XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 1883
indicator of a well-fitted model this finding is supported
by additional research. Subsequently, the correlation plots
displayed in Figure 6 shows the strong agreement between
the predicted and actual experimental recoveries.
Response Surface Plots and Analysis
The plots showing the individual effects of the various input
factors for the recovery of Ni from the rougher tailings are
presented in Figure 7. The individual effect plots for Cu
is also depicted in Figure A2. Considering the assumption
that all other factors remain constant, these one-factor plots
examine the overall effect of a particular process condition
on the response variables. This analysis of the results shows
that increasing the grind time increased the leaching recov-
ery to an optimum time of 5 mins. This could be due to
the metal distribution in the fine fraction as observed in the
grindability tests (refer to Figure 3). Conversely, increas-
ing the particle size further led to gradual decrease in the
recovery. Theoretically, increasing the grind time further
could lead to the liberation of other gangue minerals that
compete with the valuable metals for the leaching reagents
(Amara et al. “The Influence of Integrated Gravity Circuit
on the Efficiency of Gold Extraction at a Carbon-in-Leach
Plant,” 2022). Similarly, increasing the sulfuric acid con-
centration had no positive effect on leaching recoveries
after 1 M. This is expected as, typically, increasing the
lixiviant concentration enhances recovery by increasing
the available reactive ions for dissolution (Kumari et al.,
2024). However, in sulfidic minerals, these ions are locked
and need a higher oxidizing environment. Conversely, the
Table 6. The analysis of variance for Ni recovery (%)
Source Sum of Squares DF Mean Square F-Value p-Value
Model 14280.2 20 714.01 288.24 0.0001 Significant
Grind time: n1 120.3 1 120.28 48.56 0.0001
Catalyst dosage:
n
2
4667.7 1 4667.71 1884.33 0.0001
H
2 SO
4 concentration: n
3
1352.7 1 1352.69 546.07 0.0001
Leaching time: n4 795.7 1 795.74 321.23 0.0001
Leaching temp.:
n5
3330.6 1 3330.64 1344.56 0.0001
n
1
2 34.2 1 34.23 13.82 0.0030
n
2
2 186.9 1 186.89 75.44 0.0001
n32 84.9 1 84.93 34.28 0.0001
n42 14.2 1 14.23 5.74 0.035
n
5
2 608.1 1 608.13 245.5 0.0001
n
1n 2 351.3 1 351.28 141.81 0.0001
n1n3 41.8 1 41.76 16.86 0.002
n1n4 12.8 1 12.83 5.18 0.044
n
1 n
5 43.5 1 43.46 17.54 0.002
n
2 n
3 1214.7 1 1214.7 490.37 0.0001
n2n4 352.6 1 352.59 142.34 0.0001
n2n5 867.2 1 867.16 350.07 0.0001
n
3 n
4 2.5 1 2.49 1 0.338
n
3 n
5 224.5 1 224.48 90.62 0.0001
n4n5 0.8 1 0.76 0.31 0.509
Residual 27.2 11 2.48
Lack of fit 23.9 6 3.98 15.93 0.35 Not significant
Pure error 3.4 5 0.67
Corrected total 14307.5 31
Table 7. The fit statistics of the experimental data and regression model
Std. Dev. R2 Adjusted R2 Predicted R2
Ni recovery (%)157 0.9981 0.9946 0.8519
Cu recovery (%)1.91 0.9983 0.9952 0.9314
indicator of a well-fitted model this finding is supported
by additional research. Subsequently, the correlation plots
displayed in Figure 6 shows the strong agreement between
the predicted and actual experimental recoveries.
Response Surface Plots and Analysis
The plots showing the individual effects of the various input
factors for the recovery of Ni from the rougher tailings are
presented in Figure 7. The individual effect plots for Cu
is also depicted in Figure A2. Considering the assumption
that all other factors remain constant, these one-factor plots
examine the overall effect of a particular process condition
on the response variables. This analysis of the results shows
that increasing the grind time increased the leaching recov-
ery to an optimum time of 5 mins. This could be due to
the metal distribution in the fine fraction as observed in the
grindability tests (refer to Figure 3). Conversely, increas-
ing the particle size further led to gradual decrease in the
recovery. Theoretically, increasing the grind time further
could lead to the liberation of other gangue minerals that
compete with the valuable metals for the leaching reagents
(Amara et al. “The Influence of Integrated Gravity Circuit
on the Efficiency of Gold Extraction at a Carbon-in-Leach
Plant,” 2022). Similarly, increasing the sulfuric acid con-
centration had no positive effect on leaching recoveries
after 1 M. This is expected as, typically, increasing the
lixiviant concentration enhances recovery by increasing
the available reactive ions for dissolution (Kumari et al.,
2024). However, in sulfidic minerals, these ions are locked
and need a higher oxidizing environment. Conversely, the
Table 6. The analysis of variance for Ni recovery (%)
Source Sum of Squares DF Mean Square F-Value p-Value
Model 14280.2 20 714.01 288.24 0.0001 Significant
Grind time: n1 120.3 1 120.28 48.56 0.0001
Catalyst dosage:
n
2
4667.7 1 4667.71 1884.33 0.0001
H
2 SO
4 concentration: n
3
1352.7 1 1352.69 546.07 0.0001
Leaching time: n4 795.7 1 795.74 321.23 0.0001
Leaching temp.:
n5
3330.6 1 3330.64 1344.56 0.0001
n
1
2 34.2 1 34.23 13.82 0.0030
n
2
2 186.9 1 186.89 75.44 0.0001
n32 84.9 1 84.93 34.28 0.0001
n42 14.2 1 14.23 5.74 0.035
n
5
2 608.1 1 608.13 245.5 0.0001
n
1n 2 351.3 1 351.28 141.81 0.0001
n1n3 41.8 1 41.76 16.86 0.002
n1n4 12.8 1 12.83 5.18 0.044
n
1 n
5 43.5 1 43.46 17.54 0.002
n
2 n
3 1214.7 1 1214.7 490.37 0.0001
n2n4 352.6 1 352.59 142.34 0.0001
n2n5 867.2 1 867.16 350.07 0.0001
n
3 n
4 2.5 1 2.49 1 0.338
n
3 n
5 224.5 1 224.48 90.62 0.0001
n4n5 0.8 1 0.76 0.31 0.509
Residual 27.2 11 2.48
Lack of fit 23.9 6 3.98 15.93 0.35 Not significant
Pure error 3.4 5 0.67
Corrected total 14307.5 31
Table 7. The fit statistics of the experimental data and regression model
Std. Dev. R2 Adjusted R2 Predicted R2
Ni recovery (%)157 0.9981 0.9946 0.8519
Cu recovery (%)1.91 0.9983 0.9952 0.9314