Minerals Engineering 99 (2016) 151157

Contents lists available at ScienceDirect

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

Correlation between the top of froth grade and the operational variables
in rougher flotation circuits q
J. Yianatos , L. Bergh, L. Vinnett, I. Panire, V. Iriarte
Automation and Supervision Center for Mining Industry, CASIM, Department of Chemical and Environmental Eng., Federico Santa Maria Technical University, Chile

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

Article history: Cu grades at the top of the froth (TOF) were compared to the froth depth (hF) profiles and froth discharge
Received 17 December 2015 velocity profiles along rougher flotation circuits. Measurements were performed in two self-aspirated
Revised 5 July 2016 flotation banks, one consisting of nine 130 m3 cells (1-2-2-2-2 arrangement), and the other one of six
Accepted 24 August 2016
250 m3 cells (1-1-1-1-1-1 arrangement). Two behaviors for the TOF grade were observed: (i) for
Available online 28 August 2016
hF 6 10 cm, a decrease in the TOF grade along the bank was observed because either the decrease in min-
eral liberation or a potential recovery of slow floating gangue, (ii) for hF P 10 cm, an increase in the TOF
Keywords:
grades with the froth depth was observed, mainly due to an increase in froth selectivity. In addition, an
Top of froth
Froth discharge velocity
inverse relationship between TOF grades and froth discharge velocities was obtained in most cases.
Liberation However, in the first rougher cell, both higher froth discharge velocities (714 cm/s) and higher TOF
Flotation grades (2028%Cu) were observed. The froth in the first cell is typically loaded, which favors the froth
Froth depth stability as well as the concentrate discharge velocity.
The TOF grades of Mo as a function of the operating variables showed the same dependency as those
observed with the TOF grade of Cu. This result indicated that Cu and Mo minerals had similar flotation
rates in the rougher operation, which was in good agreement with the comparable ranges of Cu and
Mo rougher recoveries.
2016 Elsevier Ltd. All rights reserved.

1. Introduction drainage, diffusion of gas from small bubbles to large bubbles, the
breakup of bubble films and gas coalescence (Ata, 2012; Pugh,
Flotation processes are used for the selective separation of par- 1996; Yianatos et al., 1988). Particle detachment is selective if less
ticles, taking advantage of the differences in the superficial proper- hydrophobic particles are firstly detached. Selective particle
ties of materials. In the flotation processes, fine particles are detachment increases the mineral grade through the froth zone
suspended in a pulp zone, where these particles collide with bub- (Yianatos et al., 1988).
bles of gas (Yianatos, 2007). The hydrophobic particles are selec- Gourram-Badri et al. (1997) studied the separation selectivity as
tively attached to bubbles, and then the particle-bubble a function of the mineral hydrophobicity (induction time) and lib-
aggregates are transported to the top of the equipment to form a eration during the coalescence processes. The coalescence of min-
froth zone (Ata, 2012; Welsby et al., 2010). In the froth zone, the eralized bubbles was caused and non-detached particles were
particle-bubble aggregates that overcome the drainage and coales- recovered in a Hallimond tube. The experiments showed that the
cence processes are carried to the froth surface to be recovered in more hydrophilic and locked particles left the bubble surface dur-
the concentrate stream. ing the coalescence processes. Yianatos et al. (1988) observed the
The recovery of particles to the concentrate stream depends effect of the froth depth on the Mo grade through the froth zone
mainly on the chance of reaching the froth surface (Ata, 2008). of an industrial flotation column. Selectivity in the Mo recovery
Some particles constituting the particle-bubble aggregate might was observed for deep froths, greater than 1 m, with an increase
be detached from the bubbles, commonly because of a decrease in the Mo grade of 1015% along the froth zone. Seaman et al.
in the bubble surface area in the froth zone (Ata, 2009). The bubble (2004) showed a comparison between the froth surface (touch of
size growth can be caused by several sub-processes, such as liquid froth) and bubble load grades of zinc measured in the cleaner
and retreatment circuits at Teck Comincos Red Dog mine, Alaska.
q Low differences between the froth surface and bubble load grades
Presented at Flotation 15
Corresponding author at: P.O. Box 110-V, Valparaso, Chile. were observed in the cleaner circuit. Nevertheless, froth selectivity
E-mail address: [email protected] (J. Yianatos). was identified in the retreatment circuit, where the bubble load

