Minerals Engineering xxx (2012) xxx–xxx

Contents lists available at SciVerse ScienceDirect

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

The effect of surface liberation and particle size on flotation rate constants
Graeme J. Jameson ⇑
Centre for Multiphase Processes, University of Newcastle, NSW 2308, Australia

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

Article history: The recovery of mineral particles by flotation is a strong function of particle size. As the size of floatable
Available online xxxx particles increases, the recovery increases also, until it reaches a maximum, before decreasing monoton-
ically. Previous work has focused on liberated material or ore particles of unknown individual composi-
Keywords: tion. Until recently, there has been no data on the size-by-size behaviour of partially liberated minerals.
Flotation This paper presents a re-interpretation of recent experimental results for the flotation of galena parti-
Coarse particle flotation cles in an operating concentrator. The rate constants for each size fraction and liberation class were mea-
Flotation kinetics
sured. Composites floated more slowly than liberated particles, but a fresh analysis of the data shows that
Flotation machines
Composites
the general shape of the distribution of rate constant with particle size is unaffected by liberation (com-
posite formation). For each liberation class, the ratio of the rate constant k to the maximum rate constant
for completely liberated particles kmax, was independent of particle size. A flotation liberation function
L = k/kmax can be defined, which is a function of the fractional liberation. For this ore, the liberation func-
tion is of the form L = k/kmax = ax exp (bxc), where x is the fractional liberation (0 6 x 6 1), and a, b and c
are constants. The liberation function is expected to depend on the ore type.
The effect of contact angle on the size-by-size recovery of fully liberated chalcopyrite particles in a
mechanical cell has been examined. The recovery-particle size response for these particles followed
the classical shape. A plot of k/kmax vs contact angle, where kmax is the rate constant at the greatest contact
angle, showed that the flotation response was essentially independent of particle size.
The observed poor recovery of coarse particles cannot be attributed to lack of liberation. Partial surface
liberation affects the rate constants of all particles in the same way, independently of size. The distribu-
tion of recoveries with particle size is determined by the response of fully liberated particles. The rate
constants for coarse composites follow those for fully-liberated particles of the same size. The decline
in recovery of coarse particles is related to the hydrodynamic conditions in the flotation cell.
Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction others in which the mineral to be floated is contained within com-

In the flotation process for the separation of valuable minerals
from ores, it is well known that the flotation rate constant is a func-
tion of particle size. For a given type of flotation machine, the rate
constant is low for ultrafine particles, but with increasing size, it
increases until a maximum is observed, after which it declines
monotonically. Such behaviour was has been seen in operating
plants (Gaudin et al., 1931) and in experimental investigations
(Jowett, 1980; Trahar, 1981).
In practical applications, the valuable mineral to be floated is
initially embedded in a host rock, from which it must be liberated,
by grinding. The degree of liberation is dependent on the ore type
and grind size. Thus the feed to flotation can consist of particles
that are fully liberated, predominantly the smaller particles, and
⇑ Tel.: +61 2 4921 6181; fax: +61 2 4960 1445.
E-mail address: graeme.jameson@newcastle.edu.au
posite particles, along with unwanted gangue mineral. Although it
has been surmised that the reduction in rate constant for coarse
particles can be attributed to the formation of composites (see
for example Runge et al., 2007), this aspect has not been confirmed
by observation.
Until the advent of measurement techniques such as the
QEM-Scan (Sutherland and Gottlieb (1991) and the Mineral Liber-
ation Analyser (MLA) (Gu, 2003), it was impossible to measure eas-
ily, the surface liberation of a population of particles. Without this
information, the particle size and liberation class of particles in the
feed and concentrate cannot be determined, and accordingly, the
effect on the recovery or rate constant of either of these two vari-
ables cannot be identified. Recently however, Welsby et al. (2010a)
have quantified the feed and concentrate particles using the MLA
in a way that allows us to investigate the true reasons for the de-
cline in rate constant with increasing particle size. In another re-
cent work, Muganda et al. (2011) have reported on the effect of

0892-6875/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.mineng.2012.03.011

