Enhanced Removal of Pb from Electrolytic Manganese Anode Slime by Vacuum Carbothermal Reduction

Electrolytic manganese anode slime (EMAS) is produced during the production of electrolytic manganese metal. In this study, a method based on vacuum carbothermal reduction was used for Pb removal in EMAS. A Pb-removal eciency of 99.85% and MnO purity in EMAS of 97.34 wt.% was obtained for a reduction temperature of 950°C and a carbon mass ratio of 10% for a holding time of 100 min. The dense structure of the EMAS was destroyed, a large number of multidimensional pores and cracks were formed, and the Pd-containing compound was reduced to elemental Pb by the vacuum carbothermal reduction. A recovery eciency for chemical MnO 2 of 36.6% was obtained via preparation from Pd-removed EMAS through the “roasting-pickling disproportionation” process, with an acid washing time of 100 min, acid washing temperature of 70°C, H 2 SO 4 concentration of 0.8 mol/L, liquid-solid mass ratio of 7 mL/g, calcination temperature of 60°C and calcination time of 2.5 h. Moreover, the crystal form of the prepared chemical MnO 2 was found to be basically the same as that of electrolytic MnO 2 , and its specic surface area, micropore volume and discharge capacity were all higher than that of electrolytic MnO 2 . This study provides a new method for Pd removal and recycling for EMAS.


Introduction
Electrolytic manganese metal (EMM) is an important raw material for industrial production that is widely used in various industrial elds and occupies an important position in the national economy (Zhang et al., 2020). China is a major producer of electrolytic manganese metal . In 2020, 96.5% of the world's electrolytic manganese metal was produced in China. The production process for EMM mainly includes leaching, impurity removal, electrolysis, and product posttreatment (Tao et al., 2018). Electrolytic manganese anode slime (EMAS) is a solid waste found in the anode chamber during the production of electrolytic manganese metal (Tran et al., 2020). At present, the global annual output of EMM is approximately 1.5 million tons, and approximately 75,000 to 225,000 tons of EMAS is generated each year.
In China, most EMAS is dumped as hazardous waste or sold for small additional value. Therefore, EMAS has become a bottleneck hindering the development of the electrolytic manganese industry.
Many scholars have proposed the method of resource-based treatment of EMAS, but the main methods are reduction leaching and acid leaching roasting activation (Zhang et al., 2018). For example, Guo et al. used the roasting leaching method to remove Pd from EMAS and then prepared lithium manganate material from Pd-removed EMAS (Guo et al, 2018). However, the process for this method is more complicated, and the removal e ciency of Pd is low (Chen et al, 2019). Many researchers recycle manganese and Pd from EMAS by the use of different reducing agents (Chen et al., 2018;Cheng et al., 2009;Li et al., 2017;Niu et al., 2012;Gui et al., 2014;Wei et al., 2017). However, these methods have certain limitations, such as high energy consumption, complex operation, low e ciency and added value (Cheng et al., 2009;Ye et al., 2015;. The removal of Pd and the regulation of the crystal form of manganese oxides are the key factors limiting the application of the above techniques. Regulation of the manganese oxide crystal form and Pb removal have become urgent problems to be solved in the electrolytic manganese metal industry.
The vacuum carbothermal reduction method has always been a research hotspot in the eld of vacuum metallurgy, and great research progress has been made . The vacuum carbothermal reduction method has the characteristics of both reduction roasting and vacuum smelting and has the characteristics of low energy consumption, simple processing, and environmental friendliness; it is widely used in the elds of chemical engineering and metallurgy (Brkic et  In this study, the vacuum carbothermal reduction method was rst used for Pb-removal in EMAS. The effects of process parameters such as reduction temperature, holding time and mass ratio of carbon on the Pd-removal process were investigated, and the removal mechanism for Pd was analyzed. In addition, the preparation process for chemical MnO 2 by the "roasting-acid washing disproportionation" process was studied. This study provided a new idea for the high-value resource utilization of EMAS.

