Journal Information
Share
Share
Download PDF
More article options
Visits
0
Original Article
DOI: 10.1016/j.jmrt.2019.09.028
Open Access
Available online 29 September 2019
Systematic study on separation of Mn and Fe from ferruginous manganese ores by carbothermic reduction roasting process: Phase transformation and morphologies
Visits
0
Lihua Gao, Zhenggen Liu
Corresponding author
liuzg@smm.neu.edu.cn

Corresponding authors.
, Yuzhu Pan, Cong Feng, Mansheng Chu
Corresponding author
chums@smm.neu.edu.cn

Corresponding authors.
, Jue Tang
School of Metallurgy, Northeastern University, Shenyang 110819, China
This item has received
0
Visits

Under a Creative Commons license
Article information
Abstract
Full Text
Bibliography
Download PDF
Statistics
Figures (15)
Show moreShow less
Tables (1)
Table 1. Reaction equations for reduction of Mn–Fe ore based on the following equation.
Abstract

Carbothermic roasting reduction followed by magnetic separation process is reported as an effective technological process to separation and recovery of Fe and Mn from low-grade ferruginous manganese ores (Fe–Mn ores) as an acceptable feed to meet the demands of the developing manganese industry. In this study, the effects of operating variables on the recovery and grade of Fe and Mn are initially investigated during the carbothermic roasting reduction process followed by magnetic separation. Then, the phase transformation and morphologies of spinel (FeyMn1−y)Al2O4, fayalite FeyMn2−ySiO4 and (FeO)x(MnO)1−x are investigated by SEM-EDS, XRD, TG/DTG and thermodynamic analyses. Finally, the stepwise reduction and interfacial reaction of three types of phases are investigated at 700–1100 °C for 10–120 min to clarify the reason for the poor separation of Mn and Fe from Fe–Mn ores. The effect of MnO on the stepwise reduction behavior of Fe2SiO4 and FeAl2O4 is analyzed in detail. Finally, the tight integration of the MnO phase with a metallic Fe–Mn alloy derived from the stepwise reduction of (FeO)x(MnO)1−x are closely associated with the stepwise reduction of the FeyMn2−ySiO4, (FeyMn1−y)Al2O4 and (Fe,Mn)2O3 phases. In addition, the formation mechanism and the interfacial reduction reaction of (FeO)x(MnO)1−x, (FeyMn1−y)Al2O4 and FeyMn2−ySiO4 are systematically analyzed to show the effect of the MnO phase on the stepwise reduction of the Fe2SiO4 and FeAl2O4 phases. Finally, some suggestions were recommended for the carbothermic reduction followed by magnetic separation for utilizing the Fe–Mn ores effectively. The manganese-rich product an acceptable feed containing 53.60 wt.% total Mn with the Mn recovery of 89.38% and mass fraction of Mn/Fe of 5.43 are obtained.

Keywords:
Ferruginous manganese ores
Carbothermic reduction roasting
Phase transformation
Separation
Morphologies
Full Text
1Introduction

Low-grade ferruginous manganese ores (Fe–Mn ores) have received increasing attention as substitutes for high-grade manganese ores, which are one kind of significant manganese resources that can satisfy the demands of the developing manganese industry [1–6]. Unfortunately, Fe–Mn ores with high iron content are unacceptable as feedstocks for smelting ferromanganese alloys. Therefore, many studies have been carried out to upgrade the mass ratio of Mn/Fe (to over 5) in Fe–Mn ores for use as an acceptable feed in manganese alloy production [7–9]. Carbothermic reduction roasting followed by magnetic separation has been considered as an effective technology for the separation and recovery of Mn and Fe from Fe–Mn ores [7–12].

In the abovementioned technological process, which is satisfactory and acceptable for industrial application, carbothermic roasting reduction is the essential procedure for the recovery and separation of Mn and Fe from Fe–Mn ores [1,7,8,10]. However, Mn and Fe cannot be effectively separated from Fe–Mn ores, and approximately 20% of Mn content was presented in magnetic product by carbothermic roasting reduction process [7–10]. Gao et al. [11] found that during magnetic roasting reduction, the Mn grade in the rich-manganese product was as high as 27–34 wt.%; in this study, the efficiency of the upgrading process was characterized using comprehensive full factorial experiments. Cao et al. focused mainly on the optimization of the technological parameters by comprehensive full factorial experiments but did not clarify why the Mn and Fe were not separated well from Fe–Mn ores. Swamy et al. [13] presented that the mass ratio of Mn/Fe in rich-manganese products to 6.2, while the Mn recovery was only 50 wt.%. Thus, their technique cannot be used to effectively separate and recover Fe and Mn from Fe–Mn ores. Liu et al. [1] clarified the underlying reason for the inferior separation efficiency of Mn and Fe and recommended low temperature (below 800 °C) and prolonged reduction time for the magnetic reduction roasting of Fe–Mn ores. The Fe recovery reached 90.6 wt.%, but the Mn recovery only reached 65.3 wt.%. The formation of spinel manganese ferrite MnxFe3−xO4 and the olivine phases of FexMn2−xSiO4 resulted in the poor separation of Fe and Mn from the Fe–Mn ores. In a direct reduction roasting-magnetic separation process, Huang et al. [14,15] the Fe and Mn was still not separated effectively, and the Mn recovery of non-magnetic product was only 78.46%. Few studies in the literature have investigated the underlying reason for the poor separation of Fe and Mn from Fe-Mn ores. This poor separation may be partially caused by the phase composition of the Mn, Fe, Al and Si constituents in the Fe–Mn ores and the isomorphic substitution of Fe and Mn elements in the mineral lattice structure during the natural mineralization of Fe–Mn oxides in Fe–Mn ores [15–25]. In this paper, the essential features of the carbothermic reduction roasting process for Fe–Mn ores are systematically characterized based on observations of phase transformation and intrinsic morphology evolution.

The objective of the present study is to perform a detailed analysis of the underlying reason for the poor Fe–Mn separation based on observations of phase transformation and morphology evolution. First, the effects of the operating variables on the separation and recovery of Mn and Fe from Fe–Mn ores are investigated. The resulting inference is that the poor separation of Fe and Mn may be related to the phase transformation and intrinsic morphology evolution of composite oxides containing Mn, Fe, Al and Si elements during carbothermic reduction. To further investigate this hypothesis, the phase transformations of spinel (FeyMn1−y)Al2O4, fayalite FeyMn2−ySiO4 and (FeO)x(MnO)1−x are examined by SEM-EDS, XRD and thermodynamic analyses. Finally, the stepwise reduction behavior and the interfacial reaction of three types of phases are investigated at 700–1100 °C for 10–120 min to understand the poor separation of Mn and Fe. Recommendations are given for the carbothermic roasting reduction conditions for Fe–Mn ores to obtain credible and reliable experimental results.

