XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 1315
studies on the recovery of scandium from that precipitate
suggested that its processing cost would be significant. As
a result, L3 evaluated adding a scandium specific circuit
upstream of the iron extraction circuit. Bench scale test
work was performed on High Density Leaching HCl-PLS
using a 20 v% D2EHPA diluted with kerosene organic
phase at a O:A ratio of 1:10. The bench scale extraction
data summary is presented as Figure 8. It should be noted
that smoothed lines between the data points are presented
as a support for ease of reading, and that it should not be
interpreted as representing a continuity in the dataset. The
Fe assay at the 6th contact was not determined due to an
operator error during the sample preparation process. The
extraction value reflects the extraction during a specified
contact in the succession. Negative extractions represent
stripping of previously loaded element from the organic to
the PLS.
The test work revealed that D2EHPA’s scandium
extraction was sufficiently preferential that it also displaced
all other elements from the organic phase. Thorium’s
extraction potential was for its part in the same range as
titanium and niobium, with a HCl-PLS concentration four
times greater. Yttrium was used to understand the behavior
of REEs in this circuit, having more affinity for D2EHPA
than our heaviest REE product, dysprosium. According to
this series of experiments, thorium is sufficiently favored
over all iron that a selective D2EHPA thorium, titanium
and niobium extraction circuit is possible. Thus, a two-
stage D2EHPA extraction circuit was added to the circuit.
The first circuit operates a very low organic to aqueous ratio
and aims at recovering the majority of the scandium as a
high purity extract while the second extraction circuit aims
are scavenging any residual scandium while recovering all
the thorium present in the HCl-PLS.
Bench scale stripping experiments on Th-loaded
organic identified 2M sulfuric acid as the optimum strip-
ping acid for thorium. L3 evaluated HCl, H2SO4 and
HNO3 from 0.01 M to 6 M. The initial operation of the
thorium extraction circuit presented as Figure 9 revealed
a buildup of titanium correlated with a gradual reduction
in thorium extraction. Building on work by Verbaan et al.,
L3 evaluated different combinations of sulfuric acid and
hydrogen peroxide solution. At day 9 of the operating
sequence presented as Figure 9, the thorium strip acid solu-
tion was changed to 2.5 M sulfuric acid and 3 v% hydrogen
peroxide. This new solution resulted in a significant strip-
ping of the titanium from the organic and a correlated full
extraction of the thorium from the thorium extraction PLS
solution.
PRIMARY SOLVENT EXTRACTION
FLOWSHEET PERFORMANCE
Primary Solvent Extraction Flowsheet
The primary solvent extraction flowsheet block flow dia-
gram is presented as Figure 10.
Scandium Recovery Circuit
The demonstration scandium recovery unit is comprised of
four stages of extraction and a manual bench scale scrub-
bing and stripping stage.
Figure 6. Effect of strip solution acidity and activity on the stripping efficiency of DG6
studies on the recovery of scandium from that precipitate
suggested that its processing cost would be significant. As
a result, L3 evaluated adding a scandium specific circuit
upstream of the iron extraction circuit. Bench scale test
work was performed on High Density Leaching HCl-PLS
using a 20 v% D2EHPA diluted with kerosene organic
phase at a O:A ratio of 1:10. The bench scale extraction
data summary is presented as Figure 8. It should be noted
that smoothed lines between the data points are presented
as a support for ease of reading, and that it should not be
interpreted as representing a continuity in the dataset. The
Fe assay at the 6th contact was not determined due to an
operator error during the sample preparation process. The
extraction value reflects the extraction during a specified
contact in the succession. Negative extractions represent
stripping of previously loaded element from the organic to
the PLS.
The test work revealed that D2EHPA’s scandium
extraction was sufficiently preferential that it also displaced
all other elements from the organic phase. Thorium’s
extraction potential was for its part in the same range as
titanium and niobium, with a HCl-PLS concentration four
times greater. Yttrium was used to understand the behavior
of REEs in this circuit, having more affinity for D2EHPA
than our heaviest REE product, dysprosium. According to
this series of experiments, thorium is sufficiently favored
over all iron that a selective D2EHPA thorium, titanium
and niobium extraction circuit is possible. Thus, a two-
stage D2EHPA extraction circuit was added to the circuit.
The first circuit operates a very low organic to aqueous ratio
and aims at recovering the majority of the scandium as a
high purity extract while the second extraction circuit aims
are scavenging any residual scandium while recovering all
the thorium present in the HCl-PLS.
Bench scale stripping experiments on Th-loaded
organic identified 2M sulfuric acid as the optimum strip-
ping acid for thorium. L3 evaluated HCl, H2SO4 and
HNO3 from 0.01 M to 6 M. The initial operation of the
thorium extraction circuit presented as Figure 9 revealed
a buildup of titanium correlated with a gradual reduction
in thorium extraction. Building on work by Verbaan et al.,
L3 evaluated different combinations of sulfuric acid and
hydrogen peroxide solution. At day 9 of the operating
sequence presented as Figure 9, the thorium strip acid solu-
tion was changed to 2.5 M sulfuric acid and 3 v% hydrogen
peroxide. This new solution resulted in a significant strip-
ping of the titanium from the organic and a correlated full
extraction of the thorium from the thorium extraction PLS
solution.
PRIMARY SOLVENT EXTRACTION
FLOWSHEET PERFORMANCE
Primary Solvent Extraction Flowsheet
The primary solvent extraction flowsheet block flow dia-
gram is presented as Figure 10.
Scandium Recovery Circuit
The demonstration scandium recovery unit is comprised of
four stages of extraction and a manual bench scale scrub-
bing and stripping stage.
Figure 6. Effect of strip solution acidity and activity on the stripping efficiency of DG6