1292 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
The process developed on the micro-pilot scale was
transferred to the pilot plant (Figure 5) which processed
~8 tons of coal ash. Between 20% and 50% of the REEs
were extracted during the leaching step, with a relative con-
tent of ~0.1%-1% (depending on the quality of the feed
material). The liquid-liquid extraction stage produced near-
quantitative yields and was demonstrated to produce REE
concentrates with 15% yield and 50% purity (relative
content).
The pilot plant was exercised over several campaigns to
produce a rare earth oxide concentrate. The initial product
from the plant had a total REE content of ~20 wt.% (on
oxide basis). As detailed below, subsequent modifications
to the cold side and post-processing operations resulted in
increased product purity. The REE product from the pilot
facility contained significant quantities of the valuable Nd,
Y, Sc and HREE content as exemplified by the product
analysis presented in Figure 9.
The ash cake materials that were filtered out follow-
ing the digestion step were functionally qualified for use
as a pozzolanic cement substitute. ASTM C-618 proto-
col was used to qualify the samples using as criterion the
75% of strength activity index (SAI) after 28 days metric.
As shown in Figure 10, five out of six samples (originating
from both ash feedstocks) passed the ASTM C-618 require-
ments. These results indicate that the ash cake is suitable
as a cement substitute and the digestion process does not
greatly impact cement suitability.
To produce additional high purity product samples, a
counter current liquid-liquid extraction (CCLLX) circuit
(Figure 11) was operated in continuous campaigns. These
campaigns produced REE solutions that were further
refined to meet material production and purity goals.
The post-processing steps were developed and PSI
and scaled-up in the pilot plant operations by WWS. The
REY oxide product was generated over multiple produc-
tion runs. Figure 12 outlines the results with the amount of
REY oxide produced shown in blue and the corresponding
REY relative content (defined as the weight of REE in the
sample divided by the total amount of metal in the sample)
shown in red for each REE oxide production run. As shown
in Figure 12, consistent high product purity was obtained
after post processing optimization:
The LLX optimization resulted in an LLX product with
an REE purity consistently in the range of 40–60% relative
content. The post-processing processing steps were demon-
strated to consistently generate REY oxide material (~90%
relative content) from the LLX products. The scaled-up
post processing steps generated the REY oxide concentrates
deliverable materials (22 g of 85% REY content) and 16 g
of 90% REE content.
A commercial plant model (Figure 13) was developed
using the following assumptions: (1) plant located at ash
Figure 9. REE oxides produced in the pilot plant and representative analysis of a product sample
The process developed on the micro-pilot scale was
transferred to the pilot plant (Figure 5) which processed
~8 tons of coal ash. Between 20% and 50% of the REEs
were extracted during the leaching step, with a relative con-
tent of ~0.1%-1% (depending on the quality of the feed
material). The liquid-liquid extraction stage produced near-
quantitative yields and was demonstrated to produce REE
concentrates with 15% yield and 50% purity (relative
content).
The pilot plant was exercised over several campaigns to
produce a rare earth oxide concentrate. The initial product
from the plant had a total REE content of ~20 wt.% (on
oxide basis). As detailed below, subsequent modifications
to the cold side and post-processing operations resulted in
increased product purity. The REE product from the pilot
facility contained significant quantities of the valuable Nd,
Y, Sc and HREE content as exemplified by the product
analysis presented in Figure 9.
The ash cake materials that were filtered out follow-
ing the digestion step were functionally qualified for use
as a pozzolanic cement substitute. ASTM C-618 proto-
col was used to qualify the samples using as criterion the
75% of strength activity index (SAI) after 28 days metric.
As shown in Figure 10, five out of six samples (originating
from both ash feedstocks) passed the ASTM C-618 require-
ments. These results indicate that the ash cake is suitable
as a cement substitute and the digestion process does not
greatly impact cement suitability.
To produce additional high purity product samples, a
counter current liquid-liquid extraction (CCLLX) circuit
(Figure 11) was operated in continuous campaigns. These
campaigns produced REE solutions that were further
refined to meet material production and purity goals.
The post-processing steps were developed and PSI
and scaled-up in the pilot plant operations by WWS. The
REY oxide product was generated over multiple produc-
tion runs. Figure 12 outlines the results with the amount of
REY oxide produced shown in blue and the corresponding
REY relative content (defined as the weight of REE in the
sample divided by the total amount of metal in the sample)
shown in red for each REE oxide production run. As shown
in Figure 12, consistent high product purity was obtained
after post processing optimization:
The LLX optimization resulted in an LLX product with
an REE purity consistently in the range of 40–60% relative
content. The post-processing processing steps were demon-
strated to consistently generate REY oxide material (~90%
relative content) from the LLX products. The scaled-up
post processing steps generated the REY oxide concentrates
deliverable materials (22 g of 85% REY content) and 16 g
of 90% REE content.
A commercial plant model (Figure 13) was developed
using the following assumptions: (1) plant located at ash
Figure 9. REE oxides produced in the pilot plant and representative analysis of a product sample