Raw material
The EMAS samples used in this paper were all sampled in an electrolytic manganese plant in Chongzuo, Guangxi, according to the "technical speci cation for sampling and preparation of industrial solid waste". The EMAS samples were dried at 80°C, milled by ball milling and screened through 200 mesh for standby. H 2 SO 4 and other chemical reagents were analytically pure and purchased from Chongqing Boyi Chemical Reagent Co., Ltd.

Experiment for Pd removal
First, the EMAS was pretreated by washing and drying. Secondly, a certain mass of pretreated EMAS and a certain mass ratio of activated carbon (4 wt%, 6 wt%, 8 wt%, 10 wt%, 12 wt%) were transferred to an agate mortar for full mixing. Finally, the sample boat was placed into the middle of a tubular furnace, and the vacuum pump and tubular furnace were started. The sample was then roasted at a set reaction temperature (850°C, 900°C, 950°C, 1000°C and 1050°C), held for a certain time (70 min, 80 min, 90 min, 100 min and 110 min), and then naturally cooled to room temperature.

Preparation of chemical MnO 2
Chemical MnO 2 was prepared from Pb-removed EMAS by a "roasting-acid washing disproportionation" process. The roasting process is the key step affecting the preparation of chemical MnO 2 from Pd-removed EMAS (MnO), while the acid washing process has little effect on the conversion e ciency of chemical MnO 2 ). Therefore, the effects of calcination temperature (450°C, 500°C, 550°C, 600°C, 650°C) and calcination time (1.5 h, 2 h, 2.5 h, 3 h, 3.5 h) on the conversion of chemical MnO 2 were investigated. The process parameters used for acid washing disproportionation were as follows: acid washing time of 100 min, acid washing temperature of 70°C, H 2 SO 4 concentration of 0.8 mol/L, and liquid-solid mass ratio of 7 mL/g.

Discharge performance test for chemical MnO 2
The prepared sample electrode (0.032 g sample) was used as the working electrode, a zinc sheet (1 cm×1 cm) was used as the reference electrode and counter electrode, and a 9 mol/L KOH saturated solution was used as the electrolyte. The constant current discharge method was used to measure the speci c capacity of the sample, and the termination voltage was 1.0 V. The speci c capacity was determined using the calculation method shown in equation 1: Where C (mAh·g −1 ) is the speci c capacity of the sample to be tested; I (mA) is the discharge current; T (h) is the discharge time; and m (g) is the mass of the sample.

Analysis method
X-ray uorescence (XRF) (XRF-1800, Japan) was used to analyze the elemental composition of the EMAS sample. X-ray diffraction (XRD) (D/Max-2500, Japan) and scanning electron microscopy (SEM) (JSM-7800F, Japan) were used to analyze the phase composition and microstructure of EMAS and EMAS after vacuum carbothermal reduction and chemical MnO 2 . The speci c surface area and pore diameter of chemical MnO 2 were analyzed by the Brunauer-Emmett-Teller (BET) method (3H-2000PS1, Best Instrument Technology Co., Ltd., China.). The discharge performance of electrolytic MnO 2 and chemical MnO 2 was analyzed by using an electrochemical workstation (CHI660E, Shanghai Chen hua Instrument Co., Ltd., China.). The Pb content was determined by XRF, and the following formula was used to calculate the Pbremoval e ciency: Where φ Pb (%) is the removal e ciency for Pb; C 0 (mg·g −1 ) is the original Pb content; C e (mg·g −1 ) is the Pb content after vacuum carbothermal reduction treatment; φ Mn (%) is the conversion e ciency for chemical MnO 2 ; m 0 (mg·g −1 ) is the original weight of Mn in EMAS after vacuum carbothermal reduction treatment; and m e (mg·g −1 ) is the weight of Mn in MnO 2 after "roasting -acid washing disproportionation" treatment.