2Experimental2.1Raw materials

The Fe–Mn ore samples used in this study were obtained in South Africa. The Fe–Mn ore was a typical high-alumina and low-silicon ferruginous manganese ore that contained 29.50 wt.% TMn (total manganese), 28.28 wt.% TFe (total iron), 3.72 wt.% SiO2 and 7.82 wt.%. Al2O3. The manganese mainly comprised manganite MnO(OH) and bixbyite ((Fe,Mn)2O3), and the iron compounds mainly existed in the form of hematite (Fe2O3) and bixbyite ((Fe,Mn)2O3). The Si, Al and Mg in the Fe–Mn ore, which have an undesirable effect on manganese recovery, were detected in the form of hercynite (Al15.41Mg1.42Fe7.18O32.00) and (Fe, Mn)2SiO4. The reduction coal was obtained from Tianjin province in China. The bitumite coal contained 52.98 wt.% fixed carbon, 40.70 wt.% volatiles, 5.30 wt.% ash and 1.02 wt.% moisture.

2.2Carbothermic roasting reduction process

The schematic diagram of the industrial recovery and separation of Mn and Fe from Fe–Mn ores by carbothermic reduction roasting process followed by magnetic separation is displayed in Fig. 1(a). In this work, The Fe–Mn ores as a raw material were first crushed to a prescribed range of grain sizes and then subjected to a reduction reaction for Mn and Fe oxide in a rotary kiln; the simulated experiment for the carbothermic reduction was conducted in a vertical resistance furnace that is shown in Fig. 1(b). The raw materials were dried at 105 °C for 3 h in a draft-drying cabinet. The Fe–Mn ores and bitumite coal that were used in this study were crushed by a jaw crusher (PE60 × 100 type) to sizes ranging from 8 to 13 mm and below 3 mm, respectively. Approximately 500 g of raw ores and reduction coal were adequately mixed in a prescribed proportion of FC/O. Noted that, the FC/O is the molar ratio of the fixed carbon in coal to the reducible oxygen of iron and manganese in the Fe–Mn ores. Then, the mixtures were charged into a corundum crucible and placed into the vertical resistance furnace. The mixtures were reduced at a setting temperature for a fixed period of time. Before the carbothermic reduction roasting process, the furnace was preheated to the setting temperature at a speed of 10 °C/min. After the reduction roasting process, the reduced samples were cooled to room temperature in an Ar atmosphere. Other detailed descriptions of the carbothermic reduction process was reported in our previous reports [26–28].

Fig. 1.

(a) Schematic diagram of the industrial separation of Mn and Fe from Fe–Mn ore inside a rotary kiln and (b) schematic diagram of experimental equipment for the carbothermic reduction process.

(0.34MB).
2.3Magnetic separation process

The reduced Fe–Mn ores obtained from carbothermic reduction roasting were ground in a 2-MZ centrifugal grinding machine. Next, approximately 20 g of the reduction product powder was separated in a DTCXG-ZN50 Davies magnetic tube at a prescribed magnetic field intensity to facilitate the separation and recovery of Mn and Fe from the Fe–Mn ores. Subsequently, the resulting magnetic product containing an iron-rich phase and a nonmagnetic product containing a MnO-rich phase were filtered and dried in a vacuum drying oven at 100 °C for 4 h at a negative pressure of 0.06 MPa. The Mn- and Fe-grades of the separation products were measured by chemical titration. The chemical analyses were conducted at the Metallurgy Research Institute of Northeastern University, Shenyang, China. Further detailed descriptions of the carbothermic roasting reduction technological process were illustrated in our previous reports [27,28].

2.4Characterization methods2.4.1Instrumental analyses

After the carbothermic roasting reduction process, the main phase identification of the treated Fe–Mn ores was identified using XRD (Almelo, the Netherlands), which was operated under the conditions of copper Kα radiation: Cu Ka, tube current and voltage: 40 mA, 40 kV, scanning range: 5–90° and scanning speed of 0.2°/s.

For the morphology analysis, manganese-rich or iron-rich particles were observed by scanning electron microscopy (SEM, Jena, Germany) with energy dispersive X-ray spectroscopy (EDS). The diffusion coupling method was conducted in conjunction with SEM-EDS to determine the intrinsic morphology evolution during carbothermic roasting reduction.

Thermogravimetric experiments were performed using a Netzsch STA 409 C/CD in an Ar atmosphere (99.99%, volume fraction). The Fe–Mn ores were preground to particle sizes that could pass through a 200 mesh. A sample that weighed approximately 10 mg was heated from room temperature to 1200 °C at a heating rate of 10 °C/min by passing a steady Ar flow (40 mL/min) through the reactor tube.

The particle size distribution (PSD) of the reduced Fe–Mn ore was measured using a laser particle size analyzer (Master size 2000) and corrected with calibration curves to produce the PSD data that are presented in this paper. The equivalent particle sizes (EPS: D10, D20, D40, D60, and D80) and the specific surface areas (SSA) of the reduced Fe–Mn ore were calculated using the software of the laser particle size analyzer.

2.4.2Separation indexes

The following experimental assessment indicators were chosen: the Mn (RMn) and Fe (RFe) recoveries, the mass ratio of Mn to Fe in the nonmagnetic product (Mn/Fe), the total content of Mn (TMn) and Fe (TFe) in the nonmagnetic product, and the metallization ratio (MR) of the reduced product [5]. The efficiency of the carbothermic reduction and the magnetic separation of the Fe–Mn ore were evaluated using the following equation:

where RMn is the recovery of Mn, %; mMn is the mass of the nonmagnetic product, g; MMn is the mass of the reduced product, g; βMn is the Mn grade in the nonmagnetic product, wt.%; and γMn is the Mn grade in the reduced product, wt.%. A similar expression was used to describe how the recovery yield of iron varied with the composition of the magnetic product.

3Results and discussion3.1Results of the carbothermic reduction process-magnetic separation of Fe–Mn ores3.1.1Effect of carbothermic reduction condition on separation indexes of Mn and Fe

The effects of the temperature (a, b), the duration time (c, d) and FC/O (e, f) on the grade and recovery of Mn and Fe from the Fe–Mn ores are investigated in Fig. 2. As shown in Fig. 2(a), with the increase of temperature from 1000 to 1100 °C, the Fe recovery continuously increased from 74.84 to 86.27%, while the Mn grade increased from 49.02 to 53.30% and then tardily decreased to 52.15%. This demonstrates that the higher temperature is not favorable to the separation of Mn and Fe. In Fig. 2(c, d), it can be seen that when the duration time reached 6 h, the Mn and Fe recovery increased from 68.28 to 73.72%; however, the separation indexes of Mn and Fe decreased when the duration time exceeded 6 h. That is, increasing the duration time did not improve the separation and recovery of Mn and Fe from the Fe–Mn ores. From Fig. 2(e, f), it can be seen that the FC/O had little influence on the Fe recovery, and the Mn recovery increased from 70.76 to 78.83% as the FC/O increased from 1.5 to 3.5. In Fig. 2(g, h), it can be seen that the Fe recovery decreased from 86.81 to 77.64%, while the Mn recovery increased from 70.54 to 83.30%, and the particle size of the Fe–Mn ores increased to 13 mm. It is well known that both the stepwise reduction reaction of Fe2O3 to metallic Fe and the stepwise reduction of MnO2 to MnO increase with the duration time and the temperature [1]. However, the change in the Mn recovery was different from that of the Fe recovery, and the variation in the recovery and grade of Mn and Fe are not accordant with the theoretical prediction. Thus, it was deduced that the phase transformation of Mn may have been related to the Fe oxide, that is, a complex phase may have formed that prevented the separation and recovery of Mn and Fe from the Fe–Mn ores during carbothermic roasting reduction.