Please cite this article in press as: Jameson, G.J. The effect of surface liberation and particle size on flotation rate constants. Miner. Eng. (2012), http://
dx.doi.org/10.1016/j.mineng.2012.03.011
2 G.J. Jameson / Minerals Engineering xxx (2012) xxx–xxx

Table 1
Data from Table 3, Welsby et al. (2010a).

kij, min1 Liberation class (mean percent)

Size class Mean (lm) 5 15 25 35 45 55 65 75 85 95 100
 +
Rate constants of floatable particles
+106 115.5 0 0.01 0.02 0.03 0.07 0.12 0.13 0.19 0.24 0.26 1.39
106 75 90.5 0.02 0.05 0.11 0.17 0.2 0.26 0.47 0.63 0.77 1.26 2.52
75 38 56.5 0.05 0.1 0.16 0.27 0.4 0.63 0.77 0.83 1.12 1.99 3.83
38 28 33 0.07 0.1 0.17 0.28 0.39 0.6 0.78 0.96 1.62 2.69 4.36
28 19 23.5 0.06 0.09 0.16 0.27 0.38 0.63 0.75 0.98 1.64 2.74 4.23
19 10 14.5 0.06 0.1 0.16 0.24 0.32 0.39 0.4 0.49 0.68 1.07 2.13
10 0 5 0.03 0.04 0.04 0.08 0.08 0.1 0.12 0.13 0.14 0.27 0.67
mij  100
Mass of floatable mineral in each liberation class, percent
+106 115.5 0.65 0.65 0.47 0.38 0.17 0.13 0.09 0.07 0.05 0.1 0.37
106 75 90.5 0.22 0.17 0.14 0.16 0.14 0.13 0.07 0.07 0.1 0.22 1.15
75 38 56.5 0.32 0.32 0.28 0.28 0.25 0.24 0.23 0.24 0.29 0.92 6.95
38 28 33 0.07 0.07 0.06 0.07 0.07 0.08 0.08 0.07 0.09 0.38 11.37
28 19 23.5 0.07 0.07 0.06 0.06 0.08 0.07 0.07 0.07 0.09 0.35 10.4
19 10 14.5 0.07 0.09 0.08 0.08 0.12 0.11 0.11 0.12 0.14 0.46 14.81
10 0 5 0.05 0.12 0.11 0.11 0.12 0.13 0.17 0.18 0.22 0.35 23.8

