Brazilian Journal of Chemical Engineering BATTERY RECYCLING : EFFECT OF CURRENT DENSITY ON MANGANESE RECOVERY THROUGH ELECTROLYTIC PROCESS

This work aims to verify the possibility of using depleted batteries as a source of manganese dioxide applying the electrolytic process. An electrolyte solution containing the following metal ions was used: Ca (270 mgL), Ni (3.000 mgL), Co (630 mgL), Mn (115.3 mgL), Ti (400 mgL) and Pb (20 mgL). The production of electrolytic manganese dioxide (EMD) was performed through electrolysis at 98 °C (± 2 °C) applying different current densities (ranging from 0.61 A.dm to 2.51 A.dm). The materials obtained were analyzed through X-ray fluorescence spectrometry, X-ray diffraction, specific surface area (BET) and scanning electron microscopy (SEM). The best results regarding the current efficiency, purity grade and specific surface area were obtained with a current density ranging between 1.02 A.dm and 1.39 A.dm. The allotropic ε-MnO2 variety was found in all tests.


INTRODUCTION
Electrolytic manganese dioxide (EMD) is the main active component present in the cathodes of several cell systems.The most common system among those using manganese dioxide is the system of primary alkaline batteries Zn/MnO 2 .This component is also found in the aqueous system of zinc-carbon batteries and in some lithium batteries.Currently, manganese dioxide has been investigated as a precursor in the production of spinel (LiMn 2 O 4 ), one of the main components to be used as cathodic material for rechargeable lithium batteries (Strobel et al., 2003;Komaba et al., 2005).Recycling of batteries has been studied in recent years, but the recovery of materials present remains a challenge (Espinosa, et al., 2004;Bernardes et al., 2004;Espinosa and Tenório, 2004;Espinosa and Tenório, 2006).
The use of electrolytic manganese dioxide in the alkaline battery industry exceeded 23.000 t/year in 2.002 (Zhang and Cheng, 2007).The high consumption of this substance is because of the growing demand for batteries in the market worldwide, generated by the mass production and consumption of portable electronic devices such as digital cameras, telephones, computers and high-tech toys, whose power supplies are almost exclusively batteries based on a Zn/MnO 2 system.These cells represent a significant percentage of the total number of batteries available in the market, mainly due to the characteristics of the manganese dioxide cathode performance (Browning and Donne, 2005) and its low cost.
Because of the growing need for products containing manganese, the use of secondary sources of this metal has become increasingly necessary.For this purpose, studies have been done on: the use of Brazilian Journal of Chemical Engineering low-grade manganese ores (lower than 40%), of ocean manganese nodules and manganese-containing materials such as steel scrap, spent electrodes, used catalysts, and spent batteries (Zhang and Cheng, 2007).
Considering these studies, this work focuses on the recovery of manganese from spent batteries, which is feasible due to the high content of manganese dioxide (28 to 38 wt.%) present in the composition of cells produced with this material.
The hydrometallurgical process has been widely used for separation of the metals contained in batteries (Sayilgan et al., 2009;Li et al., 2009;Vatistas et al., 2001;Provazi et al., 2011).The solution from the hydrometallurgical treatment process for battery recycling can be a potentially important source of manganese.
The objective of this work focuses on the electrolytic process for recovering the manganese present in the sulfuric leaching liquor of exhausted batteries.

MATERIALS AND METHODS
The purpose of this study was to obtain electrolytic manganese dioxide from the synthetic solution prepared, taking into consideration results obtained elsewhere (Provazi et al., 2011).Provazi et al. (2011) studied the sulfuric acid leaching of a mixture of household batteries from a municipal collection box.
The solution was prepared by adding the following reagents: solution of titanium and enough Mili-Q water (a water purification system that involves steps of filtration and deionization) to complete a liter of the solution.The amount of each reagent added was calculated to provide the ions and the pH value corresponding to the solution found in a previous study (Provazi et al., 2011).
Table 1 provides the values of concentrations in mgL -1 of metal ions and the final pH of the electrolytic solution.Acid solution TiO 2+ 400

Electrolysis Tests
The cell used during the electrolytic process consisted of a 250 mL beaker, for which a special round plastic lid was created.The lid contains openings to hold: two counter electrodes, a working electrode, a thermometer and a condenser, as shown in Figure 1 and Figure 2. A lead plate 5.0 cm high, 2.0 cm wide and 0.1 cm thick was used as the working electrode (anode) and two graphite plates 2.5 cm high, 1.0 cm wide, and 0.5 cm thick as counter electrode (cathode).Five samples of electrolyte solution with pH 0.30, containing 150 mL each, were anodically electrolyzed during 4 hours, at a temperature of 98 °C (± 2 °C).For each sample, a different range of current density was used, varying from 0.61 to 2.51A.dm - .After each test, the anode was removed from the electrolytic cell and washed with Mili-Q water and dried at 100 °C.EMD deposits were mechanically removed from the anode and were repeatedly washed with Mili-Q water.Afterwards, the EMD was dried at a temperature of 100 °C.Table 2 contains a summary of the tests performed.

Characterization
The characterization of the EMD deposits, collected from the anodes was performed to identify the characteristics of the product obtained.
The following techniques were used: X-ray fluorescence spectrometry, X-ray diffraction, determination of specific surface area by the BET method and scanning electron microscopy (SEM).
As samples were composed of oxides, it was necessary to sputter all samples with gold.Consequently, this element appeared in the spectrum when EDS analysis was carried out.

