2928 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
reagents that were dosed in the bench cell after precondi-
tioning. The samples were floated in each case by scraping
the froth at 15-second intervals for thirty minutes, the con-
centrates being collected at cumulative intervals of 2, 6, 14
and 30 minutes.
RESULTS AND DISCUSSION
The flotation retention time on the SMC concentrator
amounts to 15, 26 and 36 minutes respectively for the
rougher, rougher scavenger and cleaner banks. Thus, in
order to interpret the bench flotation results more repre-
sentatively in terms of the plant scenario, a scale-down fac-
tor of 1 /2.5 =0.4 was applied. As point of reference, the
bench rougher, rougher scavenger and cleaner kinetics were
therefore taken as 6, 10 and 14 minutes respectively. For
comparison purposes, an arbitrary bench retention time
of 14 minutes was chosen for a tailings retreatment plant
(TRP) that the company might consider, with a rougher
capacity of 36 minutes.
Rougher feed
For the rougher feed, four sets of tests were conducted,
each comprising four kinetic bench tests. Results from the
last two sets for the Rougher feed were averaged and are
reported here as ultimate data sets for this stream.
The profiles for the mass pull-recovery and grade-
recovery relationships for the rougher feed are shown in
Figures 2 and 3 respectively, where a residence time of 6
minutes served as reference. This is represented by the sec-
ond data point on the curves. Note that, rather than the
conventional approach of showing actual figures, these graphs
represent the recovery and grade differences with respect to
the 6-minute data points of the baseline, and thus contain
both negative and positive values. As indicated, the Mach
returned a 4E recovery improvement after 5 passes of 8.8%
compared to the baseline, demonstrating the higher flota-
tion selectivity that is being achieved by preconditioning in
the Reactor. A further increase in the preconditioning time
to 10 passes resulted in a smaller improvement of 5.7%,
whereas a further 5 passes proved detrimental to the float
by resulting in a loss, compared to the baseline, in recov-
ery of 4.7%. These trends are in line with earlier findings
on PGM ores, where a peak is observed in the recovery
response with the number of passes (i.e., conditioning
time), rather than a constant increase. It is also evident that
under these conditions, the mass pull differences with the
baseline were minimal throughout.
As is to be expected from the above results, Figure 3
shows the increased kinetics and grade that were achieved
after 5 and 10 passes of preconditioning in the Mach
Reactor, being most noticeable during the early stages of flo-
tation. Care should be taken with the interpretation of the
results, however, as the calculated head grades of the respec-
tive Rougher feed samples varied considerably from 2.47 g/t
and 2.87 g/t for the base case and 15-pass test respectively,
to 4.0 and 4.26 for the 5 and 10-pass tests. This is a result
-30
-20
-10
0
10
20
30
-10 0 10 20
Mass pull diff. (%)
0 Pass 5 pass 10 pass 15 pass
Figure 2. Mass pull – recovery differential profiles for the Rougher feed (baseline
reference point of 6 minutes)
4E
Recovery
d
iff .
(%)
reagents that were dosed in the bench cell after precondi-
tioning. The samples were floated in each case by scraping
the froth at 15-second intervals for thirty minutes, the con-
centrates being collected at cumulative intervals of 2, 6, 14
and 30 minutes.
RESULTS AND DISCUSSION
The flotation retention time on the SMC concentrator
amounts to 15, 26 and 36 minutes respectively for the
rougher, rougher scavenger and cleaner banks. Thus, in
order to interpret the bench flotation results more repre-
sentatively in terms of the plant scenario, a scale-down fac-
tor of 1 /2.5 =0.4 was applied. As point of reference, the
bench rougher, rougher scavenger and cleaner kinetics were
therefore taken as 6, 10 and 14 minutes respectively. For
comparison purposes, an arbitrary bench retention time
of 14 minutes was chosen for a tailings retreatment plant
(TRP) that the company might consider, with a rougher
capacity of 36 minutes.
Rougher feed
For the rougher feed, four sets of tests were conducted,
each comprising four kinetic bench tests. Results from the
last two sets for the Rougher feed were averaged and are
reported here as ultimate data sets for this stream.
The profiles for the mass pull-recovery and grade-
recovery relationships for the rougher feed are shown in
Figures 2 and 3 respectively, where a residence time of 6
minutes served as reference. This is represented by the sec-
ond data point on the curves. Note that, rather than the
conventional approach of showing actual figures, these graphs
represent the recovery and grade differences with respect to
the 6-minute data points of the baseline, and thus contain
both negative and positive values. As indicated, the Mach
returned a 4E recovery improvement after 5 passes of 8.8%
compared to the baseline, demonstrating the higher flota-
tion selectivity that is being achieved by preconditioning in
the Reactor. A further increase in the preconditioning time
to 10 passes resulted in a smaller improvement of 5.7%,
whereas a further 5 passes proved detrimental to the float
by resulting in a loss, compared to the baseline, in recov-
ery of 4.7%. These trends are in line with earlier findings
on PGM ores, where a peak is observed in the recovery
response with the number of passes (i.e., conditioning
time), rather than a constant increase. It is also evident that
under these conditions, the mass pull differences with the
baseline were minimal throughout.
As is to be expected from the above results, Figure 3
shows the increased kinetics and grade that were achieved
after 5 and 10 passes of preconditioning in the Mach
Reactor, being most noticeable during the early stages of flo-
tation. Care should be taken with the interpretation of the
results, however, as the calculated head grades of the respec-
tive Rougher feed samples varied considerably from 2.47 g/t
and 2.87 g/t for the base case and 15-pass test respectively,
to 4.0 and 4.26 for the 5 and 10-pass tests. This is a result
-30
-20
-10
0
10
20
30
-10 0 10 20
Mass pull diff. (%)
0 Pass 5 pass 10 pass 15 pass
Figure 2. Mass pull – recovery differential profiles for the Rougher feed (baseline
reference point of 6 minutes)
4E
Recovery
d
iff .
(%)