Recovery and separation of Fe and Mn from simulated chlorinated vanadium slag by molten salt electrolysis

1.   Introduction

Recently, the recycling of solid waste and wastewater as resources has received increasing attention as minimization of solid waste and wastewater is required by environmental policy [1–2]. In China, a large portion of solid waste is produced annually from steel plants. For example, V₂O₅ is prepared from vanadium slag. Nowadays, the extraction processes of vanadium in the industry mainly include calcification roasting–acid leaching and sodium roasting–water leaching, which have achieved successful industrial application [3–5]. However, these processes also have some disadvantages that are difficult to resolve. More than 1.2 million tons of tailings containing a small amount of hazardous V⁴⁺, V⁵⁺, and Cr⁶⁺ produced during extraction of vanadium are discarded every year as solid waste, which contain approximately 30wt% of Fe [6–8]. Only a small number of vanadium tailings are used [8]. From sodium roasting–water leaching process, 30000–50000 kg wastewater with 4–16 g/L ${\rm NH_4^+ }$ and above 20 g/L Na⁺ per 1000 kg V₂O₅ were generated [9]. Recently, the treatment of tailings was investigated. As waste of vanadium production, Fe in tailings was recovered by the carbothermal reduction–magnetic separation process [8,10]. Wang et al. [6,11] reported a method having two steps: the vanadium from vanadium slag was first extracted by CaO roasting–H₂SO₄ leaching method; then, from a mixture of chromite and Fe-containing tailings, Fe–Cr–C alloy was obtained by carbothermal reduction. High-temperature reduction processes for recovering Fe in tailings need relatively high energy consumption.

In the metallurgical extraction processes, chlorinating method has an important role and it is a clean method for recycling many metals [12]. Du et al. [13] reported that the vanadium in the form of vanadium oxytrichloride (VOCl₃) was separated from the vanadium slag by carbochlorination with Cl₂, and the iron-containing residue after dechlorination was used for ironmaking. Compared to gaseous chlorinating agents (Cl₂), the solid chlorides (FeCl₃, FeCl₂, NH₄Cl, and AlCl₃) are more environmentally friendly. Sun [14] proposed selective chlorination of vanadium from vanadium slag using FeCl₂ and FeCl₃ in an oxidizing atmosphere. However, the chlorination ratio of vanadium was only 57%. For minimization of solid waste and wastewater produced during extraction of vanadium, we proposed a new process to preferentially extract Fe and Mn by chlorination of NH₄Cl, and then utilize V and Cr by chlorination of AlCl₃ [15–16]. Fig. 1 shows the flow diagram of the new method to minimize the production of solid waste containing Fe, Mn, V, Cr, and Ti during the usage of vanadium slag. After chlorination, Cr and V were produced in the form of CrCl₃ and VCl₃, respectively. Previously, VCr alloy was prepared by electrodeposition of CrCl₃ and VCl₃ [17]. The molten salt FeCl₂–MnCl₂ was obtained by chlorination of NH₄Cl. Manganese is widely used in steelmaking [18]. The traditional separation of elements Fe and Mn from manganese ores containing Fe is mainly realized by reduction roasting followed by a magnetic separation process [19]. We proposed a novel method for electrolytic separation of Mn and Fe in a molten salt produced from simulated chlorinated vanadium slag.

Fig. 1.  Flow diagram of the new method for utilization of vanadium slag.

DownLoad: Full-Size Img PowerPoint

In previous studies, productions of Fe from molten salts (NaCl–KCl/NaF–KCl) containing FeCl₂ were tried [20–21]. To produce iron, Fe₂O₃ was dissolved in the molten salt (chloride and fluoride) [22–23]. The solubility of Fe₂O₃ can be increased by adding AlCl₃. Haarberg et al. [24] studied the electrochemistry of Fe in CaCl₂–CaF₂. The electrochemical behavior of Fe²⁺ in the molten mixtures of eutectic NaCl–KCl was mostly investigated in KCl–LiCl system, MgCl₂–KCl–NaCl system, CaCl₂–CaF₂ system, and ZnCl₂–NaCl system [24–27]. The electrochemical behavior of Fe³⁺ in LiCl–KCl–NaCl melt was reported by Khalaghi et al. [28]. The electrodeposition of Mn²⁺ in NaCl–KCl system was studied by Xiao et al. [29]. However, the NaCl–KCl–FeCl₂–MnCl₂ system has not been reported.