contact angle on the flotation kinetics of single-mineral chalcopy-
rite, on a size-by-size basis. These papers provide detailed evidence
concerning the flotation kinetics of pure and composite particles.
In this paper, the experimental evidence is reviewed, and used
to investigate the effect of liberation fraction and particle size on
rate constants in the flotation of a galena ore.
2. Liberation and particle size
In data obtained by early workers in the field, the particle size
effect was reported in terms of the recovery. Presumably, it was
simpler to report the overall recovery after a specified flotation
time in a batch test, or residence time in a continuous flotation cell
or bank of cells, than to calculate the rate constants. Naturally, the
two are closely linked. The recovery is a manifestation of the rate
constant of a given size particle, in the case of a pure mineral.
For a collection of composite particles, the recovery will be the
aggregate of the recoveries of particles in a given size band, with
a range of fractional liberations, weighted according to the mass
fractions of the particles in the feed. While the recovery is a valu-
able concept, especially from an operational point of view, the rate
constant as a function of particle size and liberation class is more
useful for interpreting the behaviour of a system with distributions
of both particle size and liberation class.
Welsby et al. (2010a) described an investigation in which a
ground ore was fed continuously to a 40-L pilot plant. The mineral
was a galena ore from the Cannington deposit in Queensland, Aus-
tralia. It was taken from the plant feed to the roughers, on site. So-
dium ethyl xanthate was used as collector. Samples of concentrate
were separated in a Warman Cyclosizer, and analysed by MLA. Rate
constants were determined for each particle size band, and for ele-
ven liberation classes. The data were analysed with reference to
the Floatability Component Model (FCM), and the Physical Prop-
erty Based Model, developed by the Julius Kruttschnitt Mineral Re-
search Centre at the University of Queensland. Australia. The paper
contains an excellent description of each of these models and the
way they can be utilised. However, the extensive data reported
in this paper has other uses. With further interpretation, some very
interesting results emerge.
Table 1 shows the data reported in Table 3 of Welsby et al. The
liberation classes in the original were reported as brackets of 10%
by weight, but for present purposes, these have been replaced by
the average percentages in each class, which are used in discus-
sions as the ‘‘fractional liberation’’.
The authors separated the particles into the two groups because
the floatability component model (FCM) assumes that all particles in
the concentrate can be classified as slow floating, fast floating and
non-floating (not shown here). We see that the fast-floating size-
by-liberation classes (shown bolded in Table 3) contains particles
that are there for two reasons: either they are highly liberated, or
they are in a size class that, for whatever reason, happens to have
high flotation rates. We can explore the flotation behaviour more
deeply by plotting the rate constant as a function of the percentage
liberation class and particle size, as shown in Fig. 1.
We see that although the rate constant decreases as liberation
decreases, the shapes of the distributions remain approximately
the same, suggesting similarity in the response to liberation.
Accordingly, k/kmax, the ratio of the rate constant at a given particle
size and liberation class, to the rate constant for fully-liberated
particles of the same size, has been calculated as shown in Fig. 2.
It is seen that the rate constant ratio as a function of liberation is
essentially the same for each particle size. The line of best fit
shown in Fig. 2 can be described as a flotation liberation function
L = k/kmax. In this instance it has the form
L ¼ axebx
c
ð1Þ
where L = k/kmax and x is the fractional liberation (0 6 x 6 1); the
constants have the values a = 0.27, b = 1.30 and c = 10.80. This rela-
tion is purely empirical. The constants will probably be a function of
cell hydrodynamics and system surface chemistry.
The result shown in Fig. 2 is quite remarkable. It allows us to
make generalisations about the effect of liberation on the rate con-
stant in a way never before possible. More importantly, it suggests
that the rate constant for a fully-liberated ore is determined solely
by hydrodynamics and the surface chemistry of the system. If the
chemical regime is constant, the peak in the k–dp curve for fully lib-
erated particles depends only on cell hydrodynamics. The rate con-
stant of a particle of a given liberation class is firstly determined by
that of the fully-liberated particle of the same size. A correction
factor can then be applied to allow for the effect of liberation.
We can compare these results with the data for batch flotation
of chalcopyrite reported by Sutherland (1989). The chalcopyrite
data are for 38 + 32 lm particles, so data from Welsby et al. for
the mean size of 33 lm have been used. Sutherland reported the
recovery after specific time intervals, but not the rate constants.

Please cite this article in press as: Jameson, G.J. The effect of surface liberation and particle size on flotation rate constants. Miner. Eng. (2012), http://
dx.doi.org/10.1016/j.mineng.2012.03.011
 G.J. Jameson / Minerals Engineering xxx (2012) xxx–xxx 3

Fig. 1. Rate constant data of Welsby et al. (2010a) plotted against mean size in
Fig. 3. Comparison of fractional recovery ratios for galena (Welsby et al., 2010a)
band.
and chalcopyrite (Sutherland, 1989), for 33 lm particles.

powerful effect on the rate constant. Sutherland’s (1989) work was

done with naturally-floating chalcopyrite, and the contact angle is
not known.
Another reason for the reduction in rate constant may relate to
the way the hydrophobic patches are distributed over the surface
of the particles. For example, if the surface of a 50% liberated par-
ticle consists of two halves, one being galena and the other the
gangue, it is likely that the way in which it attaches to bubbles,
and the strength of the bond, may be very different if the galena
is scattered over the surface of the particle like salt and pepper.
The surface topography or disposition of the valuable component
of the surfaces of the galena mineral particles was not described
in the paper of Welsby et al. (2010a).

2.1. Overall rate constants and contributions from liberation classes

Fig. 4 shows the overall rate constant plotted against particle

diameter. Contributions from all liberation classes are included.
Fig. 2. Ratio of the rate constant k to the maximum rate constant kmax as a function
of the liberation class, for particles of the mean size shown.