Current Efficiency
The anodic current efficiency behavior during the electrolytic process was analyzed varying the current density.Figure 3 shows the anodic current efficiency for tests performed at different current densities.
The current efficiency in the process for obtaining manganese dioxide initially rose with increasing current density, achieving a maximum (94.3%), and then decreased.The maximum efficiency was noted at the current density of 1.02A.dm - , which is the value usually adopted in the industrial process for obtaining EMD.A sharp decrease of current efficiency (70.5%) with increasing current density (2.51A.dm - ) was noticed.This variation of current efficiency takes place because, under high current density, the electrolysis of water occurs and the evolution of oxygen can become a competitive reaction in place of the deposition of manganese dioxide, reducing the current efficiency (Ghaemi and Binder, 2002).Another aspect that should be considered is the fact that, under high current density, the rate of deposition is higher than the transport of ions to the electrode and the concentration of metal ions near the anode is reduced (Ghaemi and Binder, 2002).
These results are close to those found by Ghaemi and Binder (2002), who verified that efficiencies higher than 90% (from 91 to 99%) can be obtained at a current density of around 1 A.dm -2 .A current efficiency of 80% was obtained in the range of 1.35 A.dm -2 to 1.39 A.dm -2 and at a very high density such as 3.75 A.dm -2 , the current efficiency decreased to 36%.

Specific Surface Area
The values of specific surface area obtained through the BET method in the EMD samples produced at different current densities, are shown in Figure 4.
The results showed that the specific surface area of EMD grows with increasing current density from 0.61 A.dm -2 (131.54 m 2 .g - ) to 1.02 A.dm -2 (150.85 m 2 .g - ) and the specific surface area of EMD remained almost constant from this point until the current density of 1.93 A.dm -2 (150.85 m 2 .g - , 154.44 m 2 .g - , 145.95 m 2 .g - ).At the current density of 2.51 A.dm -2 , against all expectations, there was a dramatic decrease in specific surface area (128.7 m 2 .g - ).This result may have been due to ineffective control of the variables of the process during the execution of the experiment for this sample.Davis et al. (2001) state that the EMD is a dense solid with significant internal porosity, which leads to a large BET surface area (20 -100 m 2 g -1 ).

Brazilian Journal of Chemical Engineering
The increase in the specific surface area with the current density is presumably due to a high nucleation rate with high current density, which results in many small primary particles (Tamura et al., 1997).

Chemical Analysis
The chemical composition of EMD samples produced at different current densities is shown in Table 3.
The range used for the variation of current density (0.61 A.dm -2 to 2.51 A.dm -2 ) does not significantly affect the content of MnO 2 present in the final products, which showed levels above 90%.Comparing these results with the specifications of commercial electrolytic manganese dioxide, in regard to the concentration of MnO 2 , the results obtained were in accordance with the specifications required for use in battery production where 91% MnO 2 (dry basis) is the minimum concentration of MnO 2 allowed.
Analyzing the impurities present in the results, there was contamination in the materials obtained by the electrolytic process.Such contamination is due to the impurities in the chemicals used for the solution preparation.

X-Ray Diffraction
Figure 5 shows a typical X-ray diffractogram obtained in the tests performed at current densities of 0.61 A.dm -2 , 1.02 A.dm -2 , 1.93 A.dm -2 , 1.39 A.dm -2 , and 2.51 A.dm -2 after a 4-hour period, at a temperature of 98 °C.The X-ray diffraction analyses of all samples presented consistent results for the crystallographic file JCPDS 30-0820, referring to the ε-MnO 2 allotropic form.The ε-MnO 2 phase is the desirable allotropic form of EMD usually used in batteries.

Scanning Electron Microscopy
Figure 6 shows features of the surfaces of particles deposited on the lead electrode at different anodic current densities after a 4 hour period.
The surfaces of EMD samples were covered with fine cracks that appeared when the deposits were cooled after the removal from the hot solution.A greater number of cracks in the deposits were especially observed (Figure 6e) at higher current density (2.51A.dm - ).The increase in the current density caused an increase in the thickness of the deposit.The observed cracks on the surface of the deposit should be a consequence of the handling of a thicker material.
The increase in current density leads to the formation of slightly dendritic deposits, as shown in Figure 6 (d and e).The deposits obtained at lower current density presented a smoother surface aspect, as shown in Figure 6a and 6b; however, the deposit obtained at a current density of 1.02 A.dm -2 showed a structure with fairly smooth appearance (Figure 6b).
Figure 7 shows a typical EDS spectra of the surface of EMD produced at different anodic current densities.
The EDS spectra were similar.Manganese was indicated as the predominant element in the composition of the electroplated material obtained.This fact is compatible with the chemical analysis carried out earlier.
The spectra also present peaks of lead and titanium as contaminants in the EMD.This result indicates that Ti oxide is carried to the deposit during the electrolysis.In all samples, the presence of lead was greater than that of titanium.Lead contamination in the EMD could occur when the deposit is removed.This can be seen in Figure 8, which shows the face of the deposited material that is in contact with the electrode.In this image, the white area (A) is predominantly composed of lead, and a dark area can be seen (B), which is predominantly composed of manganese.

Figure 1 :
Figure 1: Electrochemical cell developed for production of EMD.

Figure 2 :
Figure 2: In detail, the lid of the electrolytic cell.

Figure 3 :
Figure 3: Effect of current density on the anodic current efficiency in tests for the production of EMD.

Figure 4 :
Figure 4: Effect of current density on the surface area of electrolytic manganese dioxide.