Fig. 2.

Effect of (a, b) carbothermic reduction temperature; (c, d) time; and (e, f) FC/O and particle size of Fe–Mn ores on Mn and Fe separation indexes from Fe–Mn ores.

(0.98MB).
3.1.2Effect of grinding time and magnetic intensity on separation indexes of Mn and Fe

Fig. 3 displays the effects of the grinding fineness and magnetic intensity field on Fe–Mn separation efficiency. As the magnetic intensity increased, the Fe recovery increased from 81.32 to 93.60%, while the Mn recovery decreased from 85.20 to 71.05%, and both the Mn/Fe and Mn grades concurrently increased. It is obviously found that the recoveries of Mn and Fe were clearly inversely related. It is concluded that the Mn oxide may have compounded with Fe oxide or metallic iron to form a new composite oxide during the carbothermic roasting reduction process. As shown in Fig. 3(c) and (d), as the grinding time increased from 1 to 4 min, the Fe recovery decreased from 84.87 to 71.56%, while the Mn recovery increased from 71.24 to 82.16%, and the Mn grade increased from 46.14 to 53.21%. Thus, prolonging the grinding time improved the separation of Fe and Mn. Fig. 3(e) shows the effect of the grinding time on the particle size distribution of the ground samples, as measured by the particle size analyzer. In Fig. 3(f), the equivalent particle sizes (EPS; D10, D20, D40, D60, and D80) and the specific surface areas (SSA) of the reduced Fe–Mn ores are shown for different grinding times. In this study, the EPS is defined as the particle size distribution when the cumulative passing of the particles reaches a prescribed value [29–33]. The EPS decreased while the SSA increased as the grinding time increased, indicating that the overall fineness of the Fe–Mn ore powder increased with the grinding time. The EPS of the Fe–Mn ore powder and the SSA increased with the grinding time, indicating that fine particles could upgrade the separation and grade of Mn and Fe from the Fe–Mn ores.

Fig. 3.

Effect of (a, b) magnetic intensity and (c, d) grinding time on separation indexes of Mn and Fe, and effect of grinding time on (e) particle size distribution and (f) equivalent particle size and specific surface area.

(0.78MB).
3.2Discussion on the carbothermic reduction behavior of Fe–Mn ores3.2.1Phase transformation of Fe–Mn ores during carbothermic reduction

Fig. 4 displays the XRD patterns of Fe–Mn ore reduced at 1100 °C for the duration time of 5–120 min. As illustrated in Fig. 4(a), the intensity of the diffraction peaks of the Fe-C phase were detected for 10 min, and the Fe diffraction peaks gradually decreased with the duration time from 5 to 120 min. It is observed that the diffraction peaks of the Fe-C group were observed in the diffraction peaks of the standard substance of the Fe-C group in Fig. 6(a). Fig. 4(b) displays the diffraction peaks of (FeO)x(MnO)1−x (1 ≥ x ≥ 0) in the Fe–Mn ores. In particular, it was found that a switching phenomenon for the characteristic diffraction peaks (2 0 0) of (FeO)x(MnO)1−x from 41.802 (FeO, PDF#46-1314) to 40.663 (MnO, PDF#01-0437) was observed as the x-values increased from 1 to 0 as the duration time increased from 5 to 120 min. The x-value decreased as the duration time increased because of the stepwise reduction of (FeO)x(MnO)1−x to Fe and MnO. The intensity of the diffraction peaks of (FeO)x(MnO)1−x remained the same as the duration time increased. In addition, the diffraction peaks of the standard substance of the (FeO)x(MnO)1−x group are also shown in Fig. 6(d). Theoretically, the diffraction intensity peaks of (FeO)x(MnO)1−x in the reduced Fe–Mn ores inevitablly descends as the increase of duration time. On the contrary, the changing rules of x-value in (FeO)x(MnO)1−x were not consistent with the theoretical prediction. It was inferred that the (FeO)x(MnO)1−x phase was continually generated via the phase transformation of an unknown phase or a composite oxide of Fe oxide and Mn oxide during the carbothermic reduction process. Fig. 4(c) displays the XRD patterns of FeyMn2−ySiO4 and the FeyMn1−yAl2O4 group in the Fe–Mn ores. As observed, the diffraction intensity peaks of Mn2SiO4 and FeMnSiO4 gradually increased with the reduction time. The ionic Fe2+ species in fayalite Fe2SiO4 was replaced by ionic Mn2+ to form FeyMn2−ySiO4, which promoted the stepwise reduction of fayalite FeyMn1−yAl2O4 in the Fe–Mn ores because of the similar ionic radii of Mn2+ (0.067 nm) and Fe2+ (0.061 nm) [1,34–37]. The diffraction peaks of the standard substance FeyMn2−ySiO4 are illustrated in Fig. 6(b). Besides, the diffraction intensity peaks of FeyMn1−yAl2O4 group were little changed with the rising of duration time, indicating that spinel FeAl2O4 was gradually transformed into spinel MnAl2O4. However, it was difficult to determine the y-value of the FeyMn1−yAl2O4 group because there was little change in the two-theta degrees of the characteristic diffraction peaks (3 1 1) of FeyMn1−yAl2O4. The diffraction peaks of the standard substance of the FeyMn1−yAl2O4 group are shown in Fig. 6(c). The formation of the FeyMn1−yAl2O4 phase is a novel means of reducing the spinel FeAl2O4 phase via the phase transformation of FeAl2O4/ FeyMn1−yAl2O4/ MnAl2O4.

Fig. 4.

XRD patterns of Fe–Mn ores reduced at 1100 °C for different reduction time.

(0.76MB).
Fig. 6.

XRD patterns of standard substances of (a) FexC group; (b) FeyMn2−ySiO4 group; (c) (FeyMn1−y)Al2O4 group; and (d) (FeO)x(MnO)1−x group.