Removal behavior for Pd
The Pb-removal e ciency in EMAS increased with increasing reduction temperature (Fig. 1a). The Pdremoval e ciency was 85.12% and 99.85% when the reduction temperature was 850°C and 950°C, respectively. Then, the reduction temperature continued to increase, and although the Pb-removal e ciency increased slightly, this increase was not obvious. Therefore, considering the energy consumption, the reduction temperature was suggested to be 950°C. As shown in Fig. 1b, within the time range selected in the experiment, the removal e ciency for Pb in EMAS increased obviously with time, but when the holding time exceeded 100 min, the change in the Pb-removal e ciency was small, so it was better to select a holding time of 100 min. When the carbon mass ratio reached 10%, the Pb-removal e ciency basically tended to be stable, which changed from 99.85-99.88% (Fig. 1c). This can be explained by the fact that the amount of reducing agent was not enough to fully reduce the Pb compounds in the EMAS when the proportion of carbon was relatively small, which led to a poor removal effect for Pb. When the proportion of activated carbon reached 10%, the Pb compounds in the EMAS were fully reduced, so it was better to choose the proportion of carbon as 10%. The optimum Pb-removal e ciency was 99.85%.

Removal Mechanism for Pd
In Table 1 and were reduced to PbO. MnO was the main phase in EMAS, which indicated that MnO 2 was completely reduced to MnO when the temperature ranged between 700°C and 950°C. All Pd compounds in EMAS were reduced to Pd when the temperature was higher than 700°C. The main phase was MnO with good crystallinity and no impurity peak observed when the temperature was 950°C, which indicated that Pb was almost completely removed from EMAS. As shown in Fig. 2c, the black condensate collected in the vacuum tubular furnace was metallic Pb, which further proved that the Pb-containing compounds in the EMAS were removed by Pd vapor. As shown in Fig. 3a, SEM for the raw EMAS particles presented a relatively dense state. As shown in Fig. 3b to Fig. 3d, the dense structure of the EMAS main body was gradually destroyed, and a large number of multidimensional pores and cracks were formed as the temperature was increased. These multidimensional CO 2 + C = 2CO (10) 2. Pb-removal reactions: 2 PbSO 4 + C = 2 PbO + 2 SO 2 (g)+ CO 2 (g) (11) PbSO 4 + CO = PbO + SO 2 (g)+ CO 2 (g) (12) 2 Pbo + Co = Pbo + Co (G) (13) Pb 2 O + CO = 2Pb(g) + CO 2 (g) (14) CO 2 + C = 2CO (15)