In this work, as part of the basic research on the extraction of valuable elements from vanadium slag by a molten salt chlorination-electrolysis method, a study on the electrochemical behaviors of Fe²⁺ and Mn²⁺ in NaCl–KCl–MnCl₂ was carried out to separate Fe and Mn. The effective separation of the elements Mn and Fe was realized by electrodeposition of FeCl₂ and MnCl₂ from simulated chlorinated vanadium slag. This investigation provides some guidelines for the effective utilization of vanadium slag.

2.   Experimental

2.1   Materials

The solid reagents used in each experiment, such as NaCl, KCl, MnCl₂, and FeCl₂, were of analytical grade. To remove moisture, before each experiment, the solid KCl and NaCl were dried at 200°C. The Ag/AgCl electrode with NaCl–KCl–2mol%AgCl molten salt served as the reference electrode. A tungsten wire with a diameter of 1 mm and graphite rod with a diameter of 6 mm were used as the working electrode (WE) and the counter electrode, respectively.

2.2   Experimental procedure

2.2.1   Electrochemical experiments

During the electrochemical test, WE has a depth of 1 cm in the molten salt. All electrochemical measurements were performed using a PAPSTAT 2273 system and were used to reveal the electrochemical behaviors of FeCl₂ and MnCl₂ in NaCl–KCl molten salt (NaCl/KCl molar ratio = 50.6:49.4).

2.2.2   Preparation and analysis of products

A uniform mixture of 50.4 g KCl and 40.4 g NaCl with certain weighed amounts of FeCl₂ and MnCl₂ were placed in an Al₂O₃ crucible, which was placed in a SiC rod shaft furnace and vacuumed for 15 min. After heating at 200°C for 24 h to remove volatiles (e.g., moisture), the samples were heated to 800°C under high purity grade argon (Ar). The electrodepositions of Fe and Mn in the two-electrode cell were executed in the molten salt at a constant cell voltage. A graphite rod with a diameter of 6 mm and a tungsten wire with a diameter of 1 mm were used as the anode material and cathode material, respectively. After electrodeposition process, the W cathode was taken out of the molten salt and cooled under high purity grade Ar at the upper part of the SiC rod shaft furnace. With deionized water, the product obtained was washed and dried. The phase composition of deposited product was identified using X-ray diffraction (XRD) with a Cu K_α radiation source. Then, the morphology and element contents of deposited product were analyzed using a scanning electron microscope (SEM, Zeiss Ultra 55) equipped with an energy dispersive spectroscopy (EDS, INCA X-MAX 50, Oxford Instruments). The Fe and Mn contents of the electrodeposited product were measured using inductively couple plasma optical emission spectroscopy (ICP-OES).

3.   Results and discussion

3.1   Electrochemical behavior of MnCl₂ in NaCl–KCl melt system

Fig. 2(a) illustrates cyclic voltammetry (CV) analyses obtained on the WE before and after mixing MnCl₂ in NaCl–KCl melt. In the blank NaCl–KCl melt (the red dotted curve in Fig. 2(a)), only one couple of oxidation and reduction peaks between −1.8 and −2.4 V were associated with oxidation of Na and reduction of Na⁺, respectively. No other oxidation and reduction peaks within the electrochemical window were found except for the oxidation and formation peaks of metal Na. Therefore, the blank NaCl–KCl molten salt is suitable to study the electrochemical behavior of MnCl₂. In the NaCl–KCl–MnCl₂ molten salt (as represented by the solid curves in Fig. 2(a)), the anodic current peaks (A′) and the cathodic current peaks (A) were considered the oxidation and formation of metal Mn at the WE. The potential of the cathode peak (A) moved negatively, and the current of the cathode peak (A) significantly increased as the scanning rate increased.

