XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 2863
Kruttschnitt Mineral Research Centre (JKMRC). It is com-
prised of two transparent columns separated by a porous
sintered disc. The air and fluidised water are introduced
in the lower column (or plenum) below the sintered disc.
The presence of the sintered disc enables uniform fluidisa-
tion and bubble distribution in the top chamber. For more
details regarding the design of this equipment, please refer
to (Verster, Awatey, Forbes et al., 2023).
Bubble Size Analysis
All the bubble size measurements were conducted using a
ParticleView ™ V19 (Mettler Toledo). The probe comprises
a waterproof tube with a mounted camera and a process-
ing block. The probe camera has a viewing area measur-
ing 1300 × 890 μm. Online images could be observed and
recorded using the iC PVM ™ software (Wang, Forbes, &
Forbes, 2022). Throughout the tests, images were captured
at one frame every two seconds. This rate is sufficient to
ensure that no bubbles are recorded twice.
The image analysis was conducted using in-house
MATLAB ® code developed for this purpose. The code
employs the circular Hough transform on binary images
to identify bubbles. It enables circle detection with a desig-
nated radius and various properties such as sensitivity and
edge threshold. This flexibility allows the readjustment of
parameters to identify bubbles with diverse boundary prop-
erties (Wang, Forbes, &Forbes, 2022). Figure 3 illustrates
an example of the image analysis steps.
Test Set Up
The 2-phase tests with the use of air and water were
conducted with the use of synthetic process water, con-
tinuously recycled from a 200-litre water tank. The pro-
cess water was prepared on-site by adding the necessary
amounts of salts to the water tank. The water pump’s high
recirculation flow ensured the salts’ rapid and efficient dis-
solution. Throughout all the tests, the water flow to the
unit remained constant at 2 lpm. The airflow was adjusted
using an inbuilt rotameter, ranging from 0.05 lpm to 0.5
lpm. This translated to superficial gas velocity (Jg) values
ranging from 0.02 to 0.17 s–1. This range was chosen based
on previous published work on the operation of fluidised
bed flotation, where it was shown to provide optimum gas
distribution throughout the fluidised bed (Zanin, Chan, &
Skinner, 2021). It is important to note that this range dif-
fers significantly from that used by Cappuccitti and Nesset
(2010), with a small overlap between the two.
Figure 3. Image processing steps with the developed MATLAB code

Previous Page Next Page

Extracted Text (may have errors)

XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 2863
Kruttschnitt Mineral Research Centre (JKMRC). It is com-
prised of two transparent columns separated by a porous
sintered disc. The air and fluidised water are introduced
in the lower column (or plenum) below the sintered disc.
The presence of the sintered disc enables uniform fluidisa-
tion and bubble distribution in the top chamber. For more
details regarding the design of this equipment, please refer
to (Verster, Awatey, Forbes et al., 2023).
Bubble Size Analysis
All the bubble size measurements were conducted using a
ParticleView ™ V19 (Mettler Toledo). The probe comprises
a waterproof tube with a mounted camera and a process-
ing block. The probe camera has a viewing area measur-
ing 1300 × 890 μm. Online images could be observed and
recorded using the iC PVM ™ software (Wang, Forbes, &
Forbes, 2022). Throughout the tests, images were captured
at one frame every two seconds. This rate is sufficient to
ensure that no bubbles are recorded twice.
The image analysis was conducted using in-house
MATLAB ® code developed for this purpose. The code
employs the circular Hough transform on binary images
to identify bubbles. It enables circle detection with a desig-
nated radius and various properties such as sensitivity and
edge threshold. This flexibility allows the readjustment of
parameters to identify bubbles with diverse boundary prop-
erties (Wang, Forbes, &Forbes, 2022). Figure 3 illustrates
an example of the image analysis steps.
Test Set Up
The 2-phase tests with the use of air and water were
conducted with the use of synthetic process water, con-
tinuously recycled from a 200-litre water tank. The pro-
cess water was prepared on-site by adding the necessary
amounts of salts to the water tank. The water pump’s high
recirculation flow ensured the salts’ rapid and efficient dis-
solution. Throughout all the tests, the water flow to the
unit remained constant at 2 lpm. The airflow was adjusted
using an inbuilt rotameter, ranging from 0.05 lpm to 0.5
lpm. This translated to superficial gas velocity (Jg) values
ranging from 0.02 to 0.17 s–1. This range was chosen based
on previous published work on the operation of fluidised
bed flotation, where it was shown to provide optimum gas
distribution throughout the fluidised bed (Zanin, Chan, &
Skinner, 2021). It is important to note that this range dif-
fers significantly from that used by Cappuccitti and Nesset
(2010), with a small overlap between the two.
Figure 3. Image processing steps with the developed MATLAB code

