1470 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
Silver Grain in Ore Particles
The main silver mineral in the selected ore sample is acan-
thite (Ag2S). To differentiate acanthite from barite, which
has a similar effective atomic number, and from galena,
which has a similar density, small pieces of acanthite,
galena, and barite were used as references and placed in the
sample containers, about 1 mm diameter packed bed con-
taining a few thousands of particles. The 425 µm × 212 µm
and 104 µm × 74 µm size fractions were used for the 3D
characterization. Once prepped, the samples were scanned
using the Zeiss Xradia 620 system at the voxel size of about
700 nm using a 4× scintillator for magnification and a 2K
× 2K CCD camera as detector.
RESULTS AND DISCUSSIONS
Effect of Fines in 4-inch Columns
Three 4-inch columns were set up to determine the effects
of fines on permeability. The first 4-inch column contained
a similar number of fines (–74 μm) compared to normal
operation (an averaged PSD for the past year), which is
about 10% fines. The second 4-inch column contained no
fines, and the third column contained double the number
of fine particles compared to normal operation, which was
targeted for 20% but measured to be 18.2% by screening.
The bottom sections of 4-inch columns were scanned using
the Zeiss Xradia 620 system at the voxel size of 68 μm,
which was supposed to show more details of the materials
inside columns.
As shown in Figure 3, the connected pore network and
LBM simulated flow channels in the bottom sections of
4-inch columns decrease in the sequence of no fines, regu-
lar fines, and double fines. As listed in Table 1, the perme-
ability in the regular fines column was 790 darcy, slightly
higher than the no fines column, which was 751 darcy. The
double fines column permeability was much lower than
the other columns at 447 darcy. The drastic decrease was
most likely caused by the narrow pore throats in the con-
nected pore network. Narrow pore throats tend to restrict
the flow greatly. The high moisture content confirms that
poor permeability was present in this column. The techni-
cians at the Rochester metallurgical laboratory also noted
that ponding occurred within the first few days of leaching
for the double fines column. The solution had to be turned
off to allow it to disperse before the solution was turned
back on a few days later.
In addition to the rock (coarse particles) and pore
network, another phase of agglomerated fines (including
–74 μm particles but possibly involving particles close to
this size) was found in the images from high-resolution
XCT scans. As shown in Figure 4, the 2D and 3D images
were created to demonstrate the phases present in each col-
umn: pore, rock, and agglomerated fines. In the middle
row of Figure 4, rocks are dark, pores are white, and the
agglomerated fines are shown in a grey color. As seen in
the 3D rendering (Figure 4 bottom row), the double fines
column had many more agglomerated fines than the other
two columns. The no-fines column had no agglomerated
fines present, which was expected because of the lack of
fine material in the column. The regular fines column
has 19.4% volume filled by the agglomerated fines, and
the double fines column has 28.1% volume filled by the
agglomerated fines, which is much more significant.
The uniform PSD (except the fines below 74 μm) for
the three 4-inch columns implied that fine particle move-
ment through pores may have been localized and not neces-
sarily a bulk movement through the different sections, like
from top to bottom following the solution flow direction.
As listed in Table 1, the pore size region between the three
4-inch columns is consistent with each other.
The mean pore size for the double fines column was
slightly larger than the regular fines columns. However,
the mean pore size for the no fines column was about half
of those containing fines. Considering the much smaller
porosity of the double fines column (7.9%), the double
fines column had the least number of pores, but the pores
inside the double column are close to the larger pores in the
two other columns. The visual of flow channels in Figure 3
supports this finding. Because the mean pore size is an
average of all pores in the connected network, the double
fines and regular fines columns have significant amounts
of agglomerated fines to fill their small pores. There is no
agglomerated fines phase for the no fines column to occupy
its pore spaces and resist gravity compression. As a result,
the pores in no fines column were more compressed and
thus smaller. However, the no fines column still has the
largest porosity and permeability.
Cracks in HPGR Crushed Ore Particles
9.5 × 6.4 mm Particles
3D images of two-particle sets crushed by HPGR at the
specific pressures of 5 N/mm2 and 10 N/mm2 in the 9.5 ×
6.4 mm size fraction are shown in Figure 5. This size frac-
tion has pores and some minor close-to-surface cracks for
5N/mm2 particles. However, significant particle damage
is observed in the 10N/mm2 particles. Different than the
close-to-surface cracks in the smaller particles from abra-
sion, the shape of fractures inside the two 10N/mm2 par-
ticles, as shown in Figure 5 (right column, especially the
right top particle), suggests that they are cracks propagated
Silver Grain in Ore Particles
The main silver mineral in the selected ore sample is acan-
thite (Ag2S). To differentiate acanthite from barite, which
has a similar effective atomic number, and from galena,
which has a similar density, small pieces of acanthite,
galena, and barite were used as references and placed in the
sample containers, about 1 mm diameter packed bed con-
taining a few thousands of particles. The 425 µm × 212 µm
and 104 µm × 74 µm size fractions were used for the 3D
characterization. Once prepped, the samples were scanned
using the Zeiss Xradia 620 system at the voxel size of about
700 nm using a 4× scintillator for magnification and a 2K
× 2K CCD camera as detector.