Preparation of chemical MnO 2
As shown in Fig. 4a and Fig. 4b, the conversion e ciency of chemical MnO 2 (CMD) increased with increasing temperature, increasing from 20.4-37.2%. When the temperature exceeded 600°C, the conversion e ciency for CMD increased slightly. Therefore, it was better to choose a roasting temperature of 600°C at which the conversion e ciency for CMD was 36.6%. When the roasting time was less than 2.5 h, the CMD conversion e ciency increased obviously with time (from 28.3-36.6%); when the roasting time was more than 2.5 h, the CMD conversion e ciency increased slightly (from 36.6-38.5%), and with the increasing calcination time, the economy worsened. Therefore, the optimum roasting time was 2.5 h for a CMD conversion e ciency of 36.6%.
As shown in Fig. 5, the crystal forms of chemical MnO 2 (CMD) and electrolytic MnO 2 (EMD) were basically the same. Compared with the standard card, the main crystal form was γ -MnO 2 . There were many lattice defects, no ideal ratio and vacancies in the crystal form of γ -MnO 2 , which has the characteristics of a large cross-sectional area of crystal tunnels and high electrochemical activity. This type of crystal form for γ -MnO 2 is widely used in power batteries and alkaline manganese batteries and other industries, which affords the prepared CMD with natural advantages due to its crystal form and provides a large number of application scenarios for the high-value resource utilization of CMD (Guo et al., 2007;Xiao Chai et al., 1978). As shown in Fig. 6a and Fig. 6b, the CMD particle size was larger than that of EMD, the particles were staggered and stacked together, there was a substantial agglomeration phenomenon, and the particle size of the aggregate was approximately 1 µm.
3.4 Discharge performance of chemical MnO 2 Figure 7a and 7b shows discharge curves for EMD and CMD in 9 mol/L KOH (ZnO saturated) solution with different discharge currents at 1.0 V, respectively. The potential plateau in the discharge curve is attributed to the transformation between manganese dioxide and metal . The discharge platforms for CMD and EMD were basically the same. The discharge plateau for CMD was more stable and lasted longer at a discharge current of 0.1 A/g, 0.3 A/g and 0.4 A/g. As shown in Fig. 7c, the discharge capacity of CMD was 240.84 mAh/g, and the discharge capacity of EMD was 223.96 mAh/g when the discharge current was 0.1 A/g. The results showed that the discharge performance of CMD was better than that of EMD.
A comparative analysis of CMD and EMD by the BET test is shown in Fig. 8. The adsorption capacity of the prepared CMD was greater than that of EMD. The speci c surface area of CMD was 45.2607 m 2 ·g −1 , while that of EMD was only 28.3444 m 2 ·g −1 , which is consistent with the SEM results. According to the IUPAC classi cation, the adsorption isotherms for CMD and EMD can be classi ed as type II with H 3 -type hysteresis, respectively (Thevenot et al., 1996). In the range of P/P 0 ≤0.40, the N 2 adsorption capacity of CMD and EMD increased gradually with increasing relative pressure, and the adsorption curve and analytical curve overlapped in this region, which indicated a small amount of microporous adsorption and monolayer adsorption (Shu et al., 2017). When P/P0 > 0.40, the adsorption capacity of N 2 increased rapidly, and H 3 -type hysteresis was observed at relatively high pressure, which was due to the capillary condensation and multilayer adsorption of N 2 in mesopores and macropores, indicating that the pores in CMD and EMD are narrow slit-like pores (Bakandritsos et al,. 2004;Bu et al., 2010). According to the Dubinin-Radushkevich model, the micropore volume of CMD was 0.0896 cm 3 ·g −1 , while that of EMD was only 0.0489 cm 3 ·g −1 . The average pore size of CMD was 7.92 nm and that of EMD was 6.90 nm. Therefore, it was considered that the speci c surface area and pore size of CMD is higher than those for EMD, which may be one of the reasons for the better discharge performance of CMD.

Conclusion
In this study, the vacuum carbothermal reduction method was used for Pd-removal from EMAS. In this process, the main phase of EMAS was reduced to MnO, and the Pd-containing compounds were gradually reduced to metal Pd and volatilized. A Pd-removal e ciency of 99.85% and MnO powder purity in EMAS of 97.34% was obtained at a reduction temperature of 950°C, carbon mass ratio of 10%, and holding time of 100 min. A Mn recovery e ciency of 36.6% was obtained by the "roasting-acid washing disproportionation" process under the following conditions: acid washing time of 100 min, acid washing temperature of 70°C, H 2 SO 4 concentration of 0.8 mol/L, liquid-solid mass ratio of 7 mL/g, calcination temperature of 60°C and calcination time of 2.5 h. The crystal form of CMD was basically the same as that of EMD, and the speci c surface area and micropore volume of CMD were higher than those of EMD. The discharge capacity of CMD was 16.88 mAh/g, which was higher than that of EMD at a discharge current of 0.1 A/g. This study provides a new method for the recycling of EMAS.

Declarations
Ethics approval and consent to participate Approval was obtained from the ethics committee of Southwest University of Science and Technology. The procedures used in this study adhere to the tenets of the Declaration of Helsinki.

Consent to Publish
Informed consent was obtained from all individual participants included in the study.

Funding
All sources of funding for the research reported should be declared. The role of the funding body in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript should be declared.

Availability of data and materials
All data generated or analysed during this study are included in this published article (and its supplementary information les).

Supporting Information
Supporting Information is not available with this version.