2652 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
The equation used for prediction and surface response
are given below. In order to maximise both the recovery and
the enrichment ratios, same variables that were in agree-
ment with the ANOVA study were kept. In order to maxi-
mise both recovery and ER, variables that were significant
are the dosage of amine, which showed also both quadratic
contribution and interaction with the desliming size. pH
had only a linear contribution on the studied ranges.
..
..
.
R x x
x xdesliming
x x
94 03 1 12 10.23
12 11 10 05
8 24
pH
2
Li O
desliming
2
=++
-+
-
6amine@
6amine@
6amine@
...
..
.
ER x x
x x
x x
3 11 0 14 0 45
0 11 0 10
0 14
min
min
Li O pH
e@ desliming
e@ desliming
2
2
=--
++
-
6a
6a
6amine@
These results could be expected as collector consump-
tion is correlated to the size range floated, with higher col-
lector consumption for fine grained samples (Farrokhpay et
al., 2021). This might explain the interaction effect between
desliming size and collector dosage. One interesting fact is
the low contribution of the pH value over the studied bor-
ders. However from pH values 1.5 to 2.5 the consumption
of acid is well reduced (H+ concentration reduced from
0.03 mol/L to 0.003 mol/L). Therefore, this can reduce the
operational cost of this flotation step.
A better way to look at interaction between the dif-
ferent factors studied is to look at surface responses. They
were plotted for flotation stage only (Figure 3) and for Li2O
recoveries calculated along the whole flowsheet, i.e., by tak-
ing into account the impact of desliming on the lithium
recovery (Figure 4). Along with the surface response of lith-
ium recovery, a plot with isolines of both lithium recover-
ies and lithium enrichment ratios was computed (Figure 3,
Figure 4). By looking at lithium recovery along the stud-
ied work area, a maximum value can be obtained for an
amine dosage of about 250 g/ton. Then the desliming size
of the feed of flotation impacts greatly the lithium recovery.
High desliming size is linked with high lithium recoveries,
whereas by reducing the size up to 10µm can decrease the
Figure 2. Partition curves of the samples used for flotation tests
Table 5. Statistical analysis of the DOE
Response R2 RMSE p-value
Recovery %Li
2 O 0.86 7.1026 0.0005
ER (Li2O) 0.85 0.2111 0.0007
The equation used for prediction and surface response
are given below. In order to maximise both the recovery and
the enrichment ratios, same variables that were in agree-
ment with the ANOVA study were kept. In order to maxi-
mise both recovery and ER, variables that were significant
are the dosage of amine, which showed also both quadratic
contribution and interaction with the desliming size. pH
had only a linear contribution on the studied ranges.
..
..
.
R x x
x xdesliming
x x
94 03 1 12 10.23
12 11 10 05
8 24
pH
2
Li O
desliming
2
=++
-+
-
6amine@
6amine@
6amine@
...
..
.
ER x x
x x
x x
3 11 0 14 0 45
0 11 0 10
0 14
min
min
Li O pH
e@ desliming
e@ desliming
2
2
=--
++
-
6a
6a
6amine@
These results could be expected as collector consump-
tion is correlated to the size range floated, with higher col-
lector consumption for fine grained samples (Farrokhpay et
al., 2021). This might explain the interaction effect between
desliming size and collector dosage. One interesting fact is
the low contribution of the pH value over the studied bor-
ders. However from pH values 1.5 to 2.5 the consumption
of acid is well reduced (H+ concentration reduced from
0.03 mol/L to 0.003 mol/L). Therefore, this can reduce the
operational cost of this flotation step.
A better way to look at interaction between the dif-
ferent factors studied is to look at surface responses. They
were plotted for flotation stage only (Figure 3) and for Li2O
recoveries calculated along the whole flowsheet, i.e., by tak-
ing into account the impact of desliming on the lithium
recovery (Figure 4). Along with the surface response of lith-
ium recovery, a plot with isolines of both lithium recover-
ies and lithium enrichment ratios was computed (Figure 3,
Figure 4). By looking at lithium recovery along the stud-
ied work area, a maximum value can be obtained for an
amine dosage of about 250 g/ton. Then the desliming size
of the feed of flotation impacts greatly the lithium recovery.
High desliming size is linked with high lithium recoveries,
whereas by reducing the size up to 10µm can decrease the
Figure 2. Partition curves of the samples used for flotation tests
Table 5. Statistical analysis of the DOE
Response R2 RMSE p-value
Recovery %Li
2 O 0.86 7.1026 0.0005
ER (Li2O) 0.85 0.2111 0.0007