1846 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
most of the iron oxide have been converted into metallic
iron. However, since the magnetite peak at around 35.5°
was very similar to the fayalite peak (Figure 5b). Therefore,
the fayalite peak at about 24.8° and 31.6° was observed
(Figure 5d~e), it was found that when the temperature
exceeded 800 °C, the characteristic peak of fayalite began
to appear and the diffraction peak was strongest at 900 °C,
indicating that a large amount of FeO and SiO2 formed
fayalite at this temperature. This phenomenon explained
why the metallization rate no longer increases and the fer-
rous content increased as the temperature rises above 800
°C, it was difficult for the newly generated fayalite to be
further reduced by hydrogen.
Figure 6 illustrated the XRD patterns at different times.
The XRD patterns in Figure 6(a) revealed the presence of
iron diffraction peaks at 5 min, gradually intensifying from
5 to 20 min and the changes were no longer apparent after-
ward. The magnetite diffraction peak was most prominent
at 2.5 min, this time all hematite was converted to magne-
tite, and then gradually weakened and was no longer evi-
dent after 20 min. While the diffraction peak of wüstite
was generated at 2.5 min and increased to the maximum at
5 min, and then gradually decreased, indicating that in the
early stage of the reaction, the generation of magnetite was
dominated, and it continued to be reduced to wüstite with
the increase of time, while wüstite was further converted to
metallic iron. It was noteworthy that, as can be seen from
the Figure 6, magnetite, wüstite, and metallic iron were
present simultaneously from 2.5 to 20 min, which meant
that the magnetite and wüstite in the ore undergo simulta-
neous reduction. After 20 min, the magnetite and wüstite
diffraction peaks were no longer apparent. And as known
from Figure 6(b), the characteristic peaks of fayalite pro-
duced during the reaction existed after 5 min, indicating
that at the later stage of the reaction, the form of ferrous
iron produced may be mainly in the form of fayalite, while
partly in the form of magnetite, wüstite and other minerals.
Therefore, it was difficult to be continued to be reduced.
Figure 7 depicted the XRD diffraction peak patterns at dif-
ferent concentrations. It was observed that at a concentra-
tion of 10%, iron mainly existed in the form of wüstite and
magnetite, with only a small amount of metallic iron gen-
erated. With the increase in hydrogen concentration, when
the hydrogen concentration reached 20%, the diffraction
peaks of metallic iron appeared and strengthened, until the
intensity of metallic iron diffraction peaks stabilized after
exceeding 50%. At 10% and 20% of hydrogen concentra-
tion, the wüstite diffraction peaks were more pronounced
and gradually weakened with the increase in concentration,
disappearing at concentrations above 40%. This indicated
that the majority of hematite in the mineral required a
concentration of over 50% to be completely reduced to
metallic iron. Meanwhile, the reaction process at different
concentrations was similar to that at different time states,
with the coexistence of magnetite, wüstite and fayalite, and
a small peak of magnetite was still present at concentrations
above 40%, Which may indicate that after a large amount
of metallic iron was produced, the unreacted magnetite had
been wrapped up, making it difficult to continue the reduc-
tion reaction.
SEM-EDS Analysis
To explore the microstructural changes in the roasted sam-
ple throughout the reduction process, we prepared both
powder particle and cross-section samples. The morphol-
ogy of the micro-regions was examined using SEM-EDS.
Figure 8 illustrated the surface microstructural changes
and EDS analysis results of the reduced products at differ-
ent temperatures. At 600°C, the particle surface appeared
smooth, with no evident generation of metallic iron. The
EDS spectrum (Point 1) indicated a relatively high oxy-
gen content. Combined with the results of the XRD tests,
this illustrated that even at the mineral surface, there were
still some iron minerals in the form of magnetite that have
not been completely reduced to metallic iron. At 700°C,
the oxygen content was significantly reduced (Point 2).
This indicated that hematite had completely reacted into
wüstite or metallic iron. Additionally, the protrusions were
observed, and the metallic iron was covered the ore surface
in a “striped” or “fine-line” pattern (Figure 8b). By 800°C,
as can be seen from Figure 8c, the surface was entirely com-
posed of metallic iron, transitioning from a “striped” or
“fine-line” pattern to interconnected and thicker structures,
with some forming a “honeycomb” or “plate-like” pattern.
Upon reaching 900°C, the “plate-like” and “honeycomb”
structures became more pronounced, yet the purity of
metallic iron (Point 4) showed no significant change com-
pared to 800°C (Point 3)
The cross-sectional microstructures of the reduced
samples at different temperatures were illustrated in
Figure 9. At 600°C, no metallic iron was generated in the
interior of the ore, with a small amount present on the sur-
face. Metallic iron and wüstite (Point 1,2) were generated
at the edges, while the interior was primarily composed of
magnetite. Upon increasing the temperature to 700°C, it
was clear that the content of oxygen in the periphery was
much lower (Point 4). As we approached the core, the oxy-
gen mass fraction in the iron minerals gradually increased.
