XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 719
effects on the initial settling rate response were flocculant
dose and pH, with values of 2.1 and 1.47, respectively.
Following closely is the quadratic interaction of pH, con-
tributing with a value of 2.4. Hence, pH emerges as the
most significant contributor to the ISR response. In terms
of the water recovery response, the primary factors influ-
encing the recovery rates are the mixing, followed by the
flocculant dose, with coefficient estimate values of 3.2 and
1.83, respectively. The pH variable exerts a less pronounced
effect, with a coefficient estimate value of 1.09. The math-
ematical equation modelling the recycled water turbidity
highlights that the main factors exerting substantial influ-
ence on turbidity are the quadratic interaction of pH, the
interaction effect of flocculant dose and pH, and the pH.
With a less pronounced impact, the quadratic interac-
tion effect of mixing rate also contributes to the turbidity.
The coefficient estimate values for these factors are 3.86,
2.51, 1.81, and 1.30, respectively. The regression models
were employed to predict the optimal values of the studied
responses through graphical optimization within the soft-
ware’s module. This approach generates contour plots based
on response intervals, delineating comfortable areas for
achieving optimal outcomes (By Russell R. Barton, 1999).
While useful, this method can become intricate when deal-
ing with a large number of factors under study (Abraham
et al., 2009). The response behaviour of ISR, WR, and
WT concerning simultaneous changes in the three studied
factors and their interaction effects is depicted in the con-
tour plots presented in Figure 5. In each contour plot, the
other two factors were held constant at their center level. As
clearly illustrated in Figure 5, there is a notable curvature in
the contour plots, indicating interdependence among these
four factors. In other words, the interaction effects between
the studied factors were significant.
Interpretation of Response Surfaces and Contour Plots
Figure 6 (a, b, c) illustrates the interaction effects of pH and
flocculant dose, mixing rate and flocculant dose, as well as
mixing rate and pH for the ISR response. ISR is influenced
mainly by floc properties (size, density, structure, etc.) and
slurry viscosity, with the former primarily impacted by
flocculation operating conditions. It can be clearly seen
that high pH values exceeding 8 and elevated mixing rates
yield in the highest settling velocities. The peak of ISRs is
observed at alkaline pH levels and very high flocculant dos-
ages which is confirmed by a previous study conducted on
the dewatering of oil mine tailings (Sworska et al., 2000). A
detailed analysis reveals that, under alkaline pH, increasing
flocculant dose enhances settling velocities, while increas-
ing mixing speed reduces settling velocities. This could
be explained by the fact that, at high flocculant dosages,
the formation of medium to large flocs leads to significant
particle aggregation, resulting in the highest settling rates.
Conversely, high mixing rates break down the formed flocs,
generating smaller flocs, hence, the lowest settling veloci-
ties (Jarvis et al., 2005 Ofori et al., 2011 OruÇ &Sabah,
2006 Yu et al., 2011). This study aimed to preliminarily
investigate the influence of flocculant dose (FD), pH, and
mixing rate (MR) on ISR. However, a more detailed exami-
nation of the impact of flocculation conditions on ISR is
essential to understand the characteristics of the formed
flocs. This can be achieved by establishing the relationships
between floc properties and flocculation conditions, as well
as between floc properties and dewatering effectiveness.
In Figure 6 (a', b', c'), the interaction effects of pH
and flocculant dose, mixing rate and flocculant dose, as
well as mixing rate and pH for water recovery response
are depicted. Notably, the figure illustrates the consider-
able impact of flocculant dose on water recovery. At high
Figure 5. Predicted vs. actual values plot for responses: (a) ISR (b) W.R and (c) water turbidity
effects on the initial settling rate response were flocculant
dose and pH, with values of 2.1 and 1.47, respectively.
Following closely is the quadratic interaction of pH, con-
tributing with a value of 2.4. Hence, pH emerges as the
most significant contributor to the ISR response. In terms
of the water recovery response, the primary factors influ-
encing the recovery rates are the mixing, followed by the
flocculant dose, with coefficient estimate values of 3.2 and
1.83, respectively. The pH variable exerts a less pronounced
effect, with a coefficient estimate value of 1.09. The math-
ematical equation modelling the recycled water turbidity
highlights that the main factors exerting substantial influ-
ence on turbidity are the quadratic interaction of pH, the
interaction effect of flocculant dose and pH, and the pH.
With a less pronounced impact, the quadratic interac-
tion effect of mixing rate also contributes to the turbidity.
The coefficient estimate values for these factors are 3.86,
2.51, 1.81, and 1.30, respectively. The regression models
were employed to predict the optimal values of the studied
responses through graphical optimization within the soft-
ware’s module. This approach generates contour plots based
on response intervals, delineating comfortable areas for
achieving optimal outcomes (By Russell R. Barton, 1999).
While useful, this method can become intricate when deal-
ing with a large number of factors under study (Abraham
et al., 2009). The response behaviour of ISR, WR, and
WT concerning simultaneous changes in the three studied
factors and their interaction effects is depicted in the con-
tour plots presented in Figure 5. In each contour plot, the
other two factors were held constant at their center level. As
clearly illustrated in Figure 5, there is a notable curvature in
the contour plots, indicating interdependence among these
four factors. In other words, the interaction effects between
the studied factors were significant.
Interpretation of Response Surfaces and Contour Plots
Figure 6 (a, b, c) illustrates the interaction effects of pH and
flocculant dose, mixing rate and flocculant dose, as well as
mixing rate and pH for the ISR response. ISR is influenced
mainly by floc properties (size, density, structure, etc.) and
slurry viscosity, with the former primarily impacted by
flocculation operating conditions. It can be clearly seen
that high pH values exceeding 8 and elevated mixing rates
yield in the highest settling velocities. The peak of ISRs is
observed at alkaline pH levels and very high flocculant dos-
ages which is confirmed by a previous study conducted on
the dewatering of oil mine tailings (Sworska et al., 2000). A
detailed analysis reveals that, under alkaline pH, increasing
flocculant dose enhances settling velocities, while increas-
ing mixing speed reduces settling velocities. This could
be explained by the fact that, at high flocculant dosages,
the formation of medium to large flocs leads to significant
particle aggregation, resulting in the highest settling rates.
Conversely, high mixing rates break down the formed flocs,
generating smaller flocs, hence, the lowest settling veloci-
ties (Jarvis et al., 2005 Ofori et al., 2011 OruÇ &Sabah,
2006 Yu et al., 2011). This study aimed to preliminarily
investigate the influence of flocculant dose (FD), pH, and
mixing rate (MR) on ISR. However, a more detailed exami-
nation of the impact of flocculation conditions on ISR is
essential to understand the characteristics of the formed
flocs. This can be achieved by establishing the relationships
between floc properties and flocculation conditions, as well
as between floc properties and dewatering effectiveness.
In Figure 6 (a', b', c'), the interaction effects of pH
and flocculant dose, mixing rate and flocculant dose, as
well as mixing rate and pH for water recovery response
are depicted. Notably, the figure illustrates the consider-
able impact of flocculant dose on water recovery. At high
Figure 5. Predicted vs. actual values plot for responses: (a) ISR (b) W.R and (c) water turbidity