(0.6MB).
3.2.2Morphological characteristics of Fe–Mn ores during carbothermic reduction process

It is of great importance to study the microstructure of the samples that were reduced at different reduction time, as described in Fig. 7, to reveal the intrinsic morphology evolution and reduction mechanisms during the carbothermic reduction process. The results displayed an irregular surface structure for the primary particles of the Fe–Mn ores and finer reagent grade iron particles in the matrix of the reacted material. Then, as the irregular fragmentation became increasingly apparent, and the initial shape of an iron whisker began to take shape. As the reduction time was increased further, iron particles with different growth orientations of whiskers became interconnected with each other, possibly because of the hooking effects of the iron whiskers. The iron whiskers from Fe2O3 inside the Fe–Mn ores particles facilitated the diffusion of iron atoms, thereby enabling the iron particles to stick together during the fluidization process and promoting the aggregation and growth of iron particles. In addition, when the reduction time exceeded 30 min, the majority of the MnO phase was present on the surfaces of the metallic Fe–Mn particles, and the formation of larger metallic Fe–Mn particles were embedded on the surfaces of the silicates and aluminates. The size of the metallic Fe–Mn particles gradually growed with increasing of duration time.

Fig. 7.

SEM-EDS analysis of reduction products for different reduction times at 1000 °C: (a) 5 min, (b) 10 min, (c) 20 min, (d) 30 min, (e) 50 min, and (f) 120 min.

(1.93MB).
3.2.3Carbothermic reduction behavior of Fe–Mn ore during carbothermic reduction process

Based on phase transformation from XRD analysis and SEM of reduced Fe–Mn ores, the thermodynamic and thermogravimetric analysis of Fe–Mn ore during carbothermic reduction process are displayed in Fig. 8. The stepwise reduction reaction equations of the Fe–Mn ores are listed in Table 1. Note that MnO2 was easily reduced to Mn2O3, Mn3O4 and MnO, but could not be further reduced to Mn at low temperatures, whereas Fe2O3 was readily reduced to Fe3O4, FeO and Fe. The stepwise reduction reactions of Fe2SiO4 and FeAl2O4 were considered in the XRD and SEM/EDS analysis. The FeyMn2−ySiO4 and FeyMn1−yAl2O4 phases are inevitably generated during the carbothermic reduction of Fe–Mn ores. Therefore, the metallic Fe phases formed from both the direct reduction of Fe2SiO4, FeAl2O4 and Fe2O3 and the stepwise reduction of FeyMn2−ySiO4, FeyMn1−yAl2O4 and (FeO)x(MnO)1−x. Whereas, there was a clear discrepancy between the CO content and the temperature required for the reduction reaction of FeAl2O4 and FeyMn1−yAl2O4 or Fe2SiO4 and FeyMn2−ySiO4. The thermodynamic equilibrium of iron oxide reduction by CO (Fig. 8(a)) and C (Fig. 8(b)) are calculated by FACTSAGE 7.0 Software. The Fe2+ in FeAl2O4 and Fe2SiO4 was directly reduced to metallic iron at 841.5 °C and 815.6 °C under the CO atmosphere respectively. When MnO was involved in the reduction of FeAl2O4 and Fe2SiO4, the reduction temperature of the FeAl2O4 and Fe2SiO4, was 658.6 °C and 667.9 °C under the CO atmosphere respectively. It was demonstrated that the participation of MnO can significantly lower the reduction temperature of FeAl2O4 and Fe2SiO4. Different reduction stages of the reduction reaction of the Fe–Mn ores can clearly be observed in Fig. 8(c). The morphology evolution and the phase transformation of the treated Fe–Mn ores could be described for all of the reduction stages in the carbothermic reduction process. The crystal water precipitation in the raw ores initially occurred before all of the reduction reaction stages. The reduction stage A1 (MnxOy → MnO) occurred before the reduction stage A2 (FexOy → FeO). This result indicated that the reduction stage of MnxOy occurred more readily than that of FexOy. The stage A3 (FeyMn2−ySiO4, FeyMn1−yAl2O4 and (FeO)x(MnO)1-x → Fe) also occurred in advance of the reduction stage A4 (Fe2SiO4 and FeAl2O4 → Fe). The TG/DTG analysis results for the reduced Fe–Mn ores were consistent with the thermodynamic analysis.

Fig. 8.

(a) Gas-equilibrium diagram of Mn–Fe–C–O system; (b) ΔG0 − T of FeAl2O4 and Fe2SiO4; (c) TG/DTG of the Fe–Mn ores +5 wt% graphite from room temperature to 1100 °C at a heating rate of 10 °C/min in a pure Ar atmosphere.

(0.5MB).
Table 1.

Reaction equations for reduction of Mn–Fe ore based on the following equation.

No.  Reaction equation  No.  Reaction equation 
Fe3O4 + 4CO = 3Fe + 4CO2  FeO + CO = Fe + CO2 
Fe3O4 + CO = 3FeO + CO2  3Fe2O3 + CO = 2Fe3O4 + CO2 
C + CO2 = 2CO  FeAl2O4 + CO = Fe + Al2O4 + CO2 
FeAl2O4 + MnO + CO = Fe + MnAl2O4 + CO2  Fe2SiO4 + 2CO = 2Fe + SiO2 + 2CO2 
10  Fe2SiO4 + 2MnO + 2CO = 2Fe + Mn2SiO4 + 2CO2  11  MnO + C = Mn + CO 
12  Fe2SiO4 + 2C = 2Fe + SiO2 + 2CO  13  Fe2SiO4 + 2MnO+2C = 2Fe + Mn2SiO4 + 2CO 
14  FeAl2O4 + C = Fe + Al2O3 + CO  15  FeAl2O4 + MnO + C = Fe + MnAl2O4 + CO 
3.3Discussion on the interfacial reaction behavior of FeyMn2−ySiO4 system

Fig. 9 shows the SEM/EDS analysis for the interfacial reaction between FeyMn2−ySiO4, (FeO)x(MnO)1−x, wherein the Fe phase was reduced at 1100 °C for 30 min. As observed, the particles containing (FeO)x(MnO)1−x and the Fe–Mn alloy were observed to shrink after reduction. It’s clearly observed that the metallic Fe–Mn particles were clearly closely wrapped by the fayalite (FeyMn2−ySiO4). Numerous small cracks and pits appeared around the elliptical metallic Fe–Mn particles. The SEM for the reduced Fe–Mn ores also provided evidence of the mass motion of liquid Fe–Mn particles through small pellets with lower (FeO)x(MnO)1−x content and of the motion of the Fe phase towards the larger pellets. Some gaps formed due to particle shrinkage during the reduction of (FeO)x(MnO)1−x to metallic Fe–Mn and MnO. Thus, it was very easy separate and recover Mn and Fe from the Fe–Mn ores by refining of particles followed by magnetic separation. Hence, the metallic particles with a lower (FeO)x(MnO)1−x content and the Fe–Mn alloy entered into the magnetic product after magnetic separation.