Fig. 2.  (a) CV analyses obtained on the WE in NaCl–KCl melt before (scanning rate: 0.2 V/s) and after the addition of 1.09wt% MnCl₂ (scanning rates: 0.1–1 V/s) at 800°C; (b) relationship of I_p and v^1/2 in (a); (c) variation of cathodic peak potential as the scan rate logarithm in (a).

DownLoad: Full-Size Img PowerPoint

Fig. 2(b) indicated a linear relationship of I_p and v^1/2 (here, I_p represents the current density of cathodic peak A; v is the scanning rate). It indicated that the electrodeposition of Mn was restricted by the diffusion of Mn²⁺ in the NaCl–KCl–MnCl₂ melt. From Fig. 2(c), $E_{\rm p}^{\rm c}$ and $\text{lg}\,{v}$ showed a linear dependence (here, $E_{\rm p}^{\rm c}$ represents the cathodic peak potential). The cathodic peaks (A) and anodic peaks (A′) corresponding to the reduction of Mn²⁺ and dissolution of Mn can be considered as quasi-reversible as shown in Fig. 2(a).

When the deposition product is insoluble, the diffusion coefficient of Mn²⁺ in NaCl–KCl–MnCl₂ molten salt can be obtained by the following equation [30]:

where I_p is the peak current (A), n is the number of electrons transferred, F is the Faraday constant (C), A is the surface area of the WE in the NaCl–KCl–MnCl₂ molten salt (cm²), R is the ideal gas constant (J·mol⁻¹·K⁻¹), T is the temperature (K), D is the diffusion coefficient of Mn²⁺ (cm²/s), C₀ is the concentration of MnCl₂ in the NaCl–KCl–MnCl₂ molten salt (mol/cm³), and v is the scanning rate (V/s). The diffusion coefficient of Mn²⁺ in the NaCl–KCl–MnCl₂ molten salt was calculated as 4.76 × 10⁻⁵ cm²/s, which was close to the value (4.13 × 10⁻⁵ cm²/s) reported by Xiao et al. [29].

At the constant voltage of 3 V, the current–time curve is shown in Fig. 3(a) for electrolyzing MnCl₂ at 800°C. The current rapidly declined from 4.99 to 2.08 A in the first 0.1 min, which indicated that cell voltage had reached equilibrium. It then gradually rose to 2.48 A with the time up to 8 min, which can be ascribed to increased metal Mn deposition on the W cathode and the W cathode surface area. With the electrolysis time reaching 90 min, the current decreased to 0.39 A due to decreased concentration of Mn²⁺. Then, the current reduced to approximately 0.3 A within 90–114 min [31].

Fig. 3.  (a) Current–time curve of electrolysis of molten salt (NaCl–KCl–7.1wt%MnCl₂, 800°C); (b) XRD analysis of the electrodeposited product by constant cell voltage electrolysis at 3 V (NaCl–KCl–7.1wt%MnCl₂, 800°C); (c) SEM analysis of electrodeposited product by constant cell voltage electrolysis at 3 V (NaCl–KCl–7.1wt%MnCl₂, 800°C).

DownLoad: Full-Size Img PowerPoint

Fig. 3(b) represents the XRD pattern of deposited product, which suggested that the metal Mn can be prepared by constant voltage electrolysis at 3 V. The morphology and composition of the deposited product were analyzed by SEM, as shown in Fig. 3(c). EDS showed that Mn product with a granular shape and a composition of 99.46wt% Mn and 0.54wt% O was obtained.

3.2   Electrolysis of FeCl₂ in NaCl–KCl melt system

Fig. 4(a) shows the cyclic voltammograms of FeCl₂ to reveal the electrochemical behavior of FeCl₂ in NaCl–KCl melt. The anodic peak A′ and cathodic peak A were attributed to the oxidation and formation of metal Fe at the WE. The small peak (B) appeared at about −0.35 V and was considered the oxidation of Fe−W alloy, which was formed by the electrodeposition of NaCl–KCl–FeCl₂ on the WE. This alloy has a certain free energy and the oxidation peak of the alloy is different from the oxidation peak of pure iron [32].

