1934 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
particle size strongly affects the quality of the concentrate.
The bad performance at P90 equal to 47 µm is attributed to
the bad performance for the technique applied for this fine
size range. The recovery, especially the magnetite one, has a
higher drop below P90 of 115 µm.
Then, grinding tests were performed in the pre-con-
centrate to evaluate the energy consumption and produce
material for further concentration. Before the grinding
tests, the sample was screened using a Sinex AT450 and
the oversize was crushed at a lab roll crusher. The opera-
tion was performed in closed circuit using control screens
in a process containing 5 cycles (the respective oversize frac-
tions were crushed and screened using the apertures 2.5,
1.8, 1.05 and 0.85 mm two times. The Bond Work Index
determination, (Austin and Luckie (1973) Rowland and
McIvor (2008)), was performed in the pre-concentrate. The
grinding energy is 12.2 kWh/t. The particle size distribu-
tions of the feeding and product are presented at Figure 3
and exhibited D90 of 2,932 mm and 132 µm, respectively.
The D80 of the product was 113 µm, so the screen of 105
µm was chosen to control the time required to grind the
material to produce mass for concentration tests. After 37
minutes, 80% of the material was passing in the control
screen and test was over.
The concentration, aiming the silicates removal, was
performed in the Outotec wet drum wet low-intensity
magnetic separator model WD(20)111-15. The solids
content in the feed and feeding rate were fixed at 30% in
weight and 1,5 l/min. The variable magnet setting, allowed
for 10 different positions. In position 1, the magnet is
Figure 1. Mode of liberation (area-grade) vs size fraction
particle size strongly affects the quality of the concentrate.
The bad performance at P90 equal to 47 µm is attributed to
the bad performance for the technique applied for this fine
size range. The recovery, especially the magnetite one, has a
higher drop below P90 of 115 µm.
Then, grinding tests were performed in the pre-con-
centrate to evaluate the energy consumption and produce
material for further concentration. Before the grinding
tests, the sample was screened using a Sinex AT450 and
the oversize was crushed at a lab roll crusher. The opera-
tion was performed in closed circuit using control screens
in a process containing 5 cycles (the respective oversize frac-
tions were crushed and screened using the apertures 2.5,
1.8, 1.05 and 0.85 mm two times. The Bond Work Index
determination, (Austin and Luckie (1973) Rowland and
McIvor (2008)), was performed in the pre-concentrate. The
grinding energy is 12.2 kWh/t. The particle size distribu-
tions of the feeding and product are presented at Figure 3
and exhibited D90 of 2,932 mm and 132 µm, respectively.
The D80 of the product was 113 µm, so the screen of 105
µm was chosen to control the time required to grind the
material to produce mass for concentration tests. After 37
minutes, 80% of the material was passing in the control
screen and test was over.
The concentration, aiming the silicates removal, was
performed in the Outotec wet drum wet low-intensity
magnetic separator model WD(20)111-15. The solids
content in the feed and feeding rate were fixed at 30% in
weight and 1,5 l/min. The variable magnet setting, allowed
for 10 different positions. In position 1, the magnet is
Figure 1. Mode of liberation (area-grade) vs size fraction