http://dx.doi.org/10.1016/j.mineng.2016.08.018
0892-6875/ 2016 Elsevier Ltd. All rights reserved.
152 J. Yianatos et al. / Minerals Engineering 99 (2016) 151157

grade was significantly lower than the froth surface grade, with the discharge velocity is measured and recorded to the PI (process
exception of the columns and two mechanical cells. information) system by Visiofroth cameras (Metso Minerals, 2006).
More recently, mineral collection along industrial flotation cir- Five metallurgical samplings for mass balances were performed
cuits based on the top-of-froth measurement has been reported around the rougher circuits. In addition, radioactive tracer mea-
(Yianatos et al., 2014). The authors proposed the use of the top- surements were conducted to determine the solid and liquid flow-
of-froth instead of bubble load measurements for the evaluation rate (load) distribution between the rougher circuits (Yianatos
of rougher flotation circuits, where rather non-selective froths et al., 2015). The feed, tail and concentrate were sampled per cir-
are typically present. This approach was based on the assumption cuit during surveys of approximately 23 h. The samples were
that, for lowselectivity froths, the mineral that reaches the froth assayed for Cu and Mo to estimate the recovery of these elements.
surface (top-of-froth) is similar to the mineral that enters the froth The concentrate stream was sampled cell by cell (subject to avail-
zone by true flotation (bubble load measurement). A slight selec- ability) to obtain the incremental grades of Cu and Mo. The rougher
tivity has been observed by Yianatos et al. (2014) in the coarse concentrates were sampled as a composite. Each survey consisted
classes because the TOF grades of Cu were slightly greater than of sample cuts every 2030 min to obtain a composite of approxi-
the bubble load grades in these classes. Nevertheless, this bias mately 20 L. A total of 5 cuts per survey decreases the standard
did not significantly influence the overall froth selectivity. error of the mean grades according to Lotter and Laplante (2007)
There are several ways to quantify the liberation of particles. and Arru et al. (2007). The grade data were reconciled to satisfy
For instance, the degree of liberation is defined as the fraction fully the total and component mass balances around the circuits. Table 1
liberated particles of valuable mineral/valuable mineral (Gaudin, summarizes the feed characteristics along with the metallurgical
1939). Brub and Marchand (1984) defined the degree of libera- results for Cu and Mo in each rougher circuit. Note the similar
tion as the mean volumetric grade in valuable minerals within only ranges of the Cu and Mo recoveries in rougher circuits A and B.
the mineralized particles. At present, the surface liberation of a In addition, the Mo recoveries achieved comparable results to
valuable mineral can be measured by commercial systems, e.g., those obtained for Cu in both flotation circuits. With the exception
QEMSCAN and Mineral Liberation Analysis (MLA) (Lastra, 2007). of Sampling Campaign 3, the concentrate grades were in similar
In these cases, the surface liberation is defined by total area of valu- ranges for both circuits. The lower tonnage and higher residence
able minerals/total area of particles. time in Sampling Campaign 3 might lead to a decrease in the con-
In this work, top-of-froth (TOF) grade profiles as a function of centrate grade.
the froth depth and the froth discharge velocity profiles are pre- Top-of-froth measurements (TOF) were conducted along
sented. Measurements were performed in two industrial rougher rougher circuits A and B in parallel with metallurgical surveys.
circuits consisting of 130 m3 and 250 m3 self-aspirated flotation Fig. 2 shows the TOF sampler together with the typical sampling
cells. The circuit arrangements favoured a wide range of froth location. The sampler (10 cm diameter and 14 cm height) was
depths and froth discharge velocities, enabling the evaluation of immersed in the froth up to 0.5 cm from the surface to sample
the froth selectivity (selective particle detachment) as well as the the first layer of the froth. The upper layer of the froth enters the
mineral evolution down the rougher circuit. sampler through the sampler perimeter, which was designed with
a small curvature to prevent hydraulic entrainment, as shown in
Fig. 2(b). Each measurement corresponds to sample cuts approxi-
2. Experimental procedure mately every 1 min, with a total of approximately 10 cuts per mea-
surement, obtaining a composite ranging from 100 to 600 g of
A sampling campaign was conducted in an industrial flotation solid. This procedure reduces the effect of sampling errors and
plant (Cu/Mo) that processes 8000 tph of ore from SAG grinding. operating variability. The TOF sample makes it possible for parti-
The rougher flotation circuit consists of eight parallel banks. The cles attached to bubbles as well as particles in the Plateau Borders
rougher circuits A and B were analyzed, both of which have a com- (first froth layers) to be sampled. The sampling point was located
mon raw mineral feed from the secondary grinding. Circuit A con- between the froth crowder and the peripheral concentrate launder.
sists of six self-aspirated Dorr-Oliver cells of 250 m3 arranged in a This location allows for the minimization of the contamination by
1-1-1-1-1-1 configuration (Fig. 1a). The rougher circuit B consists entrainment and allows the stagnant zones to be avoided (Yianatos
of nine self-aspirated WEMCO cells of 130 m3 arranged in a 1-2- et al., 2014). At industrial scale, an entrainment percentage lower
2-2-2 configuration (Fig. 1b). The cell pairs in the rougher banks than 10% in the TOF sample has been observed along the rougher
B (i.e., 23, 45, 67, and 89) are connected by an open section banks. Yianatos et al. (2014) did not observe a significant change
(communicating vessels). The level control is implemented at the between the total grades of the bubble load and TOF in lowselec-
end of each pair of cells. In each flotation cell of Fig. 1, the froth tivity froths, which suggests that the TOF sample is representative