Help

Destination page number Search scope Search Text

loading

2862 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
to comprehend how alterations in hydrodynamic factors,
specifically superficial gas velocity (Jg), impact the bub-
ble size (D32) and critical coalescence concentration. The
work also aims to establish how these factors subsequently
influence bubble surface area flux (Sb) within the novel
hydrodynamic environment of fluidised bed flotation. The
study explores how CCC values and minimum bubble size
change as hydrodynamic conditions are varied compared
to the established relationships observed in conventional
flotation cells.
Ultimately, the goal is to gain insights to optimise
frother dosage and understand its subsequent impact on
bubble size.
Through these objectives, the study aims to enhance
our understanding of the intricate dynamics governing
the fluidised bed flotation system, providing a foundation
for potential performance optimisation strategies of the
process.
MATERIALS AND EXPERIMENTAL
METHODS
Frothers
The current work used pure methyl isobutyl carbinol
(MIBC). MIBC was evaluated at the concentration range
between 10 and 400 ppm. The reagent was sourced from
Sigma Aldrich.
Process Water
All the tests were conducted using synthetic process water.
The ion content was synthesised to match the water chem-
istry of the Mount Isa Mines copper concentrator (Forbes,
Brito e Abreu, Tungpalan et al., 2022). The electrolyte
composition is summarised in Table 1.
Fluidised Bed Flotation
All the tests were conducted in two-phase conditions using
air and water only. The test work was performed using
a semi-batch small-scale fluidised bed flotation device
(JKHFmini), custom-designed and built at the Julius
Figure 2. Conceptual relationship between hydrodynamic and chemical parameters in controlling bubble coalescence in
flotation processes
Table 1. Composition of the synthetic process water
Ion Name Total Ion Concentration, ppm
K+ 120
Cl– 1110
Ca2+ 570
Mg2+ 340
Na+ 430
SO4 2– 2220
CO
3
2– 24

Previous Page Next Page

Extracted Text (may have errors)

2862 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
to comprehend how alterations in hydrodynamic factors,
specifically superficial gas velocity (Jg), impact the bub-
ble size (D32) and critical coalescence concentration. The
work also aims to establish how these factors subsequently
influence bubble surface area flux (Sb) within the novel
hydrodynamic environment of fluidised bed flotation. The
study explores how CCC values and minimum bubble size
change as hydrodynamic conditions are varied compared
to the established relationships observed in conventional
flotation cells.
Ultimately, the goal is to gain insights to optimise
frother dosage and understand its subsequent impact on
bubble size.
Through these objectives, the study aims to enhance
our understanding of the intricate dynamics governing
the fluidised bed flotation system, providing a foundation
for potential performance optimisation strategies of the
process.
MATERIALS AND EXPERIMENTAL
METHODS
Frothers
The current work used pure methyl isobutyl carbinol
(MIBC). MIBC was evaluated at the concentration range
between 10 and 400 ppm. The reagent was sourced from
Sigma Aldrich.
Process Water
All the tests were conducted using synthetic process water.
The ion content was synthesised to match the water chem-
istry of the Mount Isa Mines copper concentrator (Forbes,
Brito e Abreu, Tungpalan et al., 2022). The electrolyte
composition is summarised in Table 1.
Fluidised Bed Flotation
All the tests were conducted in two-phase conditions using
air and water only. The test work was performed using
a semi-batch small-scale fluidised bed flotation device
(JKHFmini), custom-designed and built at the Julius
Figure 2. Conceptual relationship between hydrodynamic and chemical parameters in controlling bubble coalescence in
flotation processes
Table 1. Composition of the synthetic process water
Ion Name Total Ion Concentration, ppm
K+ 120
Cl– 1110
Ca2+ 570
Mg2+ 340
Na+ 430
SO4 2– 2220
CO
3
2– 24