For purposes of comparison, it is necessary to convert Welsby’s

rate constants into recoveries. For a continuous flotation cell, the
rate constant is given by (Levenspiel, 1999):

R slope = 1.14
¼ ks; ð2Þ
ð1  RÞ

where s is the residence time in the cell. Recovery was calculated

using Eq. (2), with a residence time of 4.7 min (Welsby et al.,
2010b).
Fig. 3 shows a comparison of the chalcopyrite and galena data slope = - 2.14
for particles in the size band 38 + 32 lm. The values at zero liber-
ation may result from naturally hydrophobic gangue particles in
the feed, or entrainment into the froth.
It appears that the recovery for the galena particles is much
more affected by composite formation than is the case with the
chalcopyrite. One explanation may relate to the contact angles in
the two systems. Sutherland and Wark (1955) report that the max-
imum contact angle possible with galena and potassium ethyl xan-
thate is 60°. Muganda et al. (2011) showed that the contact angle,
for which values as high as 100° are possible for chalcopyrite, has a Fig. 4. Overall rate constants from the data of Welsby et al. (2010a).

Please cite this article in press as: Jameson, G.J. The effect of surface liberation and particle size on flotation rate constants. Miner. Eng. (2012), http://
dx.doi.org/10.1016/j.mineng.2012.03.011
4 G.J. Jameson / Minerals Engineering xxx (2012) xxx–xxx

The data divides itself into two groups, each representative of com- 5.0
Contact angle
peting effects in the flotation process. The smallest particles are Contact angle
degrees
degrees
relatively unaffected by detachment, so the positive slope to the
line of best fit is an indication of the effect of particle size on the 4.0
88
attachment process. Conversely, past the maximum, detachment

Rate constant, 1/min

becomes important. 78
3.0
However, the shape of the overall rate constant curve in Fig. 4 is 73
misleading. The overall rate constant depends on the individual
rate constants for a given liberation class, weighted according to 2.0 68
the masses of that class in the feed and the concentrate. The slope 63
of the line of best fit for the coarse particles in Fig. 4 is quite differ-
1.0 53
ent to the slopes of similar lines for individual liberation classes in
Fig. 1. Thus if the performance of a flotation cell is modeled using 38
the overall recovery alone, it must be appreciated that the result 0.0
would only be valid for flotation feeds of the same ore type, frac- 10 100 1000
tional liberation and mass distribution of particles in the feed Mean size in band, µm, for contact angles shown
and the concentrate.
The prediction of the recovery starts with knowledge of the rate Fig. 5. Data of Muganda et al. (2011) re-plotted to show the effect of contact angle
on flotation rate constant for fully liberated chalcopyrite particles.
constant as a function of particle size and fractional liberation. But
it is not sufficient to know only the rate constant. To calculate the
overall rate constant, we must also know the mass fraction of each even the contact angle, has no influence on the general shape of the
size-liberation class in the feed, as shown in Table 1. It would also k–dp distribution, although the details are obviously affected.
be misleading to use the overall data in Fig. 4 to draw conclusions If we plot the rate constant ratio k/kmax from Muganda’s data,
about the effect of the fundamental variables such as particle size taking as the maximum rate constant kmax the value at the highest
and cell hydrodynamics, on the overall rate constant. The latter is contact angle, 88°, for a given particle size, the graph shown in
only the sum of the constituent parts. Fig. 6 results.
It is seen that the distribution of rate constants with particle
size follows a similar pattern to that observed by Welsby et al.
(2010a) for particles of constant fractional liberation. Since all
3. Effect of collector on fully liberated particles the particles in the work of Muganda et al. (2011) were fully
liberated, it can be concluded that liberation or composite forma-
Examination of the data of Welsby et al. (2010a), shows that the tion, or even the contact angle, has no effect on the general shape
fully liberated particles follow the classical recovery-particle size of the k–dp distribution. It is interesting to note that in Fig. 6, there
behaviour. These authors did not alter the reagent concentration is no systematic effect of particle size. There is considerable spread
in their work, but it would be expected that the rate constants of the data around the line of best fit, but it appears that particles
would change if the contact angle or collector adsorption concen- with the same contact angle are all affected equally, independent
tration were varied. Early investigators appreciated the importance of size.
of the contact angle in determining recovery (Sutherland and It is worth noting that some of the observed effects of liberation
Wark, 1955; Glembotskii et al., 1963; Klassen and Mokrousov, may be linked to the Cassie equation (Cassie and Baxter, 1944),
1963) but there has been little published data on the combined ef- which relates the effective contact angle of an inhomogeneous sur-
fects of particle size and contact angle on rate constants. Vianna face, to the contact angles and the fractional area occupied by each
(2004) reported experiments in which rate constants for galena of the components. The validity of this equation has been ques-
particles were measured as a function of liberation and collector tioned frequently over the years (see Gau and McCarthy, (2009),
coverage. The rate constants for a given particle size and liberation
class were strongly influenced by the collector coverage, but inter-
estingly, the general form of the data curves remained much the
same as those shown in Fig. 1. It was also seen that as the surface 1.0
coverage increased, there was a shift in the peak in the curves to- 0.9 20µm
wards larger particle sizes, as well as an increase in the recovery.
0.8 45µm
According to Evans (2011), Vianna considered the particle compo-
sition in terms of mass percentage (calculated from measured vol- 0.7 64µm
umetric percentage) rather than surface composition so the 0.6 90µm
liberations reported may not correspond strictly to fractional sur-
k/kmax

