventura-Medina, E., Cilliers, J.J., 2003. Dynamic froth stability in froth
flotation. Miner. Eng. 16, 1111–1116.
Bazin, C., Grant R., Cooper, M., Tessier, R., 1994. Prediction of metallurgical
performance as a function of fineness of grind. In: Proceedings 26th Canadian
Mineral Processors, Ottawa.
Bazin, C., Proulx, M., 2001. Distribution of reagents down a flotation bank to
improve the recovery of coarse particles. Int. J. Miner. Process. 61, 1–12.
Bravo, A., Torem, M., Monte, M., Dutra, A., Tondo, L., 2005. The influence of particle
size and collector on the flotation of a very low grade auriferous ore. Miner. Eng.
18, 459–461.
Bourke, P., 1995. Flash flotation in the gold industry. In: Proceedings Randol Gold
Forum, Perth 1995, pp. 61–71.
Bulled, D., Connolly, J., 2009. A case study in the flotation characterisation of an ore
body and its use in circuit design and risk assessment. In: Proceedings Mineral
Processing Plant Design 2009 (SME).
Coleman, R., 2011. Private Communication (Outotec).
Crawford, R., Ralston, J., 1988. The influence of particle size and contact angle in
mineral flotation. Int. J. Miner. Process. 23, 1–24.
Cutting, G.W., 1989. Effect of froth structure and mobility on plant performance.
Miner. Process. Extr. Metall. Rev. 5, 169–201.
Deglon, D.A., 2005. The effect of agitation on the flotation of platinum ores. Miner.
Eng. 18, 839–844.
Dippenaar, A., 1981a. The destabilization of froth by solids. I. The mechanism of film
rupture. Int. J. Miner. Process. 9, 1–14.
9
Dippenaar, A., 1981b. The destabilization of froth by solids. II. The rate determining
step. Int. J. Miner. Process. 9, 15–27.
Dobby, G., Savassi, O., 2005. An advanced modelling technique for scale-up of batch
flotation results to plant metallurgical performance. In: Proceedings Centenary
of Flotation, Symposium, pp. 99–103 (AusIMM).
Drzymala, J., 1994. Characterisation of materials by Hallimond tube flotation. Part
II: maximum size of floating particles and contact angle. Int. J. Miner. Process.
42, 153–167.
Farrokhpay, S., 2011. The significance of froth stability in mineral flotation – a
review. Adv. Colloid Interface Sci. 166, 1–7.
Feng, D., Aldrich, C., 1999. Effect of particle size on flotation performance of complex
sulphide ores. Miner. Eng. 12, 721–731.
Gontijo, C.F., Fornasiero, D., Ralston, J., 2007. The limits of fine and coarse particle
flotation. Can. J. Chem. Eng. 85, 739–747.
Grano, S., 2006. Effect of impeller rotational speed on the size dependent flotation
rate of galena in full scale plants. Miner. Eng. 19, 1307–1318.
Hay, M., 2005. Using the SUPASIM flotation model to diagnose and understand
flotation behaviour from laboratory through to plant. Miner. Eng. 18, 762–771.
Hay, M., Rule, C., 2003. SUPASIM: a flotation plant design and analysis methodology.
Miner. Eng. 16, 1103–1109.
Hunter, T.N., Pugh, R.J., Franks, G.V., Jameson, G.J., 2008. The role of particles in
stabilising foams and emulsions. Adv. Colloid Interface Sci. 137, 57–81.
Jameson, G.J., 1984. Physical aspects of fine particle flotation. In: Principles of
Mineral Flotation – the Wark, Symposium, pp. 215–232 (AusIMM).
Johansson, G., Pugh, R.J., 1992. The influence of particle size and hydrophobicity on
the stability of mineralized froths. Int. J. Miner. Process. 34, 1–21.
Johnson, N.W., 2006. Liberated 0–10 lm particles from sulphide ores, their
production and separation – recent developments and future needs. Miner.