Fig. 4.  (a) CV analyses obtained on the WE in NaCl–KCl–1.09wt%FeCl₂ melt at 800°C (scanning rates: 0.2–1 V/s); (b) relationship of I_p and v^1/2 in (a); (c) variation of cathodic peak potential as the scan rate logarithm in (a).

DownLoad: Full-Size Img PowerPoint

Fig. 4(b) indicated the linear relationship of I_p on v^1/2 for the electrochemical behavior of FeCl₂ in NaCl–KCl melt, suggesting that the electrodeposition of Fe was restricted by the diffusion of Fe²⁺ in the NaCl–KCl molten salt. In Fig. 4(c), $E_{\rm p}^{\rm c}$ and lgv showed a linear dependence. Fig. 4(a) showed the cathodic peaks and anodic peaks corresponding to the reduction of Fe²⁺ and dissolution of Fe, which were quasi-reversible. The diffusion coefficient of Fe²⁺ at 800°C was 4.64 × 10⁻⁵ cm²/s, which was larger than the value (1.4 × 10⁻⁵ cm²/s) at 500°C reported by Khalaghi et al. [28]. As the diffusion coefficient increases with increasing temperature, the diffusion coefficient in this experiment is high.

The electrolysis of molten salt (NaCl–KCl–FeCl₂) was performed at 2.3 V for 120 min. The XRD pattern of deposited product in Fig. 5(a) suggested that the metal Fe was obtained by molten salt electrolysis. Fig. 5(b) shows the morphology of deposited product, which was analyzed by SEM. It showed that the Fe product exhibited two kinds of crystals: stick-like and granular shape. The EDS analysis of Fe product indicated that the mass percentages of Fe and O were 98.07% and 1.93%, respectively. Thus, electrodeposition of FeCl₂ produced low-oxygen metallic Fe.

Fig. 5.  (a) XRD spectrum of the electrodeposited product by constant cell voltage electrolysis at 2.3 V (NaCl–KCl–1.63wt%FeCl₂, 800°C); (b) SEM analysis of the electrodeposited product (NaCl–KCl–1.63wt%FeCl₂, 800°C).

DownLoad: Full-Size Img PowerPoint

3.3   Electrochemical tests of FeCl₂ and MnCl₂ in the NaCl–KCl melt system

The mass ratio of FeCl₂ and MnCl₂ was selected according to the previous research [16]. The cyclic voltammogram of Fe²⁺ and Mn²⁺ in NaCl–KCl–FeCl₂–MnCl₂ melt is illustrated in Fig. 6(a) at the WE at 800°C. In Fig. 6(a), two reduction peaks (A and B) at a scanning rate of 50 mV/s appeared. The square wave voltammetry (SWV) was used to clear the relative position of the peaks and study the reduction of Fe²⁺ and Mn²⁺ in the NaCl–KCl–FeCl₂–MnCl₂ melt. Fig. 6(b) shows the SWV analysis obtained on the WE in the NaCl–KCl–2.13wt%FeCl₂–1.07wt%MnCl₂ melt at 800°C, in which two cathodic signal peaks (A and B) appeared at −1.079 and −1.522 V. The cathodic peak potential of MnCl₂ in NaCl–KCl–MnCl₂ molten salt at 400 mV/s was −1.64 V (Fig. 2(a)), while the cathodic peak potential of FeCl₂ in NaCl–KCl–FeCl₂ molten salt at 400 mV/s was −1.11 V (Fig. 4(a)). Thus, peak A in Fig. 6(b) was regarded as Fe²⁺ reduced to Fe. The peak position of Fe²⁺/Fe moved in the positive direction and recognized the presence of Mn²⁺. The peak B in Fig. 6(b) was associated with Mn²⁺ reduction to Mn. The peak position of Mn²⁺/Mn also moved in the positive direction, which arised from the presence of Fe²⁺. The positive direction shift of potential was considered an increasing activity of the deposited metal ions [33].

