1. Introduction
2. Materials and Method
2.1. Materials
2.2. Leaching of black mass
2.3. Precipitation of nickel and cobalt ions as their sulfide form from the leaching solution
2.4. Removal of manganese ions from the sulfide residue
2.5. Conversion of dissolved Mn+ into its solid oxide form
3. Result and Discussion
3.1. Recovery of nickel and cobalt ions as their sulfide form
3.2. Removal of contaminated manganese ions from metal sulfide precipitate
3.3. Recovery of manganese ions in the solid form from Mn ion solution
4. Conclusion
1. Introduction
Li-ion batteries (LIBs) find extensive application in consumer electronics, energy storage systems, and automotive sectors owing to their superior energy density, cell voltage, resistance to memory effect, and favorable cycle longevity. Predominantly, LIB variants incorporating lithium ion manganese-based compounds such as LiMnO2 (LMO), LiMn2O4, Li2MnO3, and lithium nickel manganese cobalt oxides (LiNi1xyMnxCoyO2, NMC) are preferred for electric vehicles and medical devices due to their reduced risk of fire or explosion compared to those utilizing lithium cobalt oxide (LiCoO2)1).
While there’s been significant research on reclaiming lithium, cobalt, and nickel from LIB waste, less attention has been given to retrieving manganese2,3,4,5). The challenge lies in the complex mixture of materials in industrial LIB waste, such as aluminum, iron, and copper, making traditional recycling methods less effective. Unlike manganese minerals, battery waste poses unique obstacles. Additionally, the relatively low cost of manganese contributes to the lack of urgency in developing improved recovery methods.
Furthermore, manganese ion (Mnn+) consistently emerges as a vexing impurity in the recycling of lithium-ion battery (LIB) waste. Its detrimental impact extends notably to the quality of cobalt alloy production and the efficacy of cutting-edge smelting methodologies6). Research elucidated by Granata et al.7) underscores this challenge, demonstrating that even a modest Mn concentration of 3000ppm markedly impedes the selective extraction of cobalt from nickel, employing the prevalent solvent bis (2,4,4-trimethylpentyl) phosphoric acid (Cyanex 272). Hence, imperative strides must be taken in the development of technologies geared towards the recovery or segregation of Mn ion from the coveted components - cobalt, nickel, and lithium.
In the hydrometallurgical treatment of manganese-rich materials, the resulting leach solutions frequently harbor a mix of divalent iron, manganese, copper, nickel, cobalt, zinc, and assorted impurities. Solvent extraction stands as a cornerstone process for the refinement and isolation of manganese from this complex mixture. Over the years, a variety of organic solvent extraction (SX) agents have been employed in fundamental investigations, with particular emphasis on phosphorus acid and carboxylic acid cation exchange reagents. These reagents play a critical role in facilitating the selective extraction and separation of manganese from the solution, thereby contributing significantly to the efficiency and effectiveness of the overall purification process.
Numerous research papers address the separation of manganese ions using solvent extraction methods from NCM (Ni, Co, Mn) leached solutions, wherein manganese ions are extracted into the organic phase containing D2EHPA, while nickel and cobalt remain in the raffinates8,9,10). However, these studies showed that the extraction of cobalt into the organic phase while keeping the manganese in the raffinate is much more difficult. Another reagent that can be used commercially is Cyanex302, [bis-(2,4,4-trimethylpentyl)mono-thio phosphinic acid], and it also possesses the capability to separate cobalt from manganese11,12).
Recently, researchers have developed several solvent extraction (SX) systems using carboxylic acids and oxime reagents13,14). These systems aim to recover nickel and cobalt while separating them from manganese, magnesium, and calcium impurities found in nickel laterite leach solutions. For instance, one system includes Versatic 10 (an alkyl monocarboxylic acid) and LIX63. In this system, there was a notable difference in pH50 values between manganese and nickel-cobalt pairs: 1.96 and 2.53 pH units, respectively. This difference suggests an efficient separation of nickel and cobalt from manganese impurities.
Similarly, the manganese ion is removed from the sulfate-leached solution, resulting in the oxidative precipitation of MnO2 using various strong oxidizing agents. This is because MnO2 itself is a potent oxidizing agent with a standard reduction potential of 1.224. Common oxidizing agents include ozone15), SO2/O2 oxidizing mixure16), Caro’s acid17), peroxy disulfuric acid18), Chloric acid19).
Manganese can be fully precipitated at pH 10.5 as manganese hydroxide by adding sodium hydroxide. However, nickel, cobalt, and other metals are also precipitated as their respective hydroxides along with manganese hydroxide.