Eng. 19, 666–674.
Jowett, A., 1979. Formation and disruption of particle-bubble aggregates in
flotation. In: Somarsundaran, P., (Ed.), Fine Particle Processing, pp. 722.
Kallioinen, J., Heiskanen, K., 1993. Effective flotation of a difficult nickel-ore based
on intelligent mineral technology. Miner. Eng. 6, 917–928.
Kallioinen, J., Nitti, T., 1985. The basis of coarse flotation kinetics and practical
realization of classifier sands flotation. In: Proceedings XV IMPC, Cannes, pp.
358–366.
Klimpel, R.R., Isherwood, S., 1991. Some industrial implications of changing frother
chemical structure. Int. J. Miner. Process. 33, 369–381.
Lamberg, P., Bernal, L., 2009. Modelling and simulation of SkimAirÒ flash flotation as
a part of grinding-flotation circuit – case study in Esperanza, Chile. 41st Annual
Meeting Can. Min. Proc., Canada, Jan. 2009, pp. 29–38.
Laplante, A., Dunne, R., 2002. The gravity recoverable gold test and flash flotation.
In: Proceedings 34th Canadian Mineral Processors, Ottawa.
Lins, F.F., Adamain, R., 1993. The influence of some physical variables on gold
flotation. Miner. Eng. 6, 267–277.
Lozano, C, 2010. Private Communication, July 2010 (SGS).
Lynch, A.J., Harbort, G.J., Nelson, M.G., 2010. History of Flotation, AusIMM, Flotation
Machines (Chapter 4).
Mackinnon, S., Yan, D., Dunne, R., 2003. The interaction of flash flotation with closed
circuit grinding. Miner. Eng. 16, 1149–1160.
McIvor, R.E., Finch, J.A., 1991. A guide to interfacing of plant grinding and flotation
operations. Miner. Eng. 4, 9–23.
Mulleneers, H.A.E., Koopal, L.K., Brunning, H., Rulkens, W.H., 2002. Selective
separation of fine particles by a new flotation approach. Sep. Sci. Technol. 37,
2097–2112.
Pietroben, M., Grano, S., Greet, C., 2004. Matching laboratory and plant performance
– a case study of the Elura lead circuit, Pasminco Australia Limited. Miner. Eng.
17, 811–824.
Pritzker, M.D., Yoon, R.H., Basilio, C., Choi, W.Z., 1985. Solution and flotation
chemistry of sulphide minerals. Can. Metall. Quart. 24, 27–38.
Ross, V.E., 1990. A study of the froth phase in large-scale pyrite flotation cells. Int. J.
Miner. Process. 30, 143–157.
Ross, V.E., 1991a. An investigation of sub-processes in equilibrium froths. I: the
mechanisms of detachment and drainage. Int. J. Miner. Process. 31, 37–50.
Ross, V.E., 1991b. An investigation of sub-processes in equilibrium froths. II: the
effect of operating conditions. Int. J. Miner. Process. 31, 51–71.
Ross, V.E., 1991c. The behaviour of particles in flotation froths. Miner. Eng. 4, 959–
974.
Ross, V.E., 1997. Particle-bubble attachment in flotation froths. Miner. Eng. 10, 695–
706.
Runge, K., 2009. Modelling of ore floatability in industrial flotation circuits. PhD
thesis, University of Queensland.
Runge, K., McMaster, J., Ijsselstijn, M., Vien, A., 2007. Establishing the relationship
between grind size and flotation recovery using modelling techniques. In:
Proceedings Flotation ’07, Cape Town.
Sandström, E., Jönsson, H., 1988. Unit flotation practice at Boliden Mineral AB. In:
Proceedings XVI International Mineral Processing Congress, pp. 525–533.
Savassi, O.N., 1998. Direct estimation of the degree of entrainment and the froth
recovery of attached particles in industrial flotation cells. PhD thesis, University
of Queensland.