Fig. 6.  (a) CV analysis of NaCl–KCl–2.13wt%FeCl₂–1.07wt%MnCl₂ melt on the WE at 800°C (scanning rate: 0.05 V/s); (b) square wave voltammogram of NaCl–KCl–2.13wt%FeCl₂–1.07wt%MnCl₂ melt on the WE at a step potential of 3 mV and scanning rate of 400 mV/s.

DownLoad: Full-Size Img PowerPoint

The separation rate between Fe and Mn can be obtained by the following equations [34]:

where c_initial and c_final are the molalities of Fe ions before and after the separation, η is the separation rate between Fe and Mn, $\Delta E$ is the difference between the reduction potential of Mn²⁺ and Fe²⁺. According to the SWV results of NaCl–KCl–FeCl₂–MnCl₂, the potential difference between Fe and Mn was 0.443 V. The number of electrons transferred during the reduction of Fe²⁺ to Fe was 2. The theoretical separation rate between FeCl₂ and MnCl₂ can reach 99.993%. Thus, it is feasible to realize the separation of Mn and Fe by electrolysis.

The separation of Fe²⁺ and Mn²⁺ was performed in NaCl–KCl–2.13wt%FeCl₂–1.07wt%MnCl₂ molten salt using two-electrode at different constant potentials at 800°C. The decomposition voltages of FeCl₂ and MnCl₂ in NaCl–KCl–2.13wt%FeCl₂–1.07wt%MnCl₂ molten salt were obtained by Eqs. (4)–(8). The standard Gibbs free energies of Eqs. (4) and (6) were obtained by calculating using the FactSage 6.4 program.

where ${\Delta}_{\mathrm{r}}{G}_{{\rm{FeCl}}_{\rm{2}}}$ and ${\Delta}_{\mathrm{r}}{G}_{{\rm{FeCl}}_{\rm{2}}}^{\ominus}$ are Gibbs free energy (J) and standard Gibbs free energy (J) of FeCl₂, respectively; ${\alpha }_{{\rm{FeCl}}_{\rm{2}}}$, ${\alpha }_{\rm{Fe}}$ and ${\alpha }_{{\rm{Cl}}_{\rm{2}}}$ are activities of FeCl₂, Fe, and Cl₂, respectively; T is the temperature (K); and R is the ideal gas constant (J·mol⁻¹·K⁻¹).

where ${\Delta}_{\mathrm{r}}{G}_{{\rm{MnCl}}_{\rm{2}}}$ and ${\Delta}_{\mathrm{r}}{G}_{{\rm{MnCl}}_{\rm{2}}}^{\ominus}$ are Gibbs free energy (J) and standard Gibbs free energy (J) of MnCl₂, respectively; ${\alpha }_{{\rm{MnCl}}_{\rm{2}}}$, ${\alpha }_{\rm{Mn}}$, and ${\alpha }_{{\rm{Cl}}_{\rm{2}}}$ are activities of MnCl₂, Mn, and Cl₂, respectively.

where ${\Delta}_{\mathrm{r}}G$ is the Gibbs free energy (J), n is the number of electrons transferred, F is the Faraday constant (C), and E is the decomposition voltage (V).

Owing to the high purity of Ar gas atmosphere in this experiment, we assume that the activity of Cl₂ is 1. In view of the low contents of FeCl₂ and MnCl₂ in NaCl–KCl–2.13wt%FeCl₂–1.07wt%MnCl₂ molten salt, the contents of FeCl₂ and MnCl₂ were used to calculate the Gibbs free energy instead of the activity. The values of ${\alpha }_{\mathrm{M}\mathrm{n}}$ and ${\alpha }_{\mathrm{F}\mathrm{e}}$ are assumed to be 1. The values of ${\Delta}_{\mathrm{r}}{G}_{{\mathrm{M}\mathrm{n}\mathrm{C}\mathrm{l}}_{2}}^{\ominus}$ and ${\Delta}_{\mathrm{r}}{G}_{{\mathrm{F}\mathrm{e}\mathrm{C}\mathrm{l}}_{2}}^{\ominus}$ at 800°C are −349169.1 and −214689.1 J, respectively. So, the decomposition voltages of FeCl₂ and MnCl₂ at 800°C are 1.530 and 2.255 V, respectively.

