1030 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
mineral grade
of species i ***
*
V V V
V
…
i i j j n n
i i t t t
t
=+++(1)
where Vi and ρi are is the volume and the density of species
i, respectively and of species Vj, …, Vn and ρj, …, ρn are
the volume and density of the other species present in the
ore.
Surface liberation was quantified in 3D by creating a
distance map for each particle. The surface was extracted
by only considering the voxels of the outer shell. Then, the
fraction of this surface that was occupied by each com-
ponent was calculated by counting the voxels that corre-
sponded to each component and dividing it by the total
particle’s surface following equation (2):
surface liberation
of species i S S
Si
j j n
=+++S (2)
where Si is the surface corresponding to component i, and
Sj, …, Sn are the surfacves that correspond to all the other
components on the surface.
It is important to stress that the main purpose of this
study is to propose an alternative interpretation of surface
liberation data, which is irrespective of the material being
analysed and the method used to quantify surface libera-
tion. Therefore, any type of ore could have been used and
iron ore was selected randomly and only for convenience.
Similar situation is the use of micro-CT for this analysis,
which was used to minimise errors in the measurements.
RESULTS
Grade Variability at the Particle Scale
A preliminary assessment of iron grade was performed for
calibration purposes and to analyse iron content at the par-
ticle level. Even when XRF provides a measurement for the
overall iron content in the sample, it does not capture the
variability within the sample.
It was possible to measure the iron grade for each of
the scanned particles using the methodology described
in Image Processing and Methodology. Figure 3 shows the
results, where it can be seen that the iron content not only
varies widely for the different sizes, but also within each size
fraction. This behaviour is expected and is due to the break-
age pattern produced by a given comminution method.
Hence, by changing the comminution device, the vertical
trends in this graph will either compact of become sparser,
with the latter being the most advantageous for concentra-
tion purposes.
Lastly, the average iron content in the sample was
quantified from Figure 3, obtaining an average iron con-
tent of 57.7%, which is slightly lower than that measured
using XRF (i.e., 58.5% of Fe). However, it must be taken
into account that this measurement is a combination of
3D measurements performed on hundreds of particles
and, most probably, a more accurate description of the
Figure 3. Iron grade as a function of the particle size
mineral grade
of species i ***
*
V V V
V
…
i i j j n n
i i t t t
t
=+++(1)
where Vi and ρi are is the volume and the density of species
i, respectively and of species Vj, …, Vn and ρj, …, ρn are
the volume and density of the other species present in the
ore.
Surface liberation was quantified in 3D by creating a
distance map for each particle. The surface was extracted
by only considering the voxels of the outer shell. Then, the
fraction of this surface that was occupied by each com-
ponent was calculated by counting the voxels that corre-
sponded to each component and dividing it by the total
particle’s surface following equation (2):
surface liberation
of species i S S
Si
j j n
=+++S (2)
where Si is the surface corresponding to component i, and
Sj, …, Sn are the surfacves that correspond to all the other
components on the surface.
It is important to stress that the main purpose of this
study is to propose an alternative interpretation of surface
liberation data, which is irrespective of the material being
analysed and the method used to quantify surface libera-
tion. Therefore, any type of ore could have been used and
iron ore was selected randomly and only for convenience.
Similar situation is the use of micro-CT for this analysis,
which was used to minimise errors in the measurements.
RESULTS
Grade Variability at the Particle Scale
A preliminary assessment of iron grade was performed for
calibration purposes and to analyse iron content at the par-
ticle level. Even when XRF provides a measurement for the
overall iron content in the sample, it does not capture the
variability within the sample.
It was possible to measure the iron grade for each of
the scanned particles using the methodology described
in Image Processing and Methodology. Figure 3 shows the
results, where it can be seen that the iron content not only
varies widely for the different sizes, but also within each size
fraction. This behaviour is expected and is due to the break-
age pattern produced by a given comminution method.
Hence, by changing the comminution device, the vertical
trends in this graph will either compact of become sparser,
with the latter being the most advantageous for concentra-
tion purposes.
Lastly, the average iron content in the sample was
quantified from Figure 3, obtaining an average iron con-
tent of 57.7%, which is slightly lower than that measured
using XRF (i.e., 58.5% of Fe). However, it must be taken
into account that this measurement is a combination of
3D measurements performed on hundreds of particles
and, most probably, a more accurate description of the
Figure 3. Iron grade as a function of the particle size