Savassi, O.N., 2005. A compartment model for the mass transfer inside a
conventional flotation cell. Int. J. Miner. Process. 77, 65–79.
Schubert, H., 2008. On the optimisation of hydrodynamics in fine particle flotation.
Miner. Eng. 21, 930–936.
10 B. Newcombe et al. / Minerals Engineering 34 (2012) 1–10
Schulze, H.J., 1977. New theoretical and experimental investigations on stability of
bubble/particle aggregates in flotation: a theory on the upper particle size of
floatability. Int. J. Miner. Process. 4, 241–259.
Schwarz, S., 2010. Private Communication (JKTech).
Seaman, D.R., Manlapig, E.V., Franzidis, J.P., 2006. Selective transport of attached
particles across the pulp-froth interface. Miner. Eng. 19, 842–851.
Senior, G.D., Shannon, L.K., Trahar, W.J., 1994. The flotation of pentlandite from
pyrrhotite with particular reference to the effects of particle size. Int. J. Miner.
Process. 42, 169–190.
Subrahmanyam, T.V., Forssberg, E., 1988. Froth stability, particle entrainment and
drainage in flotation – a review. Int. J. Miner. Process. 23, 33–53.
Sutherland, D.N., 1989. Batch flotation behaviour of composite particles. Miner. Eng.
2, 351–367.
Suthers, S.P., Clout, J., Donskoi, E., 2004. Prediction of plant performance using feed
characterisation – an emerging tool for plant design and optimisation. In:
Proceedings Metallurgical Plant Design and Operating, Strategies, pp. 203–217
(AusIMM).
Tao, D., Luttrell, G., Yoon, R.-H., 2000. A parametric study of froth stability and it
effect on column flotation of fine particles. Int. J. Miner. Process. 59, 25–43.
Trahar, W.J., 1976. The selective flotation of galena from sphalerite with special
reference to the effects of particle size. Int. J. Miner. Process. 3, 151–166.
Trahar, W.J., 1981. A rational interpretation of the role of particle size in flotation.
Int. J. Miner. Process. 8, 289–327.
Trahar, W.J., Warren, L.J., 1976. The flotability of very fine particles – a review. Int. J.
Miner. Process. 3, 103–131.
Tsatouhas, G., Grano, S.R., Vera, M., 2006. Case studies on the performance and
characterisation of the froth phase in industrial flotation circuits. Miner. Eng.
19, 774–783.
Van der Spuy, R., Ross, V.E., 1991. The recovery of coarse minerals by agglomeration
and flotation. Miner. Eng. 4, 7–11.
Van Deventer, J., Van Dyk, W., Lorenzen, L., Feng, D., 2002. The dynamic behaviour of
coarse particles in flotation froths. Part III: ore Particles. Miner. Eng. 15, 659–
665.
Vera, M., Franzidis, J.-P., Manlapig, E., 1999. Simultaneous determination of
collection zone rate constant and froth zone recovery in a mechanical
flotation environment. Miner. Eng. 12, 1163–1176.
Vianna, S., 2004. The effect of particle size, collector coverage and liberation on the
floatability of galena particles in an ore. PhD thesis, University of Queensland.
Vieira, A.M., Peres, E.C., 2007. The effect of amine type, pH and size range in the
flotation of quartz. Miner. Eng. 20, 1008–1013.
Wade, J., 2006. A big flash at gold mine. Aust. Min. 42 (September).
Yan, D., Wiseman, D., Dunne, R., 2005. Predicting the performance of a flotation
circuit that incorporates flash flotation. In: Proceedings Centenary of Flotation,
Symposium, pp. 179–184 (AusIMM).
Yoon, R.-H., 2000. The role of hydrodynamic and surface forces in bubble-particle
interaction. Int. J. Miner. Process. 58, 129–143.
Yoon, R.-H., Luttrell, G.H., Asmatulu, R., 2002. Extending the upper particle size limit
for coal flotation. J. S. African Inst. Min. Metall., 411–415 (October).