To achieve separation of Fe and Mn, the influence of different electrolytic voltages (2.0, 2.3, and 3 V) on the separation rate between Fe and Mn were investigated. At different voltages, the current–time curves are shown in Fig. 7(a) for the separation of MnCl₂ and FeCl₂ at 800°C. It can be seen that the current–time curves for different electrolytic voltages were similar. Obviously, in the first 0.2 min, the current at constant voltage of 3 V quickly declined from 1.96 to 0.5 A, which suggested that cell voltage had reached the equilibrium. It then rose gradually to 3.09 A with time up to 7 min, which can be attributed to increase in metal deposition on the W cathode and the surface area of W cathode. During the electrolysis for 54 min, the current decreased to 0.27 A due to the concentration decrease of Fe²⁺. Then, the current reduced to approximately 0.2 A within 54–115 min. The current within 0.2 min rapidly declined to 0.5 A at constant voltage of 2.3 V. This may correspond to the electrolytic cell of molten salt to reach equilibrium. It then rose gradually to 1.96 A within 20 min, which can be ascribed to increase in metal Fe deposition on the W cathode and the surface area of W cathode. During the electrolysis for 54 min, the current decreased to 0.27 A due to the decrease in concentration of Fe²⁺. Then, the current reduced to about 0.2 A within 54–78 min. In the first 0.2 min, the current rapidly declined to 0.5 A at a constant voltage of 2 V. This may correspond to the electrolytic cell of molten salt to reach equilibrium. It then rose gradually to 1.27 A within 23 min, which can be attributed to increase metallic Fe deposition on the W cathode and the surface area of tungsten cathode. During the electrolysis for 80 min, the current decreased to 0.07 A due to decrease in concentration of Fe²⁺. Then, the current reduced to about 0.01 A within 80–135 min. Fig. 7(a) shows the area of current–time curve at 3 V, which is bigger than those at 2.3 and 2.0 V. Thus, the amount of electrodeposited products (Fe and Mn) increased as the electrolytic voltage increased from 2 to 3 V.

Fig. 7.  (a) Current–time curves of electrolysis of molten salts (NaCl–KCl–2.13wt%FeCl₂–1.07wt%MnCl₂, 800°C); (b) XRD analyses of the electrodeposited products by different constant cell voltages electrolysis (NaCl–KCl–2.13wt%FeCl₂–1.07wt%MnCl₂, 800°C).

DownLoad: Full-Size Img PowerPoint

The XRD patterns in Fig. 7(b) indicated that the deposited products at 2.3 and 2.0 V were metal Fe. It was implied that the metal Fe at 2.3 and 2.0 V was electrolyzed. At 3 V, the peak characteristics for Mn appeared and the Mn²⁺ was also electrolyzed to metal Mn.

The SEM analyses of the deposited products at different voltages are presented in Fig. 8. The Fe–Mn products exhibited two morphological crystals: square and irregular granular. By using ICP-OES, the element contents of Fe–Mn products were determined. The Fe/Mn mass ratios in deposited products at 2, 2.3, and 3 V were 396.5, 16, and 3.9, respectively. Based on the above calculations, the decomposition voltage of MnCl₂ at 800°C are 2.255 V. Thus, when the electrolysis voltage exceeded 2.3 V, the Mn²⁺ in NaCl–KCl–2.13wt%FeCl₂–1.07wt%MnCl₂ melts was deposited as metal Mn.

Fig. 8.  SEM analyses of electrodeposited products under different electrolysis voltages: (a) 2 V; (b) 2.3 V; (c) 3 V.

