XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 2017
handheld analyzer manufactured by Thermo Scientific in
Waltham, USA. The spectrometer is equipped with a 50
kV Ag-anode X-ray tube and a silicon drift detector (SDD).
To ensure accuracy, two measurements were conducted for
each sample, and the average value was corrected utiliz-
ing an external calibration curve based on the compari-
son between ICP-MS (Inductively Coupled Plasma Mass
Spectrometry) and XRF (X-ray Fluorescence) data. MA-N
standard from the international working group provided
by the Centre de recherche pétrographique et géochimique
(Vandoeuvre-les-nancy, France) has been used.
Design of Experiment
A two-parameter Central Composite Design (CCD)
incorporating 5 central points was generated using the
JMP ® statistical software by SAS Institute. The Design
of Experiments (DOE) specifically focuses on investigat-
ing the impact of rotation speed and fluidization pressure
on the separation process. The coordinates of the central
point are 150 G’s and 5.5 PSI. High levels of rotation speed
and fluidization pressure are set at 200 G’s and 7.75 PSI,
respectively. Conversely, low levels are established at 100
G’s and 3.25 PSI. These settings are summarized in Table 1.
The axial value of the CCD was designated as 2, ensuring
that each axial point is equidistant from the corner point,
which, in turn, is equidistant from the central point. This
implies that the 2D design encompasses 5 levels on each of
its axes, with these five points evenly distributed between
the two extreme values of both parameters. In total, the
design comprises 13 experimental points, featuring mini-
mum and maximum values for rotation speed of 50 and
250 G’s and minimum and maximum values for fluidiza-
tion water pressure of 1 and 10 PSI. The parameters of solid
flowrate and pulp percent solid have been kept constant at
30 kg/h and 30 wt.%, respectively. In this study, the Falcon
concentrator functions as a rougher, with recovery being
the primary target. However, achieving the optimal grade
for a given recovery is also a secondary objective.
The set of 13 experiments corresponding to the Central
Composite Design (CCD) has been combined with an
additional 19 exploratory trials to enhance the dataset.
These exploratory trials were part of the initial design of
experiments but were not centered within the appropriate
zone of the two-dimensional exploratory space. We believe
that their inclusion contributes to a better understanding of
the main impacts of the parameters and responses, resulting
in a more robust model that covers a larger portion of the
experimental design.
Figure 1. Particle size distribution of the ground product (Falcon feed)
Table 1. Definition and levels of the parameters of the central composite design of experiment
Factors Symbol
Levels
Coded Variables Low (-1) Center (0) High (+1)
Rotary speed (Gs) G 100 150t 200 x1= (G-150)/50
Fluidization pressure (PSI) F 3.25 5.5 7.75 x
2 =(F-5.5)/2.25
handheld analyzer manufactured by Thermo Scientific in
Waltham, USA. The spectrometer is equipped with a 50
kV Ag-anode X-ray tube and a silicon drift detector (SDD).
To ensure accuracy, two measurements were conducted for
each sample, and the average value was corrected utiliz-
ing an external calibration curve based on the compari-
son between ICP-MS (Inductively Coupled Plasma Mass
Spectrometry) and XRF (X-ray Fluorescence) data. MA-N
standard from the international working group provided
by the Centre de recherche pétrographique et géochimique
(Vandoeuvre-les-nancy, France) has been used.
Design of Experiment
A two-parameter Central Composite Design (CCD)
incorporating 5 central points was generated using the
JMP ® statistical software by SAS Institute. The Design
of Experiments (DOE) specifically focuses on investigat-
ing the impact of rotation speed and fluidization pressure
on the separation process. The coordinates of the central
point are 150 G’s and 5.5 PSI. High levels of rotation speed
and fluidization pressure are set at 200 G’s and 7.75 PSI,
respectively. Conversely, low levels are established at 100
G’s and 3.25 PSI. These settings are summarized in Table 1.
The axial value of the CCD was designated as 2, ensuring
that each axial point is equidistant from the corner point,
which, in turn, is equidistant from the central point. This
implies that the 2D design encompasses 5 levels on each of
its axes, with these five points evenly distributed between
the two extreme values of both parameters. In total, the
design comprises 13 experimental points, featuring mini-
mum and maximum values for rotation speed of 50 and
250 G’s and minimum and maximum values for fluidiza-
tion water pressure of 1 and 10 PSI. The parameters of solid
flowrate and pulp percent solid have been kept constant at
30 kg/h and 30 wt.%, respectively. In this study, the Falcon
concentrator functions as a rougher, with recovery being
the primary target. However, achieving the optimal grade
for a given recovery is also a secondary objective.
The set of 13 experiments corresponding to the Central
Composite Design (CCD) has been combined with an
additional 19 exploratory trials to enhance the dataset.
These exploratory trials were part of the initial design of
experiments but were not centered within the appropriate
zone of the two-dimensional exploratory space. We believe
that their inclusion contributes to a better understanding of
the main impacts of the parameters and responses, resulting
in a more robust model that covers a larger portion of the
experimental design.
Figure 1. Particle size distribution of the ground product (Falcon feed)
Table 1. Definition and levels of the parameters of the central composite design of experiment
Factors Symbol
Levels
Coded Variables Low (-1) Center (0) High (+1)
Rotary speed (Gs) G 100 150t 200 x1= (G-150)/50
Fluidization pressure (PSI) F 3.25 5.5 7.75 x
2 =(F-5.5)/2.25