1860 XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3
min, and the suanite and forsterite were formed, respec-
tively. As the reduction proceeded, the intensity of metallic
iron diffraction peaks increased continuously. Finally, only
a small amount of wüstite was not reduced, which limited
the further improvement of the metallization degree.
Microstructural Characterization
Microscopic Morphology
Figure 10 and Figure 11. illustrate the SEM and EDS results
at different temperatures. Combined with Figures 10 (a–f),
the bright white areas (Spots 2 and 4) were metallic iron
with a multitude of micropores, while the light gray regions
(Spots 1 and 3) were unreacted dense magnetite. Hydrogen
could easily penetrate the microporous iron layer to reach
the magnetite reaction interface, and increasing the tem-
perature promoted the metallization degree. Besides, the
dark gray areas (Spots 5 and 6) in Figure 10 (e) were dense
and unreacted ludwigite, which verified the XRD results.
However, when the reduction temperature was
enhanced to 600 °C (Figures 10 (g–i)), the outer layer of
wüstite (spot 8) was encapsulated by dense iron (spot 9),
and the local reduction conditions may be deteriorated by
the dense iron layer, resulting in a decrease in metallization
degree. The porous iron (Figure 10 (h)) may be obtained by
one-step reduction of magnetite when the temperature was
insufficient at the initiation of the reaction, as the hydro-
gen reduction of iron oxides was endothermic (Kim et al.,
2013). At 650 °C (Figures 10 (j–l)), the wüstite (spot 11)
was entirely covered by dense iron (spot 12), and only the
outermost layer was reduced. Once a dense metallic iron
layer was formed, the limiting (slowest) step in the iron
oxide reduction process was the diffusion of solid-state oxy-
gen through the dense layer, significantly deteriorating the
kinetics conditions (Ding et al., 2017 Hagi, 1994 Takada
and Adachi, 1986). Thus, the dense iron seriously prevents
the reduction, causing a significant drop in the metalliza-
tion degree at 600–650 °C. Besides, the ludwigite (spots 7
and 13) was not reduced by hydrogen in this temperature
range.
According to Figures 11 (a) and (c), only the outer-
most layer of the wüstite (spot 1) was reduced to metal-
lic iron (spot 2), which was similar to the observation at
650 °C. However, the reduction reaction occurred between
ludwigite and hydrogen. Figure 11 (b) shows the reduced
ludwigite particle. Based on Spots 3 and 4, the O content
decreased significantly after hydrogen reduction, while the
Fe and Mg contents increased. The mass ratio of O to Mg
(20.41: 11.29 =1.81) gradually approached the theoreti-
cal value of suanite (O: Mg= 1.67). The dense ludwigite
became porous, possibly a mixture of metallic iron, wüstite
and suanite. Therefore, combined with XRD and SEM-
EDS results, the start temperature for hydrogen reduction
of ludwigite was determined to be about 700 °C, which
was rarely mentioned in previous studies (Fu et al., 2018
Han et al., 2016 Li et al., 2014 Wang et al., 2018 You et
al., 2022).
At 750–800°C, the unreduced wüstite (e.g., spot
5) particles were encapsulated by the dense iron layer
(Figures 11(d–h)). In previous studies, the morphology
of iron products reduced by the wüstite was classified as
porous iron, dense iron wrapped in porous wüstite and
dense iron wrapped in dense wüstite, which correlated with
the relative rate of the gas-solid chemical reaction and solid
mass transfer process (Farren et al., 1990 John et al., 1984
Turkdogan and Vinters, 1972). In this study, the structure
of the wüstite and metallic iron (reduced by wüstite) was
dense. Notably, many cracks were observed in the regions
reduced to metallic iron (spots 6 and 8). These cracks were
Figure 9. The XRD patterns of each product at different reduction times: (a) 550 °C (b) 800 °C
min, and the suanite and forsterite were formed, respec-
tively. As the reduction proceeded, the intensity of metallic
iron diffraction peaks increased continuously. Finally, only
a small amount of wüstite was not reduced, which limited
the further improvement of the metallization degree.
Microstructural Characterization
Microscopic Morphology
Figure 10 and Figure 11. illustrate the SEM and EDS results
at different temperatures. Combined with Figures 10 (a–f),
the bright white areas (Spots 2 and 4) were metallic iron
with a multitude of micropores, while the light gray regions
(Spots 1 and 3) were unreacted dense magnetite. Hydrogen
could easily penetrate the microporous iron layer to reach
the magnetite reaction interface, and increasing the tem-
perature promoted the metallization degree. Besides, the
dark gray areas (Spots 5 and 6) in Figure 10 (e) were dense
and unreacted ludwigite, which verified the XRD results.
However, when the reduction temperature was
enhanced to 600 °C (Figures 10 (g–i)), the outer layer of
wüstite (spot 8) was encapsulated by dense iron (spot 9),
and the local reduction conditions may be deteriorated by
the dense iron layer, resulting in a decrease in metallization
degree. The porous iron (Figure 10 (h)) may be obtained by
one-step reduction of magnetite when the temperature was
insufficient at the initiation of the reaction, as the hydro-
gen reduction of iron oxides was endothermic (Kim et al.,
2013). At 650 °C (Figures 10 (j–l)), the wüstite (spot 11)
was entirely covered by dense iron (spot 12), and only the
outermost layer was reduced. Once a dense metallic iron
layer was formed, the limiting (slowest) step in the iron
oxide reduction process was the diffusion of solid-state oxy-
gen through the dense layer, significantly deteriorating the
kinetics conditions (Ding et al., 2017 Hagi, 1994 Takada
and Adachi, 1986). Thus, the dense iron seriously prevents
the reduction, causing a significant drop in the metalliza-
tion degree at 600–650 °C. Besides, the ludwigite (spots 7
and 13) was not reduced by hydrogen in this temperature
range.
According to Figures 11 (a) and (c), only the outer-
most layer of the wüstite (spot 1) was reduced to metal-
lic iron (spot 2), which was similar to the observation at
650 °C. However, the reduction reaction occurred between
ludwigite and hydrogen. Figure 11 (b) shows the reduced
ludwigite particle. Based on Spots 3 and 4, the O content
decreased significantly after hydrogen reduction, while the
Fe and Mg contents increased. The mass ratio of O to Mg
(20.41: 11.29 =1.81) gradually approached the theoreti-
cal value of suanite (O: Mg= 1.67). The dense ludwigite
became porous, possibly a mixture of metallic iron, wüstite
and suanite. Therefore, combined with XRD and SEM-
EDS results, the start temperature for hydrogen reduction
of ludwigite was determined to be about 700 °C, which
was rarely mentioned in previous studies (Fu et al., 2018
Han et al., 2016 Li et al., 2014 Wang et al., 2018 You et
al., 2022).
At 750–800°C, the unreduced wüstite (e.g., spot
5) particles were encapsulated by the dense iron layer
(Figures 11(d–h)). In previous studies, the morphology
of iron products reduced by the wüstite was classified as
porous iron, dense iron wrapped in porous wüstite and
dense iron wrapped in dense wüstite, which correlated with
the relative rate of the gas-solid chemical reaction and solid
mass transfer process (Farren et al., 1990 John et al., 1984
Turkdogan and Vinters, 1972). In this study, the structure
of the wüstite and metallic iron (reduced by wüstite) was
dense. Notably, many cracks were observed in the regions
reduced to metallic iron (spots 6 and 8). These cracks were
Figure 9. The XRD patterns of each product at different reduction times: (a) 550 °C (b) 800 °C