face liberation. Nevertheless the work is valuable in showing the 0.5 127µm
importance of reagent coverage and presumably the contact angle 0.4 175µm
on the flotation of composite particles.
0.3 255µm
Recently, Muganda et al. (2011) reported the constants for a sin-
gle-mineral sample of chalcopyrite, as a function of contact angle. 0.2 325µm
Their data have been plotted in Fig. 5. In effect, the results repre- Best fit
0.1
sent the behaviour of a fully liberated mineral as a function of par-
ticle size and contact angle. 0.0
It is seen that the distribution of rate constants with particle 0 10 20 30 40 50 60 70 80 90
size follows a similar pattern to that observed by Welsby et al. Contact angle, degrees
(2010a) for particles of constant fractional liberation. Since all Fig. 6. Rate constant ratios k/kmax vs contact angle for the data of Muganda et al.
the particles in the work of Muganda et al. (2011) were fully liber- (2011). The maximum rate constant has been taken to be the value at the highest
ated, it can be concluded that liberation or composite formation, or contact angle, 88°, for a given particle size.

Please cite this article in press as: Jameson, G.J. The effect of surface liberation and particle size on flotation rate constants. Miner. Eng. (2012), http://
dx.doi.org/10.1016/j.mineng.2012.03.011
 G.J. Jameson / Minerals Engineering xxx (2012) xxx–xxx
for a historical account). While it has been shown to be true at the
molecular level (Zhang and Kwok, 2005), Gau and McCarthy pres-
ent convincing arguments for the proposition that it is not relevant
for macroscopic surface inclusions and irregularities such as are
found on mineral surfaces. In flotation it is necessary to distinguish
between the spreading of a contact line on a mineral particle,
which leads to the collection efficiency, and the attachment of bub-
bles to inhomogeneous surfaces that leads to Schulze’s equation,
see below. In the latter case, the bubbles are pinned to irregulari-
ties and the edges of hydrophobic particles, so the attachment
force is proportional among other things to the length of the con-
tact lines, rather than to the fractional area or surface liberation
of the hydrophobic particles as suggested by Cassie’s law. This
point deserves further attention (Gautam and Jameson, 2011).
4. The decline in recovery of coarse particles
The data examined here show that neither liberation (compos-
ite formation) nor collector concentration is the cause of the de-
cline in recovery in the coarse particle size range. This decline is
almost certainly due to hydrodynamic effects, of which two can
be singled out.
4.1. The increase in detachment forces with increasing particle size
Schulze (1977, 1982) derived an expression for a critical Bond
number Bo, such that when Bo > 1, a particle on the surface of a
bubble rotating in a turbulent field will detach. In simplified form,
with minor corrections (Goel and Jameson, 2011), Schulze’s
expression can be written:
2 1=3
3:75qp dp e2=3 =db
Bo ¼
6r sinð180  h=2Þ sinð180 þ h=2Þ
where qp is the density of the particle; dp, db are the diameters of
the particle and the bubble respectively; e is the rate of dissipation
of mechanical energy in the flotation cell; r is the surface tension
and h is an angle of contact. The equation shows that the detach-
ment force varies as the square of the particle size, so all other
things being equal, as dp increases, a point is reached where coarse
particles begin to detach from bubbles except in regions of the cell
where the dissipation rate e is so low that particles can remain at-
tached to bubbles. But the rate of collection is also related to the
turbulent dissipation rate (Bloom and Heindel, 2002), so low values
of e will lead to low rate constants and hence to low recoveries. This
argument explains the reason for the peak in the recovery vs parti-
cle size curve and the continuing reduction in the rate constant with
increasing particle size.
4.2. Loss of suspension of coarse particles