Fig. 1. Rougher flotation circuits along with level control points: (a) circuit A and (b) circuit B.
 J. Yianatos et al. / Minerals Engineering 99 (2016) 151157 153

Table 1
Operational variables in the rougher flotation circuits A and B.

Feed grade, % Recovery, % Concentrate grade, %

Sampling campaign Feed tonnage, tph hF, cm Cu Mo Cu Mo Cu Mo
Circuit A 1 781 510 0.58 0.012 91.3 91.6 11.2 0.23
2 798 314 0.54 0.015 91.0 87.2 11.2 0.29
3 732 48 0.67 0.013 91.3 93.1 9.8 0.20
Circuit B 4 798 430 0.73 0.020 91.1 93.3 11.9 0.33
5 835 325 0.55 0.012 90.1 92.5 11.8 0.27

(a) (b)
~ 0.5 cm

TOF
Sampler

Froth

(c)
Sampling
External
point
Launder

Internal
Launder

Froth
Crowder

Fig. 2. (a) Top-of-froth measurement, (b) TOF sampler and (c) sampling point.

of the bubble load sample for shallow froths. From the TOF mea- the pulp-froth interface, while the froth surface was registered in
surements, the characteristics of the mineral (grade and particle a graduated tube. This device allowed for the hF measurement in
size) that overcame drainage and coalescence processes can be each flotation cell along the rougher circuits. The measurement
determined. Repetitions of the TOF measurements in cells PR, C1, also showed good reproducibility, where the sample standard
C2 and C6 were conducted in Sampling Campaign 1. From these deviation was 1.2 cm.
datasets, the sample standard deviation of the TOF grades was The froth discharge velocity was measured in each rougher cell
found to be 1.3%, which shows an adequate repeatability and using the installed VisioFroth cameras, as shown in Fig. 3(a). The
reproducibility of the experimental procedure. VisioFroth system was developed by Metso Minerals for the char-
The contamination by entrainment in the concentrate stream acterization of the froth surface in industrial flotation machines
does not allow the identification of the material recovered by true (Metso Minerals, 2006). VisioFroth includes software that com-
flotation because the amount of entrained material is typically putes in real time the froth discharge velocity, bubble size distribu-
variable. This condition may lead to misdiagnosis in the perfor- tion, froth stability (bubble collapse rate), froth colorimetry, and
mance evaluation of flotation machines. In addition, the TOF grade changes in the froth texture, among other froth properties. The
profiles make it possible to obtain a greater span with this mea- froth discharge velocity is obtained from the displacement
surement system in comparison with the concentrate grade between consecutive images using the modified Fast Fourier Trans-
profiles. form (Runge et al., 2007). A Visiofroth camera was provided in each
The measurement of the froth depth, hF, was manually per- mechanical flotation cell of the rougher circuits, covering a region
formed using a float device, with three repetitions per sampling located at a distance of approximately 10 cm from the discharge
point. This device consists of a square piece of wood of a density launder, according to the installation facilities and plant layout
slightly less than that of the pulp. The float was kept stable at and as shown in Fig. 3. Data were continuously recorded to the
154 J. Yianatos et al. / Minerals Engineering 99 (2016) 151157