The oxidation/precipitation of Mn (in the presence of O2/SO2 proceeds according to (1): At pH < 4
Oxidation by Caro’s acid proceeds according to (2)
oxidation by peroxy-disulfuric acid proceeds according to (3)
oxidation by Chloric acid(HClO3) proceeds according to (4)
Moreover, most studies conducted on the oxidative precipitation of manganese oxide using potassium permanganate focused on applications used for water treatment and industrial waste solutions, such as those from mineral processing. The net ionic reaction for the oxidative precipitation of Mn2+ with permanganate (MnO4-) is shown in Equation 5, which shows that Mn2+ will precipitate as MnO2 in the presence of MnO4−.
In this work, we have removed the manganese ion as impurities from the sulfuric acid leached NC solution using the sulfide precipitation methods on the basis of sulfide solubility of metals at a certain pH. Sodium sulfide is usually added to the leaching solution to maintain certain pH where maximum amount of Ni and cobalt ion get changed into solid sulfide leaving manganese ion into the solution. According to the sulfide solubility diagram mentioned in the research paper20), manganese sulfide is more soluble than other metal sulfides which gives an idea for separation of Mn2+ from other metals Cu2+, Zn2+, Co2+, Ni2+ and Fe2+.
2. Materials and Method
2.1. Materials
In this study, waste NCM black mass with Mn ion 9.9% of the total mass was selected as a raw material. The contents of other metal ions (Ni, Co, Li, Na, K, Ca, Mg, Zn, Fe, Cu, Si, Al, Cr ions) are shown in Table 1. The leaching agent sulfuric acid (98%), precipitating agent sodium sulfide (70%), and Hydrochloric acid (36.5%), and NaOH (50%),was purchased from the Samchun Chemical company Korea.
Table 1.
ICP-OES data of the various elements present in the given black mass
2.2. Leaching of black mass
150 gram of black mass was digested in 25% sulfuric acid with liquid solid ratio (L/S) = 5. The black mass is stirred for 5 hours at 65~70°C and subjected to vacuum filtration to obtain the 1983 g of NCM solution. The sulfuric leached solution was analyzed using ICP-OES by a Perkin Elmer (Avio 500 ICP-OES) to determine the content of metal ions in the solution. The concentration and recoveries of the metal ions in the solution are shown in Table 2. The recovery of all metal ions (Ni, Co, Mn ions) in leachate exceeded 95%.
Table 2.
ICP-OES data of the various elements present in sulfuric acid leached solution
2.3. Precipitation of nickel and cobalt ions as their sulfide form from the leaching solution
After leaching the black mass by sulfuric acid, the leaching solution was treated with varying amount of Sodium sulfide (Na2S.XH2O) to adjust the different pH of the solution. By adding different amount of sodium sulfide in 250 g of leached solution, the pH of resulting solution was maintained from 0.75-7 and then filtered to obtain the Mn ion-rich solution. The filtered Mn ion-rich solution was analyzed by above ICP-OES and a removal manganese metal ion was expressed as percentage, calculated using the following equation:
•Where M is total mass of manganese in the initial leaching solution,
•m is mass of the manganese remaining in the solution after sulfide precipitation.
Based on the data for the highest Mn removal and minimal loss of cobalt and nickel ions, the appropriate pH for Mn ion removal from NCM leachate was determined.
2.4. Removal of manganese ions from the sulfide residue
The insoluble sulfide precipitate, enriched with nickel (Ni), cobalt (Co), and smaller amount of manganese (Mn) ions, was obtained from filtering the solution after precipitating sulfide using sodium sulfide at an optimal pH. This precipitate was subsequently treated with varying concentrations of hydrochloric acid to isolate a solution containing only Mn²⁺ ions. Consequently, the final residue obtained from the filtration of hydrochloric acid-treated solution contains only Ni and Co ions. The residue was dried at 105°C for 5 hours and subjected X-ray diffraction (XRD) analysis to confirm the absence of manganese in the Nickel-cobalt compound.
2.5. Conversion of dissolved Mn+ into its solid oxide form
The hydrochloric acid solution containing Mn and the filtrate from sulfide precipitation at a specific pH was combined, and then 25% NaOH was added until pH reaches 11.0 to precipitate insoluble Manganese hydroxide. Then resulting solution was filtered to get the insoluble manganese hydroxide. The solid Manganese hydroxide was dried at 105°C for 5 hours and analyzed by XRD to determine the form of manganese oxide and to check for contamination by Ni and Co ions in the Mn-compound.
3. Result and Discussion
3.1. Recovery of nickel and cobalt ions as their sulfide form
According to the metal sulfide solubility mentioned in the research paper21), manganese ions are more soluble up to pH 7 compared to nickel and cobalt ions, in its sulfide for. Nickel and cobalt sulfides are very insoluble even at low pH due to their solubility product, whereas manganese sulfide has greater solubility products shown in the Table 321). Therefore; we have optimized the pH to ensure that the maximum amount of nickel and cobalt ions precipitate as their insoluble sulfides, while manganese ions remain dissolved. The pH is controlled by the addition of sodium sulfide, as increased sodium sulfide raises the pH of the solution.