RESULTS AND DISCUSSIONS
Effect of Fines in 4-inch Columns
Three 4-inch columns were set up to determine the effects
of fines on permeability. The first 4-inch column contained
a similar number of fines (–74 μm) compared to normal
operation (an averaged PSD for the past year), which is
about 10% fines. The second 4-inch column contained no
fines, and the third column contained double the number
of fine particles compared to normal operation, which was
targeted for 20% but measured to be 18.2% by screening.
The bottom sections of 4-inch columns were scanned using
the Zeiss Xradia 620 system at the voxel size of 68 μm,
which was supposed to show more details of the materials
inside columns.
As shown in Figure 3, the connected pore network and
LBM simulated flow channels in the bottom sections of
4-inch columns decrease in the sequence of no fines, regu-
lar fines, and double fines. As listed in Table 1, the perme-
ability in the regular fines column was 790 darcy, slightly
higher than the no fines column, which was 751 darcy. The
double fines column permeability was much lower than
the other columns at 447 darcy. The drastic decrease was
most likely caused by the narrow pore throats in the con-
nected pore network. Narrow pore throats tend to restrict
the flow greatly. The high moisture content confirms that
poor permeability was present in this column. The techni-
cians at the Rochester metallurgical laboratory also noted
that ponding occurred within the first few days of leaching
for the double fines column. The solution had to be turned
off to allow it to disperse before the solution was turned
back on a few days later.
In addition to the rock (coarse particles) and pore
network, another phase of agglomerated fines (including
–74 μm particles but possibly involving particles close to
this size) was found in the images from high-resolution
XCT scans. As shown in Figure 4, the 2D and 3D images
were created to demonstrate the phases present in each col-
umn: pore, rock, and agglomerated fines. In the middle
row of Figure 4, rocks are dark, pores are white, and the
agglomerated fines are shown in a grey color. As seen in
the 3D rendering (Figure 4 bottom row), the double fines
column had many more agglomerated fines than the other
two columns. The no-fines column had no agglomerated
fines present, which was expected because of the lack of
fine material in the column. The regular fines column
has 19.4% volume filled by the agglomerated fines, and
the double fines column has 28.1% volume filled by the
agglomerated fines, which is much more significant.
The uniform PSD (except the fines below 74 μm) for
the three 4-inch columns implied that fine particle move-
ment through pores may have been localized and not neces-
sarily a bulk movement through the different sections, like
from top to bottom following the solution flow direction.
As listed in Table 1, the pore size region between the three
4-inch columns is consistent with each other.
The mean pore size for the double fines column was
slightly larger than the regular fines columns. However,
the mean pore size for the no fines column was about half
of those containing fines. Considering the much smaller
porosity of the double fines column (7.9%), the double
fines column had the least number of pores, but the pores
inside the double column are close to the larger pores in the
two other columns. The visual of flow channels in Figure 3
supports this finding. Because the mean pore size is an
average of all pores in the connected network, the double
fines and regular fines columns have significant amounts
of agglomerated fines to fill their small pores. There is no
agglomerated fines phase for the no fines column to occupy
its pore spaces and resist gravity compression. As a result,
the pores in no fines column were more compressed and
thus smaller. However, the no fines column still has the
largest porosity and permeability.
Cracks in HPGR Crushed Ore Particles
9.5 × 6.4 mm Particles
3D images of two-particle sets crushed by HPGR at the
specific pressures of 5 N/mm2 and 10 N/mm2 in the 9.5 ×
6.4 mm size fraction are shown in Figure 5. This size frac-
tion has pores and some minor close-to-surface cracks for
5N/mm2 particles. However, significant particle damage
is observed in the 10N/mm2 particles. Different than the
close-to-surface cracks in the smaller particles from abra-
sion, the shape of fractures inside the two 10N/mm2 par-
ticles, as shown in Figure 5 (right column, especially the
right top particle), suggests that they are cracks propagated