(a) (b)
External
Launder

Froth
Crowder

TOF sampling
Internal point
Launder
~10 cm

Discharge froth velocity

sampling point

Fig. 3. (a) VisioFroth measurement and (b) sampling points.

PI system during the TOF samplings. The plant samplings were per- condition of Fig. 4. From Fig. 4a, the froth depths in circuit A varied
formed in stable plant conditions coordinated with the control in a short range of 36 cm, with the exception of cell 2, where
room personnel who monitored the operation during the measure- hF = 10 cm was observed. Under shallow froths, such as those
ments. In addition, set-point changes were not carried out during observed in circuit A, several studies have shown that the froth
the surveys along the rougher circuit. As a result, the froth dis- zone appears to be non-selective (Ata and Jameson, 2013;
charge velocities fluctuated around the average values. Alexander et al., 2003; Vera et al., 1999). In froths with low selec-
The superficial gas rate, JG, was measured using the McGill bub- tivity, the mineral grade of the attached particles crossing the
ble size analyser (MBSA), which was adapted to determine the interface (bubble load) remains approximately constant through
local JG. In this measurement, the bubbles were collected from the froth, reaching approximately the same value at the froth sur-
the pulp zone into a tube immersed in the flotation machine, and face (Seaman et al., 2004). Particle detachment is selective if less
the JG was obtained by water volume displacement over time. hydrophobic particles are firstly detached (Yianatos et al., 1988).
The measurement was performed at the same sampling point of Therefore, the evaluation of the collected mineral may be carried
the TOF. Three repetitions were carried out in all cases. All cells out using the information provided by the TOF samples in shallow
from the rougher circuit are self-aspirated, and the JG values ranged froths. In Fig. 4a, the high TOF grades of Cu in the first cell are
from 0.5 cm/s to 0.9 cm/s in circuit A and from 0.7 cm/s to 1.6 cm/s related to the recovery of a greater amount of mineral from the
in circuit B. The JG measurements had a sample standard deviation rougher feed with a high flotation rate and high mineral grade
of 0.04 cm/s. No correlation between JG and the operational vari- and consequently a high degree of liberation (Brub and
ables such as the froth depth and froth discharge velocity was Marchand, 1984). The combined effect of decrease the liberation
observed. of floatable mineral and the recovery of slow floating gangues, such
as pyrite, can contribute to decrease the TOF grade down the flota-
tion circuit.
3. Results and discussion
As a result, at the end of the circuit, significantly lower TOF
grades of Cu were observed. The froth depths did not change signif-
The TOF grade dependency as a function of froth depth and
icantly down the circuit; thus, the froth discharge velocities (cells
froth discharge velocity was analyzed. In the rougher circuit A, a
26) were kept in a short range of 14 cm/s (Fig. 4b). In addition,
normal operation was observed with a gradual decrease of the
the froth discharge velocity was 12 cm/s in the first cell, with a
TOF grade of Cu down the circuit, as presented in the operating