Another reason for the decline in recovery is that as particles
become larger, it becomes more difficult to obtain complete off-
bottom suspension in a mechanical cell, without increasing the en-
ergy input and hence leading to detachment by Schulzes’s equa-
tion. Thus coarse particles will more easily deposit in the base of
the flotation cell, or alternatively, may exist in a low-lying ‘cloud’
in the base of the cell (Van der Westhuizen and Deglon, 2007) in
the region of the impeller. Here the rate of dissipation of energy
is so large that coarse particles can never form stable attachments
with bubbles.
From the discussion above, the shape of the recovery-particle
size distribution depends on the spatial distribution of the energy
dissipation rate throughout the flotation cell, and the ability of the
impeller to maintain the particles in suspension at high fractional
cloud heights. Accordingly, it appears likely that there will be dif-
Please cite this article in press as: Jameson, G.J. The effect of surface liberation and particle size on flotation rate constants. Miner. Eng. (2012), http://
dx.doi.org/10.1016/j.mineng.2012.03.011
5
ferences in performance between the various designs of flotation
cell currently on the market.
The phenomena of particle detachment and insufficient suspen-
sion are able to provide sufficient reasons for the reductions in
recovery observed with coarse particles. Composite formation will
of course affect the recovery of coarse un-liberated particles, but
not in a way that is different to all the other sizes present in the feed.
5. Conclusions
Analysis of recently published data has shown conclusively that
the general shape of the distribution of rate constant with particle
size is unaffected by liberation (composite formation) or contact
angle.
The data of Welsby et al. (2010a) re-affirm the classical result
for completely liberated particles, in which the rate constant is
low for ultrafines, increases gradually with increasing particle size
until a peak is reached, beyond which it declines monotonically.
For composite particles of a given liberation class, the rate constants
follow the curve for the fully liberated particles. For a given frac-
tional surface liberation, all particles are equally affected, indepen-
dently of particle size.
The ratio of the rate constant k for a particular particle size, to
the rate constant for fully liberated particles of the same size, kmax,
is dependent only on the liberation class and is independent of par-
ticle size. This observation gives rise to the concept of a flotation
liberation function L, which takes the form L = k/kmax = ax exp (bxc)
for the present data, where x is the fractional liberation (0 6 x 6 1),
and a, b and c are constants.
The experimental data were obtained during continuous pilot
plant test work at an operating site, where the flotation product
ð3Þ was galena. It is highly likely that the flotation liberation function
is ore-dependent.
In another set of recently published data, Muganda et al. (2011)
reported the effect of contact angle on the size-by-size recovery of
fully liberated chalcopyrite particles in a mechanical cell. Again,
the recovery-particle size response for these particles followed
the classical shape. A plot of k/kmax vs contact angle showed that
the flotation response was essentially independent of particle size.
Taking these experimental results together, it can be confi-
dently stated that the reason for the poor recoveries of coarse par-
ticles has nothing to do with poor liberation, since even the fully
liberated particles are affected in the same way with changes in
particle size. Partial surface liberation affects the rate constants
of all particles in the same way, independently of size. The distri-
bution of recoveries with particle size is determined by the re-
sponse of fully liberated particles, to which a correction factor
can then be applied to make allowance for the fractional surface
liberation. The decline in recovery of coarse particles is related to