Fig. 9.

SEM-EDS analysis of FeyMn2−ySiO4 system reduced at 1100 °C for 30 min.

(1.76MB).

Fig. 10(a) displays the ternary phase diagram of MnO2–Fe2O3-20 wt.% SiO2 system under 60% CO-40% CO2 by Factsage software during carbothermic reduction process. A stable region of olivine and Fe existed above 915 °C. The olivine phase consisted of Fe2SiO4, FeyMn2−ySiO4 and Fe2MnO4. Fig. 10(b) shows the crystal structure of the olivine Fe2SiO4, FeyMn2−ySiO4 and Fe2MnO4. Both the Fe2SiO4 and Fe2SiO4 exhibited similar olivine-type structures. The ionic radii of the Mn2+ (0.067 nm) and Fe2+ (0.061 nm) are in the same level [1,34–37]. Thus, Mn2+ could easily replace Fe2+ in the olivine-type structure, resulting in the formation of intermediate phase FeyMn2−ySiO4. The XRD results (Figs. 4 and 5) show the phase transformation of Fe2SiO4 to Mn2SiO4 as the duration time or the temperature increased. Fig. 10(c) displays the schematic diagram for the interfacial reaction between Fe2SiO4 and MnO, wherein a diffusion coupling model is applied to illustrate the reaction mechanism between Fe2SiO4 and MnO. Under a reducing atmosphere, a multistep reduction reaction procedure occurred between Fe2SiO4 and MnO. (a) Mn2+ ions migrated to the reaction interface of Fe2SiO4 to form an intermediate phase FeyMn2−ySiO4; some of the Fe2+ replaced Mn2+ in the olivine-type structure and were directly reduced to metallic iron, whereas the remaining Fe2+ combined with free MnO to form an intermediate phase (FeO)x(MnO)1−x that was then further reduced to metallic iron. (b) The reduction reaction of Fe2SiO4 occurred at temperatures above 805 °C. Fig. 10(d) shows the SEM-EDS analysis for the Fe2SiO4 group in Fe–Mn ores that were reduced at 1100 °C for 30 min. The (FeO)x(MnO)1−x and FeyMn2−ySiO4 were located close to each other, showing that the formation of (FeO)x(MnO)1−x was directly related to the phase transformation of FeyMn2−ySiO4.

Fig. 10.

(a) Ternary phase diagram of Fe3O4–MnO2-20 wt.% SiO2 system; (b) crystal structure of fayalite Fe2SiO4 and Mn2SiO4; (c) phase transformation of Fe2SiO4 for carbothermic reduction process; and (d) SEM of Fe–Mn ores reduced at 1100 °C for 30 min.

(1.53MB).
Fig. 5.

XRD patterns of Fe–Mn ores reduced for 30 min at different reduction temperature.

(0.74MB).

The results of the reaction mechanisms between Fe2SiO4 and MnO were used to calculate the following chemical formulas of the compounds using a valence state balance: Fe2SiO4 of [Fe2+]2[Si4+][O2−]4, FeyMn2−ySiO4 of [Fe2+]y[Mn2+]2−y[Si4+][O2−]4 and Mn2SiO4 of [Mn2+]2[Si4+][O2−]4. In brief, the reduction reaction mechanism between Fe2SiO4 and MnO is summarized by the multistep reactions given in Eqs. (16) and (17). Both Mn2SiO4 and Fe were generated in the reduction product, and the reduction reaction under a reduction atmosphere can be expressed as follows:

[Fe2+]2[Si4+][O2−]4 + (2 − y)[Mn2+][O2−] = [Fe2+]y[Mn2+]2−y[Si4+][O2−] + (2 − y)[Fe2+][O2−]
[Fe2+]y[Mn2+]2−y[Si4+][O2−] + y[Mn2+][O2−] = [Mn2+]2[Si4+][O2−]4 + y[Fe2+][O2−]
y[Fe2+][O2−] + (1 − y) [Mn2+][O2−] = [Fe2+][O2−]y[Mn2+][O2−]1−y
[Fe2+][O2−] + [C2+][O2−] = [Fe] + [C2+][O2−]2

3.4Discussion on the interfacial reaction behavior of (FeyMn1−y)Al2O4 system

The corresponding optical microstructure images for the Fe–Mn ores reduced at 1100 °C for 30 min are shown in Fig. 11, showing an enormous quantity of metallic iron particles that appear as bright white. The gray matrix phase primarily consisted of fayalite (FeyMn2−ySiO4) and spinel ((FeyMn1−y)Al2O4). The metallic particles were a mixture of Fe–Mn, Mn–Fe-C and (FeO)x(MnO)1−x phases and were elliptic in shape, as confirmed by the EDS and XRD analysis. A significant quantity of the liquid (FeyMn1−y)Al2O4 phase appeared around the metallic Fe–Mn particles. The phenomenon of (FeO)x(MnO)1−x phases wrapped by the metallic Fe–Mn phase was also observed. This phenomenon could be explained using the XRD analysis (Figs. 5 and 6). In addition, a transition phase of a (FeyMn1−y)Al2O4 product was clearly observed. However, the metallic Fe–Mn particles were closely wrapped by the transition phase product. Additionally, a significant quantity of metallic Fe–Mn particles were also present as spinel ((FeyMn1−y)Al2O4) and the (FeO)x(MnO)1−x phase, depending on the reduction conditions. The aforementioned results show that the stepwise reduction of ((FeyMn1−y)Al2O4) and (FeO)x(MnO)1−x inevitably produced a metallic Fe–Mn phase in the reduction of the Fe–Mn ores. The metallic Fe–Mn particles that were wrapped by the (FeyMn1−y)Al2O4 phase were not conducive to magnetic separation. Particles with composite structures are typically difficult to liberate during ball-milling. The (FeyMn1−y)Al2O4 phase easily entered into the magnetic product after magnetic separation. This result is one explanation for the poor separation of Fe and Mn from the Fe–Mn ores from carbothermic roasting reduction process followed by magnetic separation.

Fig. 11.

SEM-EDS analysis of (FeyMn1−y)Al2O4 system reduced at 1100 °C for 30 min.

(1.64MB).