2864 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
MIBC dosage varied between 10 and 400 ppm for each
test. Approximately 300 images were captured for every
test. Figure 4 illustrates the setup of the JKHFmini with
the PVM probe.
In the JKHFmini, the bubble generation mechanism
consists of an air sparger and a sintered disc. The sintered
disc prevents solids from moving from the top chamber of
the cylinder to the plenum at the bottom, which can lead to
the blockage of the air-sparger pores (Verster et al., 2023).
Additionally, the sintered disc has a direct influence on the
bubble formation process. However, the disc itself acts as
an additional source of bubble coalescence, thereby influ-
encing the overall bubble size. The minimum theoretical
bubble size is influenced by the size of the disc’s pores size.
RESULTS
Effect of the frother dosage on bubble diameter
Figure 5 illustrates the bubble size as a function of the
MIBC dosage for the tested airflow conditions. It can be
seen that the bubble size decreases with an increase in the
frother concentration across all superficial gas velocities.
For each value of Jg, the bubble size reaches a plateau at
the minimum D32, corresponding to the CCC value. The
concentration at which the CCC is reached increases with
increasing Jg. Moreover, the minimum bubble size obtained
at CCC increases with increasing Jg.
The D32 data displays a high level of scatter, indicated
by wide error bars. For this reason, the exact values of CCC
can be hard to determine. However, CCC values can be
roughly estimated to lie between 100 and 200 ppm. At
the CCC values, the Sauter mean diameter of the bubbles
ranged between 0.37 to 0.45. These values are commensu-
rable to the bubble size values presented by K. Demir, A.J.
Morrison, C. Evans et al. (2023). However, bubble size in
any flotation device is strongly influenced by the bubble
generation mechanism. Therefore, measured bubble sizes
should not be compared directly.
Effect of frother dosage on bubble surface area flux
Figure 6 shows the relationship between the bubble sur-
face area flux (Sb) and the frother dosage across the tested
airflow conditions. Across all superficial gas velocities, Sb
increases as the frother concentration is raised. Additionally,
Sb significantly increases with an increase in superficial gas
velocity. Regardless of the specific superficial gas velocity
(Jg) values, Sb reaches a plateau, indicating the attainment
of the maximum Sb value. The Sb value at which the plateau
is reached increases with higher Jg. Further analysis of the
data to take Sb values into account results in a smoother
Figure 4. An experimental setup was done with JKFHmini and a PVM probe

Previous Page Next Page

Extracted Text (may have errors)

2864 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
MIBC dosage varied between 10 and 400 ppm for each
test. Approximately 300 images were captured for every
test. Figure 4 illustrates the setup of the JKHFmini with
the PVM probe.
In the JKHFmini, the bubble generation mechanism
consists of an air sparger and a sintered disc. The sintered
disc prevents solids from moving from the top chamber of
the cylinder to the plenum at the bottom, which can lead to
the blockage of the air-sparger pores (Verster et al., 2023).
Additionally, the sintered disc has a direct influence on the
bubble formation process. However, the disc itself acts as
an additional source of bubble coalescence, thereby influ-
encing the overall bubble size. The minimum theoretical
bubble size is influenced by the size of the disc’s pores size.
RESULTS
Effect of the frother dosage on bubble diameter
Figure 5 illustrates the bubble size as a function of the
MIBC dosage for the tested airflow conditions. It can be
seen that the bubble size decreases with an increase in the
frother concentration across all superficial gas velocities.
For each value of Jg, the bubble size reaches a plateau at
the minimum D32, corresponding to the CCC value. The
concentration at which the CCC is reached increases with
increasing Jg. Moreover, the minimum bubble size obtained
at CCC increases with increasing Jg.
The D32 data displays a high level of scatter, indicated
by wide error bars. For this reason, the exact values of CCC
can be hard to determine. However, CCC values can be
roughly estimated to lie between 100 and 200 ppm. At
the CCC values, the Sauter mean diameter of the bubbles
ranged between 0.37 to 0.45. These values are commensu-
rable to the bubble size values presented by K. Demir, A.J.
Morrison, C. Evans et al. (2023). However, bubble size in
any flotation device is strongly influenced by the bubble
generation mechanism. Therefore, measured bubble sizes
should not be compared directly.
Effect of frother dosage on bubble surface area flux
Figure 6 shows the relationship between the bubble sur-
face area flux (Sb) and the frother dosage across the tested
airflow conditions. Across all superficial gas velocities, Sb
increases as the frother concentration is raised. Additionally,
Sb significantly increases with an increase in superficial gas
velocity. Regardless of the specific superficial gas velocity
(Jg) values, Sb reaches a plateau, indicating the attainment
of the maximum Sb value. The Sb value at which the plateau
is reached increases with higher Jg. Further analysis of the
data to take Sb values into account results in a smoother
Figure 4. An experimental setup was done with JKFHmini and a PVM probe