the hydrodynamic conditions in the flotation cell.
References
Bloom, F., Heindel, T.J., 2002. On the structure of collision and detachment
frequencies in flotation models. Chemical Engineering Science 57, 2467–2473.
Cassie, A.B.D., Baxter, S., 1944. Wettability of porous surfaces. Transactions of the
Faraday Society 40, 546–551.
Evans, C. L., 2011. Development of a methodology to estimate flotation separability
from ore microtexture. PhD Thesis, JKMRC, The University of Queensland.
Gau, L., McCarthy, T.J., 2009. An attempt to correct the faulty intuition perpetuated
by the Wenzel and Cassie ‘‘Laws’’. Langmuir 25 (13), 7249–7255.
Gaudin, A.M., Groh, J.O., Henderson, H.B., 1931. Effect of particle size in flotation.
American Institute of Mining Engineers Technical Publication 414, 3–23.
Gautam, A., Jameson, G.J., 2011. The attachment of particles to bubbles in flotation.
In: Flotation ’11, Cape Town.
Glembotskii, V.A., Klassen, V.I., Plaksin, I.N., 1963. Flotation. Primary Sources, New
York.
Gu, Y., 2003. Automated scanning electron microscope based mineral liberation
analysis. Journal of Minerals and Materials Characterisation 2 (1), 33–41.
6 G.J. Jameson / Minerals Engineering xxx (2012) xxx–xxx
Goel S., Jameson, G.J., 2011. Effect of energy input on detachment of particles from
bubbles in a turbulent flow. In: Flotation ’11, Cape Town.
Jowett, A., 1980. Formation and disruption of particle–bubble aggregates in
flotation. In: Somasundaran, P. (Ed.), Fine Particles Processing. American
Institute of Mining, Metallurgical, and Petroleum Engineers, New York, pp.
720–754.
Klassen, V.I., Mokrousov, V.A., 1963. An Introduction to the Theory of Flotation.
Butterworths, London.
Levenspiel, O., 1999. Chemical Reaction Engineering, third edition. Wiley, New
York.
Muganda, S., Zanin, M., Grano, S.J., 2011. Benchmarking flotation performance.
Single minerals. International Journal of Mineral Processing 98, 182–194.
Runge K, McMaster, J, Ijsselstijn, M., Vien, A., 2007. Establishing the relationship
between grind size and flotation recovery using modeling techniques. In:
Flotation ’07, Cape Town.
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. International Journal of Mineral Processing 4, 241–259.
Schulze, H.J., 1982. Dimensionless number and approximate calculation of the
upper particle size of floatability in flotation machines. International Journal of
Mineral Processing 9, 321–328.
Sutherland, D.N., 1989. Batch flotation behavior of composite particles. Minerals
Engineering 2 (3), 351–367.
Please cite this article in press as: Jameson, G.J. The effect of surface liberation and particle size on flotation rate constants. Miner. Eng. (2012), http://
dx.doi.org/10.1016/j.mineng.2012.03.011
Sutherland, D.N., Gottlieb, P., 1991. Application of automated quantitative
mineralogy in mineral processing. Minerals Engineering 4 (7–11), 753–762.
Sutherland, K.L., Wark, I.W., 1955. Principles of Flotation. Australasian Institute of
Mining and Metallurgy, Melbourne.
Trahar, W.J., 1981. A rational interpretation of the role of particle size in flotation.
International Journal of Mineral Processing, 289–327.
Van der Westhuizen, A.P., Deglon, D.A., 2007. Evaluation of solids suspension in a
pilot-scale mechanical flotation cell: the critical impeller speed. Minerals
Engineering 20 (3), 233–240.
Vianna, S.M.S.M., 2004. The effect of particle size, collector coverage and liberation
on the flotability of galena particles in an ore. PhD Thesis, JKMRC, The University
of Queensland.
Welsby, S.D.D., Vianna, S.M.S.M., Franzidis, J.-P., 2010a. Assigning physical
significance to floatability components. International Journal of Mineral
Processing 97, 59–67.
Welsby, S.D.D., Vianna, S.M.S.M., Franzidis, J.-P., 2010b. A continuous pilot-scale
flotation rig for the systematic study of flotation variables. Minerals Engineering
23, 826–829.
Zhang, J., Kwok, D.Y., 2005. On the validity of the Cassie equation via a mean free-
energy lattice Boltzmann approach. Journal of Colloid and Interface Science 282,
434–438.