Fig. 12 displays the ternary phase diagram of the crystal structure and phase transformation of FeAl2O4 in a Fe2O3–MnO2-20 wt.% Al2O3 system during carbothermic reduction. As shown in Fig. 12(a), the main phases are a spinel-type phase, the monoxide and an iron phase. The Al-spinel refers to FeAl2O4 and MnAl2O4. The generation of two types of phases are thermodynamically favored under roasting reduction conditions. Fig. 12(b) displays the crystal structure of the spinel FeAl2O4 and MnAl2O4. In the crystal structure of the normal spinel FeAl2O4, FeyMn1−yAl2O4 and MnAl2O4, the Al3+ atoms are form an octahedron, whereas Fe2+ or Mn2+ atoms form a tetrahedron. High temperatures increase the activity of the cations in the spinels of minerals. The distribution of Fe2+ and Al3+ atoms in the hercynite is disordered, such that these cations can occupy tetrahedral and octahedral sites [35–37]. In addition, both the FeAl2O4, FeyMn1−yAl2O4 and MnAl2O4 have spinel-type structures, and the ionic radii of the Mn2+ (0.067 nm) and Fe2+ (0.061 nm) are in the same level. These characteristics facilitated the entrance of Mn2+ atoms into the crystal lattice and the displacement of Fe2+ atoms in the hercynite structure. So it is comfortable to form easily the spinel-type MnAl2O4. Fig. 12(c) shows the three types of phase transformation from FeAl2O4 to Fe in the carbothermic reduction reaction. Scenario A is the formation of metallic Fe according to Eqs. (20) and (21). The FeAl2O4 species was converted to FeyMn1−yAl2O4 by manganese monoxide and further converted to MnAl2O4. The replaced iron atoms were reduced to metallic iron at approximately 658 °C under a reducing atmosphere. The XRD results (Figs. 4(c) and 5(c)) demonstrated that the Mn2+ easily replaced Fe2+ in the spinel-type structure of FeAl2O4, resulting in the formation of FeyMn1−yAl2O4. In scenario B, the FeAl2O4 was directly reduced to metallic iron at 841 °C in a CO atmosphere or at 831 °C by carbon. In scenario C, Fe2+ was replaced by Mn2+ in the spinel-type structure and easily combined with free MnO to form the new phase of (FeO)x(MnO)1−x, which was then further reduced to metallic iron at 731 °C in a CO atmosphere or at 724 °C by carbon, where the stepwise reduction reactions of (FeO)x(MnO)1−x are given by Eqs. (18) and (19). The general solid-state displacement reaction of FeAl2O4 by MnO can be expressed as follows:

[Fe2+][Al3+]2[O2−]4 + (1 − y)[Mn2+][O2−] = [Fe2+]y[Mn2+]1−y[Al3+]2[O2−]4 + (1 − y)[Fe2+][O2−]
[Fe2+]y[Mn2+]1−y[Al3+]2[O2−]4 + y[Mn2+][O2−] = [Mn2+][Al3+]2[O2−]4 + y[Fe2+][O2−]

Fig. 12.

(a) Ternary phase diagram of Fe3O4–MnO2–Al2O3 system; (b) crystal structure of the spinel FeAl2O4 and MnAl2O4; and (c) phase transformation of FeAl2O4 in carbothermic reduction process.

(0.88MB).
3.5Discussion on the interfacial reaction behavior of (FeO)x(MnO)1−x system

The SEM-EDS images and phase transformation of the Fe–Mn ores during the carbothermic reduction process are shown in Fig. 13. During carbothermic reduction process, the formation of (FeO)x(MnO)1−x was unfavorable to the separation of Mn and Fe from the Fe–Mn ores because metallic Fe–Mn particles were distributed among the (FeO)x(MnO)1−x phases, and the metallic Fe–Mn particles were wrapped by the (FeO)x(MnO)1−x particles associated with one another. The tight integration of the MnO phase and metallic Fe–Mn particles derived from the stepwise reduction of (FeO)x(MnO)1−x were closely associated with the stepwise reduction of the FeyMn2−ySiO4 and (FeyMn1−y)Al2O4 phases. As observed, the transition phases of FeyMn2−ySiO4 and (FeyMn1−y)Al2O4 are shown in Fig. 13. The metallic Fe–Mn particles were closely wrapped by the transition phase products of the (FeyMn1−y)Al2O4 and FeyMn2−ySiO4 phases, indicating that the formation of the new (FeO)x(MnO)1−x phase, which resulted from the stepwise reduction process of the two types of phases. With regards to the generation and stepwise reduction behavior of the (FeO)x(MnO)1−x phase, increasing the reduction temperature and shortening the reduction time could enhance the reduction reaction rate of FeyMn2−ySiO4 and (FeyMn1−y)Al2O4 phase, thereby preventing the generation of the (FeO)x(MnO)1−x phase to enhance the separation of Mn and Fe from Fe–Mn ores.

Fig. 13.

SEM-EDS analysis of (FeO)x(MnO)1−x system reduced at 1100 °C for 30 min.

(1.47MB).

A schematic diagram of the interfacial reaction between Fe2SiO4, FeAl2O4 and MnO is inferred and described in Fig. 14. The diffusion coupling model [37] was used to determine the interfacial reaction and the mechanism for the formation of the (FeO)x(MnO)1−x phase. The discussion above is summarized in Fig. 14, whereby the reduction reaction between Fe2SiO4, FeAl2O4 and MnO is roughly divided into the following steps: (a) in the main interfacial reaction, the ionic Mn2+ species migrated from the reaction interface of the MnO phase to the reaction interface of the Fe2SiO4 phase to form the FeyMn2−ySiO4 phase on the reaction interface of the Fe2SiO4 phase and the (FeO)x(MnO)1−x phase on the reaction interface of the MnO phase; (b) in the second stage, there was internal diffusion of Mn2+ through the product layer of Mn2SiO4/FeyMn2−ySiO4 phase and external diffusion of Fe2+ through the product layer of MnO/(FeO)x(MnO)1−x; and (c) the final reduction products of the Mn2SiO4 phase, the (FeO)x(MnO)1−x phase and metallic Fe were obtained. During carbothermic reduction, the CO−CO2 atmosphere played a key role in the stepwise reduction of FeyMn1−yAl2O4, FeyMn2−ySiO4, (FeO)x(MnO)1−x and metallic Fe to the final products, which was crucial for the separation and recovery of Fe and Mn by the subsequent magnetic separation processes. Similar expressions can be written for the interfacial reaction between Fe2SiO4 and MnO and the interfacial reaction between FeAl2O4 and MnO.

[Fe2+]2[Si4+][O2−]4 + (3 − y − x)[Mn2+][O2−] → [Fe2+]y[Mn2+]2−y[Si4+][O2−]4 + [Fe2+][O2−]x[Mn2+][O2−]1−x
[Fe2+][Al3+]2[O2−]4+(2 − y − x)[Mn2+][O2−] → [Fe2+]y[Mn2+]1−y[Al3+]2[O2−]4 + [Fe2+][O2−]x[Mn2+][O2−]1−x

Fig. 14.

Schematic diagram for interfacial reaction of (FeO)x(MnO)1−x in carbothermic reduction process.