In the most central region, a needle-like structure was
observed (Point 5), possibly indicating a small amount
most of the iron oxide have been converted into metallic
iron. However, since the magnetite peak at around 35.5°
was very similar to the fayalite peak (Figure 5b). Therefore,
the fayalite peak at about 24.8° and 31.6° was observed
(Figure 5d~e), it was found that when the temperature
exceeded 800 °C, the characteristic peak of fayalite began
to appear and the diffraction peak was strongest at 900 °C,
indicating that a large amount of FeO and SiO2 formed
fayalite at this temperature. This phenomenon explained
why the metallization rate no longer increases and the fer-
rous content increased as the temperature rises above 800
°C, it was difficult for the newly generated fayalite to be
further reduced by hydrogen.
Figure 6 illustrated the XRD patterns at different times.
The XRD patterns in Figure 6(a) revealed the presence of
iron diffraction peaks at 5 min, gradually intensifying from
5 to 20 min and the changes were no longer apparent after-
ward. The magnetite diffraction peak was most prominent
at 2.5 min, this time all hematite was converted to magne-
tite, and then gradually weakened and was no longer evi-
dent after 20 min. While the diffraction peak of wüstite
was generated at 2.5 min and increased to the maximum at
5 min, and then gradually decreased, indicating that in the
early stage of the reaction, the generation of magnetite was
dominated, and it continued to be reduced to wüstite with
the increase of time, while wüstite was further converted to
metallic iron. It was noteworthy that, as can be seen from
the Figure 6, magnetite, wüstite, and metallic iron were
present simultaneously from 2.5 to 20 min, which meant
that the magnetite and wüstite in the ore undergo simulta-
neous reduction. After 20 min, the magnetite and wüstite
diffraction peaks were no longer apparent. And as known
from Figure 6(b), the characteristic peaks of fayalite pro-
duced during the reaction existed after 5 min, indicating
that at the later stage of the reaction, the form of ferrous
iron produced may be mainly in the form of fayalite, while
partly in the form of magnetite, wüstite and other minerals.
Therefore, it was difficult to be continued to be reduced.
Figure 7 depicted the XRD diffraction peak patterns at dif-
ferent concentrations. It was observed that at a concentra-
tion of 10%, iron mainly existed in the form of wüstite and
magnetite, with only a small amount of metallic iron gen-
erated. With the increase in hydrogen concentration, when
the hydrogen concentration reached 20%, the diffraction
peaks of metallic iron appeared and strengthened, until the
intensity of metallic iron diffraction peaks stabilized after
exceeding 50%. At 10% and 20% of hydrogen concentra-
tion, the wüstite diffraction peaks were more pronounced
and gradually weakened with the increase in concentration,
disappearing at concentrations above 40%. This indicated
that the majority of hematite in the mineral required a
concentration of over 50% to be completely reduced to
metallic iron. Meanwhile, the reaction process at different
concentrations was similar to that at different time states,
with the coexistence of magnetite, wüstite and fayalite, and
a small peak of magnetite was still present at concentrations
above 40%, Which may indicate that after a large amount
of metallic iron was produced, the unreacted magnetite had
been wrapped up, making it difficult to continue the reduc-
tion reaction.
SEM-EDS Analysis
To explore the microstructural changes in the roasted sam-
ple throughout the reduction process, we prepared both
powder particle and cross-section samples. The morphol-
ogy of the micro-regions was examined using SEM-EDS.
Figure 8 illustrated the surface microstructural changes
and EDS analysis results of the reduced products at differ-
ent temperatures. At 600°C, the particle surface appeared
smooth, with no evident generation of metallic iron. The
EDS spectrum (Point 1) indicated a relatively high oxy-
gen content. Combined with the results of the XRD tests,
this illustrated that even at the mineral surface, there were
still some iron minerals in the form of magnetite that have
not been completely reduced to metallic iron. At 700°C,
the oxygen content was significantly reduced (Point 2).
This indicated that hematite had completely reacted into
wüstite or metallic iron. Additionally, the protrusions were
observed, and the metallic iron was covered the ore surface
in a “striped” or “fine-line” pattern (Figure 8b). By 800°C,
as can be seen from Figure 8c, the surface was entirely com-
posed of metallic iron, transitioning from a “striped” or
“fine-line” pattern to interconnected and thicker structures,
with some forming a “honeycomb” or “plate-like” pattern.
Upon reaching 900°C, the “plate-like” and “honeycomb”
structures became more pronounced, yet the purity of
metallic iron (Point 4) showed no significant change com-
pared to 800°C (Point 3)
The cross-sectional microstructures of the reduced
samples at different temperatures were illustrated in
Figure 9. At 600°C, no metallic iron was generated in the
interior of the ore, with a small amount present on the sur-
face. Metallic iron and wüstite (Point 1,2) were generated
at the edges, while the interior was primarily composed of
magnetite. Upon increasing the temperature to 700°C, it
was clear that the content of oxygen in the periphery was
much lower (Point 4). As we approached the core, the oxy-
gen mass fraction in the iron minerals gradually increased.
In the most central region, a needle-like structure was
observed (Point 5), possibly indicating a small amount