Fig. 4. (a) TOF grade of Cu versus froth depth and (b) TOF grade of Cu versus froth discharge velocity. Rougher circuit A.
 J. Yianatos et al. / Minerals Engineering 99 (2016) 151157 155

froth depth of 5 cm. The higher froth discharge velocity was attrib- Therefore, the recovery as well as the TOF grade in the first rougher
uted to the greater mineral load and froth stability typically cell are commonly higher compared to the rest of the circuit. The
observed in the first cell of the rougher banks. TOF grade of Cu in cell 2 varied between 6% and 10%, whereas cells
In normal stable operation, the TOF grade gradually decreases 4, 6, 8 and 9 (shallow froths) showed TOF grades lower than 5%.
along the cell arrangement (Yianatos et al., 2014). However, Two mechanisms might explain the decrease in the TOF grade:
Fig. 5 shows that the TOF grades as well as the froth depths and decrease in liberation of the floatable mineral and recovery
the froth discharge velocities oscillated along circuit B. This result of slow floating gangue. On the other hand, cells 3, 5, 7 and 9, with
was related to the presence of cells with high froth depths (second hF > 10 cm, showed a rather selective behavior caused by bubble
cell of each cell pair of circuit B), generating low discharge veloci- coalescence due to the larger froth residence time.
ties (Fig. 5b) and peaks in the TOF grades because of the froth selec- Fig. 6b shows the TOF grades of Cu as a function of froth depth
tivity (Fig. 5a). Cells 2, 4, 6 and 8 (without level control) for the rougher circuits A and B. The cell number of circuit A is also
significantly depend on the operation of the adjacent cells, with shown for all the datasets. In almost all cases, circuit A presented
oscillations in the froth depths and the TOF grades along circuit shallow froths, with hF 6 10 cm. Cell 4 of circuit A showed only
B. This operating condition showed the impact of hF on the froth one point above hF = 10 cm, which moved this data to the zone
selectivity and the froth discharge velocity. The difference between where the selectivity becomes significant.
froth depths in cells of the same pair was attributable to a larger Fig. 7a shows the froth discharge velocity as a function of froth
pressure difference required for the transport of pulp from the first depth in circuit B. This circuit was analyzed because of its wide
to the second cell. This pressure drop can be caused by design con- range of froth discharge velocities and TOF grades. A strong corre-
straints and partial solid deposits on the bottom of the cells, espe- lation was observed, justifying the inverse relationship between
cially coarse particles. A low pulp height is therefore necessary in the TOF grades of valuable minerals and the froth discharge veloc-
the second cell of pairs to promote the pulp transport from the first ity shown in Fig. 5b. Fig. 7b shows the TOF grades of Cu as a func-
to the second cell. Despite the difference in the hF profiles between tion of froth discharge velocity for cells of circuit B. The datasets
circuits A and B, the Cu and Mo recoveries were rather similar, as were grouped according to the range of froth depth values. The
shown in Table 1. This condition was caused by the adequate sizing highest froth discharge velocities were measured in cells 1 and 4.
of the residence times, which allowed the recoveries of the two cir- For the first rougher cell (highlighted with ellipses), a different
cuits to be close. In addition, non-significant differences in the total condition with regard to the rest of the cells was observed, which
concentrate grade of circuits A and B were observed (Table 1), was again attributed to a greater mineral content in the froth sur-
where the higher concentrate grades from the controlled cells were face. As a result, higher froth discharge velocities for the same TOF
partially compensated by the lower concentrate grades from the grade were obtained. For cells 29, higher froth discharge veloci-
non-controlled cells. ties caused the TOF grades to stabilize at a relatively constant value
Fig. 6a shows a correlation between the TOF grades of Cu and ( 2% Cu). In the case of cells with hF > 10 cm, the froth discharge
the froth depths for cells in circuit B, including all the measure- velocities were lower than 4 cm/s, whereas the TOF grades of Cu
ments. The cell number is also shown for each dataset. Two differ- were greater than 5%.
ent conditions are deployed for different ranges of hF. For Fig. 8 shows the correlation between the TOF grades of Cu and
hF P 10 cm, a TOF grade increasing with the froth depth was Mo for all the datasets from the rougher circuits A and B. The data-
observed. This behavior was attributed to the froth selectivity, sets were mainly in an error band of 20%, with a strong correla-
independent of cell number. For hF < 10 cm, a TOF grade decrease tion caused by similar Cu and Mo flotation rates in the rougher
along the circuit was observed, which might be caused by both circuits. As a result, an approximately linear trend between both
the mineral liberation decrease and/or the presence of slow float- TOF grades was observed. This correlation indicated that Cu and
ing gangue. The first cell showed the highest TOF grade values, Mo minerals had a similar floatability in the flotation plant, which
which were related to a greater amount of valuable mineral with was in good agreement with the comparable ranges of Cu and Mo
a higher degree of liberation, high grade and high flotation rate rougher recoveries shown in Table 1. Therefore, the results of
recovered at the beginning of the circuit. Flotation rates are typi- Figs. 47 showed the same dependency among the TOF grades of
cally greater in the fine fractions because particles in these size Mo and the operating variables as those observed with the TOF
classes have a higher degree of liberation (Jameson, 2012). grade of Cu.