(0.62MB).
3.6Recommendations for carbothermic reduction roasting of Fe–Mn ores

The abovementioned results show that the main reason for the poor separation efficiency of Fe–Mn ores was attributed to the formation of the composite oxide phase (FeO)x(MnO)1−x, (FeyMn1−y)Al2O4 and FeyMn2−ySiO4 during carbothermic reduction roasting. The formation of the (FeO)x(MnO)1−x phase was closely related to the reduction process of (FeyMn1−y)Al2O4 and FeyMn2−ySiO4. Therefore, to avoid the formation of (FeO)x(MnO)1−x and promote the reduction rates of (FeyMn1−y)Al2O4 and FeyMn2−ySiO4, high temperatures, short reduction times and fine Fe–Mn ores (3–8 mm) particles sizes are recommended. In an attempt to improve separation, the raw ores with particle sizes of 5–8 mm were reduced at 1100 °C for 2 h. Fig. 15 shows that the magnetic separation index for the Fe–Mn ores reduced at 1050 °C for 6 h for particle sizes of 8–13 mm (before) and at 1100 °C for 2 h with particle sizes of 3–8 mm (after). This change in the carbothermic reduction condition resulted in the Mn and Fe recoveries increasing from 83.08 to 89.38% and 75.08 to 80.66%, respectively. In addition, the Mn grade and mass fraction of Mn/Fe in the nonmagnetic product also increased from 51.33 to 53.60% and from 5.16 to 5.42, respectively. The nonmagnetic product is a Mn-rich raw material that can be used as an acceptable feed (Mn/Fe mass ratio over 5) for smelting a ferromanganese alloy product.

Fig. 15.

Magnetic separation index of Fe–Mn ores reduced at 1050 °C for 6 h with particle sizes of 8–13 mm (before) and at 1100 °C for 2 h with particle sizes of 3–8 mm (after).

(0.15MB).
4Conclusions

Carbothermic reduction roasting followed by magnetic separation process is reported an effective technological process to separation and recovery of Mn and Fe from ferruginous manganese ores, which produces an acceptable feed to meet the requirements of the developing manganese industry. Iron-rich products with a Fe recovery of 80.66% and a manganese-rich product with a Mn recovery of 89.38% were obtained at an optimal temperature of 1100 °C, a roasting reduction time of 2 h, a FC/O of 2.5, a magnetic intensity of 75 mT and a grinding time of 3 min.

The formation mechanism and the interfacial reduction reaction of (FeO)x(MnO)1−x, (FeyMn1−y)Al2O4 and FeyMn2−ySiO4 were systematically discussed to illustrate the effect of the MnO phase on the stepwise reduction of the Fe2SiO4 and FeAl2O4 phases. The formation mechanism and the stepwise reduction of FeAl2O4 proceeded as follows: FeAl2O4 was converted to FeyMn1−yAl2O4 by manganese monoxide and then further converted to MnAl2O4; FeAl2O4 was directly reduced to metallic iron in a CO atmosphere, and Fe2+ replaced Mn2+ in a spinel-type structure that easily combined with free MnO to form the new phase of (FeO)x(MnO)1−x, which was further reduced to metallic iron. The formation mechanism and stepwise reduction of Fe2SiO4 proceeded as follows: Fe2SiO4 was directly reduced to metallic iron in a CO atmosphere; Fe2SiO4 was converted to FeyMn2−ySiO4 and further converted to Mn2SiO4; Fe2SiO4 was directly reduced to metallic iron in a CO atmosphere; and Fe2+ replaced Mn2+ in a fayalite-type structure and easily combined with free MnO to form a new phase (FeO)x(MnO)1−x, which was then further reduced to metallic iron. The formation of the (FeO)x(MnO)1−x phase resulted from the stepwise reduction of (FeyMn1−y)Al2O4 and FeyMn2−ySiO4 phases. Morphology analyses indicated that the presence of Fe2SiO4, FeAl2O4 and (FeO)x(MnO)1−x as the main reason for the poor separation of Mn and Fe. The high temperature and short reduction time restrained the formation of (FeO)x(MnO)1−x and promoted the stepwise reduction of the three phases to optimize the separation of Mn and Fe from the Fe–Mn ores.

Conflicts of interest

The authors declare no conflicts of interest.

Acknowledgements

The authors wish to express their thanks to the National Natural Science Foundation of China (51704061), the China Postdoctoral Science Foundation (2016M601321) and the Fundamental Research Funds of the Central Universities of China (N162503003) for supporting this study.

