XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 1233
• TREO+Y recovery appears to be largely a function
of mass yield, with iron recovery largely following
REEs.
Figure 11 presents a regression of product mass yield
versus TREO+Y recovery for all flotation tests undertaken
on WHIMS magnetics.
A moderate correlation with an r2 of 0.75 was deter-
mined which supports the conclusion that the geochemical
similarities of allanite and hastingsite strongly influence the
mass yield-TREO recovery-TREO grade responses, more
so than the type of collector or modifiers used. Reverse flo-
tation tests (blue dots) generally produced results above the
fitted line with most of the direct float test (green dots)
below the line, indicating the former is more effective in
achieving a degree of preferential gangue rejection.
To provide further evidence of the difficulty in separat-
ing iron from REOs, Figure 12 presents a plot of TREO+Y
recovery versus Fe recovery for the same two data sets. The
fitted data had a higher r2 of 0.92, suggesting iron and
REEs are difficult to separate in this mineral system. This is
in part due to Fe being part of the allanite crystal matrix but
also a function of similar surface chemistry with hastingsite.
Based on this analysis, it was considered unlikely that a
conventional flotation solution would be found for separat-
ing allanite from hastingsite in the short term and alterna-
tive approaches are necessary to exploit other differences in
the two minerals.
Electrostatic Separation
Given the difficulty in achieving further enrichment with
flotation, alternative separation methods were sought.
Drawing parallels with the mineral sands industry where
contamination with iron silicates can be problematic,
research undertaken on an Egyptian beach monazite
resource evaluated separation of REE-bearing monazite
from gangue minerals using a Carpco HP167 high tension
roll separator (Moustafa and Abdelfattah, 2010) with a feed
temperature of 160 °C. A high purity monazite product
(97% w/w) was recovered to the non-conductor stream
was achieved whilst the majority of non-opaque miner-
als reported to the conductor fraction. Laxmi et al (2012)
tested titaniferous place material using a high tension roll,
Figure 12. WHIMS magnetics flotation TREO+Y recovery vs. Fe recovery
• TREO+Y recovery appears to be largely a function
of mass yield, with iron recovery largely following
REEs.
Figure 11 presents a regression of product mass yield
versus TREO+Y recovery for all flotation tests undertaken
on WHIMS magnetics.
A moderate correlation with an r2 of 0.75 was deter-
mined which supports the conclusion that the geochemical
similarities of allanite and hastingsite strongly influence the
mass yield-TREO recovery-TREO grade responses, more
so than the type of collector or modifiers used. Reverse flo-
tation tests (blue dots) generally produced results above the
fitted line with most of the direct float test (green dots)
below the line, indicating the former is more effective in
achieving a degree of preferential gangue rejection.
To provide further evidence of the difficulty in separat-
ing iron from REOs, Figure 12 presents a plot of TREO+Y
recovery versus Fe recovery for the same two data sets. The
fitted data had a higher r2 of 0.92, suggesting iron and
REEs are difficult to separate in this mineral system. This is
in part due to Fe being part of the allanite crystal matrix but
also a function of similar surface chemistry with hastingsite.
Based on this analysis, it was considered unlikely that a
conventional flotation solution would be found for separat-
ing allanite from hastingsite in the short term and alterna-
tive approaches are necessary to exploit other differences in
the two minerals.
Electrostatic Separation
Given the difficulty in achieving further enrichment with
flotation, alternative separation methods were sought.
Drawing parallels with the mineral sands industry where
contamination with iron silicates can be problematic,
research undertaken on an Egyptian beach monazite
resource evaluated separation of REE-bearing monazite
from gangue minerals using a Carpco HP167 high tension
roll separator (Moustafa and Abdelfattah, 2010) with a feed
temperature of 160 °C. A high purity monazite product
(97% w/w) was recovered to the non-conductor stream
was achieved whilst the majority of non-opaque miner-
als reported to the conductor fraction. Laxmi et al (2012)
tested titaniferous place material using a high tension roll,
Figure 12. WHIMS magnetics flotation TREO+Y recovery vs. Fe recovery