Fig. 5. (a) TOF grade of Cu versus froth depth and (b) TOF grade of Cu versus froth discharge velocity. Rougher circuit B.
156 J. Yianatos et al. / Minerals Engineering 99 (2016) 151157

Fig. 6. Correlation between the TOF grade of Cu and froth depth in: (a) rougher circuit B and (b) rougher circuit A and B.

Fig. 7. (a) Froth discharge velocity as a function of froth depth, rougher circuit B; and (b) TOF grade of Cu as a function of froth discharge velocity, rougher circuit B.

tion of this assumption in the first and second cells of the rougher
flotation banks (Yianatos et al., 2014).
A TOF grade decrease was observed along the rougher circuit A,
which was operated with froth depths lower than 15 cm. This
result was attributed to the combined effect of the mineral libera-
tion decrease and the slow-floating gangue that reached the froth
surface. The information provided by the liberation and mineralog-
ical analyses enables the identification of the predominating mech-
anism (liberation or slow-floating gangue) on the TOF grade
profile.
Further developments on this subject are in progress to improve
the reliability of the diagnostic methodology. A new device for
measuring the bubble load in continuous operation is being devel-
oped to obtain a significant amount of collected material down the
rougher banks. As a result, TOF and bubble load measurements will
be performed in all cells of rougher banks for comparative analy-
ses. In addition, liberation and mineralogical information will be
recorded for both measurement systems. These analyses (e.g.,
QEMSCAN, MLA) will allow the consolidation of the proposed
methodology to evaluate the floatable mineral collection along
Fig. 8. Correlation between the TOF grades of Cu and Mo in the rougher circuits. the rougher banks.

4. Future work 5. Conclusions

In this work, TOF samples from shallow froths were considered A dependence between the top-of-froth (TOF) grades of Cu and
a fairly good estimation of the material collected in the pulp zone the froth depths (hF) was observed in rougher flotation circuits.
that reached the froth surface. The comparison between the prop- Two different conditions were determined for different ranges of
erties of the TOF and bubble load samples has allowed the valida- froth depths. For hF < 10 cm (shallow froths), a decrease in the
 J. Yianatos et al. / Minerals Engineering 99 (2016) 151157 157

