XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 2717
python script, Stardist, was used to measure the bubble
sizes within the extracted frames. Figure 2 presents example
images of the bubble measurements performed.
The pressure drop data from each run were converted
to gas holdup. Samples of the silica feed were tested with a
Malvern zetasizer 3000 to measure the zeta potential of the
silica at the natural pH and different salt concentrations.
RESULTS AND DISCUSSION
Process Conditions Summary
As previously outlined, considerable effort was made to
ensure consistent feed pulp density and flow rates across
each run. A summary of the run conditions used in this
body of work is presented in Table 1. A gas flux of 1.67 cm/s
was used for all run conditions.
Feed invariability is vital to ensure that valid compari-
sons can be made, and by Table 1 it is shown that the pulp
density was maintained at ~3 wt.% across all runs. Particle
Figure 2. (A) Example frame extracted from slow-motion footage (B) shows the same frame analysed by
the Startdist machine learning python script
Table 1. Summary of experimental conditions
Run
#
NaCl
(M)
Pulp
Density
(wt%)
Feed
(cm/s)
Product
(cm/s)
Tailing
(cm/s)
Wash
(cm/s)
1a 0 3.09±0.02 1.37 0.33 1.05 0
1b 0 2.90±0.00 1.33 0.31 1.04 0
2a 0 3.03±0.05 1.37 0.32 1.86 0.83
2b 0 3.11±0.07 1.32 0.31 1.85 0.83
2c 0 3.00±0.03 1.34 0.34 1.87 0.83
3a 0.2 3.10±0.02 1.38 0.33 1.05 0
3b 0.2 3.06±0.02 1.32 0.29 1.03 0
4a 0.2 3.02±0.01 1.35 0.31 1.83 0.83
4b 0.2 2.89±0.05 1.30 0.29 1.85 0.83
4c 0.2 2.99±0.02 1.33 0.32 1.84 0.83
5a 1 2.90±0.13 1.31 0.31 1.00 0
5b 1 3.05±0.03 1.27 0.29 0.99 0
6a 1 3.08±0.03 1.28 0.30 1.77 0.83
6b 1 2.97±0.02 1.24 0.28 1.78 0.83
python script, Stardist, was used to measure the bubble
sizes within the extracted frames. Figure 2 presents example
images of the bubble measurements performed.
The pressure drop data from each run were converted
to gas holdup. Samples of the silica feed were tested with a
Malvern zetasizer 3000 to measure the zeta potential of the
silica at the natural pH and different salt concentrations.
RESULTS AND DISCUSSION
Process Conditions Summary
As previously outlined, considerable effort was made to
ensure consistent feed pulp density and flow rates across
each run. A summary of the run conditions used in this
body of work is presented in Table 1. A gas flux of 1.67 cm/s
was used for all run conditions.
Feed invariability is vital to ensure that valid compari-
sons can be made, and by Table 1 it is shown that the pulp
density was maintained at ~3 wt.% across all runs. Particle
Figure 2. (A) Example frame extracted from slow-motion footage (B) shows the same frame analysed by
the Startdist machine learning python script
Table 1. Summary of experimental conditions
Run
#
NaCl
(M)
Pulp
Density
(wt%)
Feed
(cm/s)
Product
(cm/s)
Tailing
(cm/s)
Wash
(cm/s)
1a 0 3.09±0.02 1.37 0.33 1.05 0
1b 0 2.90±0.00 1.33 0.31 1.04 0
2a 0 3.03±0.05 1.37 0.32 1.86 0.83
2b 0 3.11±0.07 1.32 0.31 1.85 0.83
2c 0 3.00±0.03 1.34 0.34 1.87 0.83
3a 0.2 3.10±0.02 1.38 0.33 1.05 0
3b 0.2 3.06±0.02 1.32 0.29 1.03 0
4a 0.2 3.02±0.01 1.35 0.31 1.83 0.83
4b 0.2 2.89±0.05 1.30 0.29 1.85 0.83
4c 0.2 2.99±0.02 1.33 0.32 1.84 0.83
5a 1 2.90±0.13 1.31 0.31 1.00 0
5b 1 3.05±0.03 1.27 0.29 0.99 0
6a 1 3.08±0.03 1.28 0.30 1.77 0.83
6b 1 2.97±0.02 1.24 0.28 1.78 0.83