DownLoad: Full-Size Img PowerPoint

After the electrolysis of NaCl–KCl–2.13wt%FeCl₂–1.07wt%MnCl₂ melt at 2.3 V, W electrode was taken out of the molten salt and held 5 cm above the molten salt. Another W electrode as the cathode was inserted into the NaCl–KCl–FeCl₂–MnCl₂ melt. Electrolysis was started again under 3.0 V. The effect of electrolysis time on current at 800°C is shown in Fig. 9(a). The current of the second electrolysis at 3 V was lower than that of the first electrolysis at 2.3 V. The XRD pattern of deposited product in Fig. 9(b) suggested that the metal Mn was obtained by the molten salt electrolysis. The morphology of deposited product was analyzed by SEM and displayed in Fig. 9(c). The morphology of Fe–Mn product exhibited irregular granular. The element contents of Fe–Mn product were determined using ICP-OES. The Mn/Fe mass ratio of deposited product of the manganese-rich phase at 3 V was 36.4. After the electrolysis of NaCl–KCl–2.13wt%FeCl₂–1.07wt%MnCl₂ melt at 2 V, electrolysis was again started under 3.0 V. The Mn/Fe mass ratio of deposited product of the manganese-rich phase at 3 V was 5. Compared the first electrolysis at 2.3 and 2 V, although the manganese content in the deposited product of the iron-rich phase at 2 V was very low, the iron content in manganese-rich phase obtained from the second electrolysis at 3 V was very high. Thus, the first electrolysis voltage of 2.3 V and the second electrolysis voltage of 3 V were better for the separation of Fe and Mn. The Mn/Fe mass ratios of magnetic concentrate and nonmagnetic material obtained by the magnetic separation process are 0.08 and 28.41, respectively [19]. Apparently, the molten salt electrolysis method on separation of Fe and Mn is better than the magnetic separation.

Fig. 9.  (a) Current–time curve of the second electrolysis of molten salt at 3 V (MnCl₂–FeCl₂–NaCl–KCl, 800°C); (b) XRD analysis of the electrodeposited product by constant cell voltage electrolysis at 3 V (MnCl₂–FeCl₂–NaCl–KCl, 800°C); (c) SEM analysis of electrodeposited product at 3 V (MnCl₂–FeCl₂–NaCl–KCl, 800°C).

DownLoad: Full-Size Img PowerPoint

4.   Conclusion

The electrochemical behaviors of Fe²⁺ and Mn²⁺ in NaCl–KCl melt at 800°C were studied using SWV and CV on a WE. The reduction processes of Fe²⁺ and Mn²⁺ were found to involve one step. The electrochemical processes of Fe²⁺ and Mn²⁺ were regarded as quasi-reversible processes. The electrodeposition processes of Fe²⁺ and Mn²⁺ were restricted by diffusion. The diffusion coefficients of Fe²⁺ and Mn²⁺ in NaCl–KCl melt were 4.64 × 10⁻⁵ and 4.76 × 10⁻⁵ cm²/s, respectively. In NaCl–KCl molten salt, the electrodeposition of Fe²⁺ and Mn²⁺ was performed using two electrodes at constant cell voltage, and metals Fe and Mn were obtained. The separation of Fe²⁺ and Mn²⁺ in NaCl–KCl melt was studied by molten salt electrolysis. The Fe/Mn mass ratio of electrodeposited product in NaCl–KCl–2.13wt%FeCl₂–1.07wt%MnCl₂ was 16 at 2.3 V. After the electrolysis of NaCl–KCl–2.13wt%FeCl₂–1.07wt%MnCl₂ melt at 2.3 V, electrolysis was again started under 3.0 V and the Mn/Fe mass ratio of deposited product was 36.4. Mn/Fe mass ratios of magnetic concentrate and nonmagnetic materials obtained by the magnetic separation process are 0.08 and 28.41, respectively. Apparently, the molten salt electrolysis method on separation of Mn and Fe is better than the magnetic separation. This process provides a novel method to effectively separate Fe and Mn from vanadium slag.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (Nos. 51904286, 51922003, 51774027, and 51734002) and the China Postdoctoral Science Foundation (No. 2019M650848).