TOF grade along the circuit was observed. For hF P 10 cm, an References
increase in the TOF grade with the froth depth was observed due
to the froth selectivity. The higher froth depths allowed the pres- Alexander, D., Franzidis, J., Manlapig, E., 2003. Froth recovery measurement in plant
scale flotation cells. Miner. Eng. 16, 11971203.
ence of higher TOF grades (>15%Cu), even in the last cells of a Arru, C., Fragomeni, D., Lotter, N.O., 2007. Improvement of Mass and Value Balance
rougher bank. This result showed that a fraction of valuable min- Data from the Statistical Benchmark Model. Canadian Mineral Processors,
eral with high grade can still be recovered at the end of the bank. Ottawa, pp. 607616, January 2007, Paper no: 36.
Ata, S., 2008. Coalescence of bubbles covered by particles. Langmuir 24 (12), 6085
An inverse relationship between the froth discharge velocity 6091.
and froth depth was observed. As a result, high TOF grades were Ata, S., 2009. The detachment of particles from coalescing bubble pairs. J. Colloid
observed for low discharge velocities and vice versa. For the first Interface Sci. 338 (2), 558565.
Ata, S., 2012. Phenomena in the froth phase of flotationa review. Int. J. Miner.
rougher cells, a different behavior was observed, in which high dis-
Process. 102103, 112.
charge velocities (714 cm/s) together with high TOF grades (20 Ata, S., Jameson, G., 2013. Recovery of coarse particles in the froth phase a case
28% Cu) were obtained. Froths with higher solid content, caused study. Miner. Eng. 45, 121127.
by the fast recovery of minerals with high flotation rate, enhanced Brub, M., Marchand, J., 1984. Evolution of the mineral liberation characteristics of
an iron ore undergoing grinding. Int. J. Miner. Process. 13 (3), 223237.
both the froth stability and the discharge velocities to the concen- Gaudin, A., 1939. Principles of Mineral Processing. McGraw-Hill, New York.
trate stream. Gourram-Badri, F., Conil, P., Morizot, G., 1997. Measurements of selectivity due to
An approximately linear correlation between the TOF grades of coalescence between two mineralized bubbles and characterization of MIBC
action on froth flotation. Int. J. Miner. Process. 51, 197208.
Cu and Mo was observed in the analyzed rougher circuits. This cor- Jameson, G., 2012. The effect of surface liberation and particle size on flotation rate
relation was attributed to the similar flotation rates between the constants. Miner. Eng. 3638, 132137.
Cu and Mo minerals, which were in good agreement with the com- Lastra, R., 2007. Seven practical application cases of liberation analysis. Int. J. Miner.
Process. 84, 337347.
parable Cu and Mo rougher recoveries (90.793.5%Cu and 87.2 Lotter, N.O., Laplante, A.R., 2007. Statistical benchmark surveying of production
93.8%Mo) obtained from the metallurgical balances. concentrators. Miner. Eng. 20 (8), 793801.
For lowselectivity froths, the TOF measurements allowed the Metso Minerals, 2006. Gua de instalacin y del usuario de OCS VisioFroth TM,
Metso Minerals. Metso Minerals, 2010. <http://www.metso.com/
identification of two flotation mechanisms that explain the MetsoMinerals>.
decrease in the grade at the froth surface along the rougher banks: Pugh, R., 1996. Foaming, foam films, antifoaming and defoaming. Adv. Colloid
the true flotation of the less liberated valuable minerals with a Interface Sci. 64, 67142.
Runge, K., McMaster, J., Wortley, M., La Rosa, D., Guyot, O., 2007. A correlation
lower grade and the recovery of slow-floating gangue. In addition,
between visiofrothTM measurements and the performance of a flotation cell. In:
the froth selectivity was clearly observed for deeper froths. The Ninth Mill Operators Conference. Fremantle, Western Australia, 1921 March.
ranges of froth depth for which the flotation mechanisms or the Seaman, D., Franzidis, J., Manlapig, E., 2004. Bubble load measurement in the pulp
froth selectivity have a more significant impact on the TOF grade zone of industrial flotation machines a new device for determining the froth
recovery of attached particles. Int. J. Miner. Process. 74, 113.
were identified. On this subject, further developments are in pro- Vera, M., Franzidis, J., Manlapig, E., 1999. Simultaneous determination of collection
gress to obtain better insight into the relative effects of the float- zone rate constant and froth zone recovery in a mechanical flotation
able mineral evolution and the slow-floating gangue on the environment. Miner. Eng. 12 (10), 11631176.
Welsby, S., Vianna, S., Franzidis, J., 2010. Assigning physical significance to
collection process down the flotation banks. floatability components. Int. J. Miner. Process. 97, 5967.
Yianatos, J., 2007. Fluid flow and kinetic modelling in flotation related processes:
columns and mechanically agitated cellsa review. Chem. Eng. Res. Des. 85
(A12), 15911603.
Yianatos, J., Bergh, L., Vinnett, L., Rojas, I., 2014. On the collection of valuable
Acknowledgements minerals along rougher flotation banks. Miner. Eng. 6668, 202206.
Yianatos, J., Finch, J., Laplante, A., 1988. Selectivity in column flotation froths. Int. J.
Miner. Process. 23, 279292.
Funding for process modeling and control research is provided Yianatos, J., Vinnett, L., Iriarte, V., Henrquez, F., Daz, F., 2015. Metallurgical
by CONICYT, Project Fondecyt 1160547 and Federico Santa Maria characterization of the collective and selective flotation plants at Minera Los
Technical University, Project 116.27.2. Pelambres. In: Conference of Metallurgists COM 2015, August 2326, Canada.