References
[1]
B.B. Liu, Y.B. Zhang, J. Wang, Z.J. Su, G.H. Li, T. Jiang.
New understanding on separation of Mn and Fe from ferruginous manganese ores by the magnetic reduction roasting process.
Appl Surf Sci, 444 (2018), pp. 133-144
[2]
Y.M.Z. Ahmed.
Synthesis of manganese ferrite from non-standard raw materials using ceramic technique.
Ceram Int, 36 (2010), pp. 969-977
[3]
Z. Cheng, G.C. Zhu, Y.N. Zhao.
Study in reduction-roast leaching manganese from low-grade manganese dioxide ores using cornstalk as reductant.
Hydrometallurgy, 96 (2009), pp. 176-179
[4]
Q.X. Ye, J. Chen, G. Chen, J.H. Peng, C. Srinivasakannan, R.S. Ruan.
Effect of microwave heating on the microstructures and kinetics of carbothermal reduction of pyrolusite ore.
Adv Powder Technol, 29 (2018), pp. 1871-1878
[5]
B.B. Liu, Y.B. Zhang, M.M. Lu, Z.J. Su, G.H. Li, T. Jiang.
A study on the carbonization and alloying process of MnO2 by methane–hydrogen gas mixture in the presence of Fe2O3.
Powder Technol, 325 (2018), pp. 271-279
[6]
S.F. Xiong, X. Li, P.L. Liu, S.H. Hao, F. Hao, Z.L. Yin, et al.
Recovery of manganese from low-grade pyrolusite ore by reductively acid leaching process using lignin as a low cost reductant.
Miner Eng, 125 (2018), pp. 126-132
[7]
Y.B. Zhang, M.H. Du, B.B. Liu, Z.J. Su, G.H. Li, T. Jiang.
Separation and recovery of iron and manganese from high-iron manganese oxide ores by reduction roasting and magnetic separation technique.
Sep Sci Technol, 52 (2017), pp. 1321-1332
[8]
M. Mpho, B. Samson, A. Ayo.
Evaluation of reduction roasting and magnetic separation for upgrading Mn/Fe ratio of fine ferromanganese.
Int J Min Sci Technol, 23 (2013), pp. 537-541
[9]
L.Y. Yi, Z.C. Huang, T. Jiang, P. Zhao, R.H. Zhong, Z.K. Liang.
Carbothermic reduction of ferruginous manganese ore for Mn/Fe beneficiation: morphology evolution and separation characteristic.
Minerals, 7 (2017), pp. 167-172
[10]
Y. Wu, B. Shi, W. Ge, C.J. Yan, X. Yang.
Magnetic separation and magnetic properties of low-grade manganese carbonate ore.
JOM, 67 (2015), pp. 361-368
[11]
Y.B. Gao, M. Olivas-Martinez, H.Y. Sohn, G.K. Han, W.K. Chan.
Upgrading of low-grade manganese ore by selective reduction of iron oxide and magnetic separation.
Metall Mater Trans B, 43 (2012), pp. 1465-1475
[12]
B.B. Liu, Y.B. Zhang, J. Wang, J. Wang, Z.J. Su, G.H. Li, et al.
A further investigation on the MnO2–Fe2O3 system roasted under CO–CO2 atmosphere.
Adv Powder Technol, 30 (2019), pp. 302-310
[13]
Y.V. Swamy, B. Bhoi, S. Prakash, H.S. Ray.
Enrichment of the manganese to iron ratio of ferruginous low-grade manganese ore using solid reductant.
Miner Metall Proc, 15 (2010), pp. 34-37
[14]
Z.C. Huang, B. Chai, L.Y. Yi, T. Jiang.
Reduction and separation of high iron content manganese ore and its mechanism.
4th international symposium on high-temperature metallurgical processing, pp. 367-374
[15]
Z.C. Huang, P. Zhao, T. Jiang, B. Chai.
Process for coal-based direct reduction and magnetic separation of a high-iron manganese ore.
Metal Mine, 3 (2012), pp. 74-77
[16]
Y.B. Zhang, B.B. Liu, Z.X. You, Z.J. Su, W. Luo, G.H. Li, et al.
Consolidation behavior of high-Fe manganese ore sinters with natural basicity.
Miner Process Extr Metall Rev, 37 (2016), pp. 333-341
[17]
J. Kim, G. Dodbiba, H. Tanno, K. Okaya, S. Matsuo, T. Fujita.
Calcination of low-grade laterite for concentration of Ni by magnetic separation.
Miner Eng, 23 (2010), pp. 282-288
[18]
T. Coetsee.
MnO reduction in high carbon ferromanganese production: practice and theory.
Miner Process Extr Metall Rev, 39 (2018), pp. 351-358
[19]
K.K. Rane, P.P. Date.
Reduction and densification characteristics of iron oxide metallic waste during solid state recycling.
Adv Powder Technol, 26 (2015), pp. 126-138
[20]
N. Peng, Q.L. Pan, H. Liu, Z.H. Yang, G.L. Wang.
Recovery of iron and manganese from iron-bearing manganese residues by multi-step roasting and magnetic separation.
Miner Eng, 126 (2018), pp. 177-183
[21]
V. Singh, T.K. Ghosh, Y. Ramamurthy, V. Tathavadkar.
Beneficiation and agglomeration process to utilize low-grade ferruginous manganese ore fines.
Int J Miner Process, 99 (2011), pp. 84-86
[22]
S.K. Tripathy, P.K. Banerjee, N. Suresh.
Effect of desliming on the magnetic separation of low-grade ferruginous manganese ore.
Int J Miner Metall Mater, 22 (2015), pp. 661-673
[23]
A.A. El-Geassy, M.I. Nasr, M.A. Yousef, M.H. Khedr, M. Bahgat.
Behaviour of manganese oxides during magnetising reduction of Baharia iron ore by CO–CO2 gas mixture.
Ironmak Steelmak, 27 (2000), pp. 117-122
[24]
V. Singh, A. Biswas.
Physicochemical processing of low grade ferruginous manganese ores.
Int J Miner Process, 158 (2017), pp. 35-44
[25]
P. Jouni, L. Markku, S. Katri, P. Erkki, K. Risto, H. Risto.
Behavior of silica-supported manganese oxides in hydrometallurgical separations.
Sep Sci Technol, 44 (2009), pp. 3045-3074
[26]
L.H. Gao, Z.G. Liu, Y.Z. Pan, Y. Ge, C. Feng, M.S. Chu, et al.
Separation and recovery of iron and nickel from low-grade laterite nickel ore using reduction roasting at rotary kiln followed by magnetic separation technique.
SME, 36 (2019), pp. 375-384
[27]
L.H. Gao, Z.G. Liu, M.S. Chu, R. Wang, Z.H. Wang, C. Feng.
Upgrading of low-grade manganese ore based on reduction roasting and magnetic separation technique.
Sep Sci Technol, 54 (2019), pp. 195-206
[28]
Z.H. Wang, M.S. Chu, Z.G. Liu, H. Wang, W. Zhao, L.H. Gao.
Preparing ferro-nickel alloy from low-grade laterite nickel ore based on metallized reduction-magnetic separation.
Metals, 313 (2017), pp. 1-2
[29]
N. Kotake, K. Suzuki, S. Asahi, Y. Kanda.
Experimental study on the grinding rate constant of solid materials in a ball mill.
Powder Technol, 122 (2002), pp. 101-108
[30]
S.H. Liu, Q.L. Li, G.S. Xie, L.H. Li, H.L. Xiao.
Effect of grinding time on the particle characteristics of glass powder.
Powder Technol, 295 (2016), pp. 133-141
[31]
F.G. Zeng.
Grinding kinetic equation of coal characterized by particle size.
J China Coal Soc, 25 (2000), pp. 304-306
[32]
J.M. Khatib.
Properties of concrete incorporating fine recycled aggregate.
Cem Concr Res, 35 (2005), pp. 763-769
[33]
S.H. Liu, Q.L. Li, J.W. Song.
Study on the grinding kinetics of copper tailing powder.
Powder Technol, 330 (2018), pp. 107-109
[34]
B.B. Liu, Y.B. Zhang, Z.J. Su, M.M. Lu, Z.W. Peng, G.H. Li, et al.
Formation mechanism of MnxFe3−xO4 by solid-state reaction of MnO2 and Fe2O3 in air atmosphere: morphologies and properties evolution.
Powder Technol, 313 (2017), pp. 201-209
[35]
T. Yamashita, P. Hayes.
Analysis of XPS spectra of Fe2+ and Fe3+ ions in oxide materials.
Appl Surf Sci, 254 (2008), pp. 2441-2449
[36]
L. Zhu, J.N. He, D.R. Yan, Y.C. Dong, J.X. Zhang, X.Z. Lia, et al.
Atmospheric reactive plasma sprayed Fe–Al2O3–FeAl2O4 composite coating and its property evaluation.
Appl Surf Sci, 257 (2011), pp. 10282-10288
[37]
Z.J. Su, Y.B. Zhang, B.L. Han, Y.M. Chen, M.M. Lu, Z.W. Peng, et al.
Interface reaction between Fe3−xSnxO4 and CaO roasted under CO–CO2 atmosphere.
Appl Surf Sci, 423 (2017), pp. 1152-1160
Copyright © 2019. The Authors
Journal of Materials Research and Technology

Subscribe to our newsletter

Article options
Tools
Cookies policy
To improve our services and products, we use cookies (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here.