Table 3.
Solubility product (mol/L) of various meal sulfides
Metal sulfide | CuS | ZnS | CoS | NiS | FeS | MnS |
LogKs | Moles per liter (mol/L) | |||||
-47.7 | -25.7 | -22.0 | -21.0 | -18.8 | -13.3 |
The concentration of nickel, cobalt, manganese ions were reduced in solution at the different pH after addition of different amount of sodium sulfide. The transformation into insoluble metal sulfide has been shown in Table 4 and Fig. 1. Using Equation 1, the transformation of nickel, cobalt, and manganese ions into their solid sulfide forms is elegantly detailed in Table 4. Based on the results presented in Table 4 and Fig. 1, the concentration nickel and cobalt ions in the filtrate is gradually decreased with increase of pH. However, the concentration of manganese ions remains high, even at high pH around 7. This data confirms that solubility of MnS is significantly higher than that of Cobalt and nickel ions.
Table 4.
Conversion to insoluble metal sulfide formation (%) and pH change with sodium sulfide addition to NCM leached solution (250 g)
The ICP-OES data confirms that nearly all nickel and cobalt ions precipitate as their sulfides when the solution pH is adjusted above 4.5. From the results shown in Table 4, it can be concluded that the optimal pH is 5.5, at which point fewer manganese ions are lost in the sulfide precipitate while nearly all soluble nickel and cobalt ions convert to their insoluble precipitate forms. By adjusting the pH to 5.5, approximately 86% of manganese ions remain in solution without contamination from Ni, and Co ions. Similarly adjusting pH =4.5 with proper addition of Na2S, we can recover 91% manganese ions although this solution is slightly contaminated with nickel and cobalt ions.
3.2. Removal of contaminated manganese ions from metal sulfide precipitate
The sulfide precipitate so obtained, which contains a small amount of manganese ions, was treated with different concentration of dilute HCl (1%-4.5% by weight) to selectively remove the manganese ions from the metal sulfide. The ICP-OES data of filtrate after the treatment of various concentration of hydrochloric acid are presented in Table 5. The added amount of HCl is determined to be 2.0 times of Mn ions remaining in metal sulfide solids on the basis balance chemical equation 6.
Table 5.
The ICP-OES data (Ni, Co, Mn) of filtrate after treatment of various concentration of hydrochloric acid with the solid metal sulfide obtained at pH 5.5 (Taking initial solution of leaching solution = 250 g)
Finally, total removal of Mn ions from the leaching solution treated with sodium sulfide at pH 5.5 is calculated using following chemical equation (2) and removal percentage of manganese ion are shown in Table 6.
Where:
•m1 = mass of manganese ions present in the filtrate after sulfide precipitation
•m2 = mass of manganese present in the filtrate after treatment of various concentration of hydrochloric acid solution.
•M = total mass of the manganese ion s present in the sulfuric acid leached solution.
Based on the data shown in Table 6 and Fig. 2, removal rate of manganese ions increases with the concentrations of hydrochloric acid solution. However, a significant loss of nickel and cobalt ions in the filtrate solution is observed when the metal sulfide solid is treated with hydrochloric acid concentrations above 3.5%. In this situation, it is concluded that hydrochloric acid concentration of 3.5% is optimal for removing the manganese ions without loss of excessive loss Ni and cobalt ions.
Table 6.
Net removal percentage of nickel, cobalt and manganese ions by treating solid metal sulfide obtained at pH 5.5 with different concentration of hydrochloric acid
No. of obs. | HCl% | Ni | Co | Mn |
Removal (%) | ||||
1 | 1.0 | 0.03 | 0.00 | 86.66 |
2 | 2.0 | 0.11 | 0.04 | 86.76 |
3 | 2.5 | 0.60 | 0.53 | 87.13 |
4 | 3.0 | 0.85 | 0.79 | 89.84 |
5 | 3.5 | 1.16 | 1.13 | 97.87 |
6 | 4.0 | 5.04 | 4.36 | 97.93 |
7 | 4.5 | 6.84 | 5.04 | 97.84 |
The solid sulfide residue obtained after the 3.5% hydrochloric acid treatment is washed with water to remove the soluble Na ions and then dried for 5 hours at 105°C. The dried sample, containing only nickel and cobalt, named D346-R3 and it is analyzed by ICP-OES and XRD to confirm the absence of manganese ions in the sample. ICP-OES result and XRD pattern of the dried sample are shown in Table 7 and Fig. 3 respectively. The ICP-OES data indicate that the final compound (D346-R3) contains only a minimal amount of manganese, confirming the successful removal of manganese ions. The removal of manganese ion from the leaching solution is approximately 99.8%, considering the manganese ions present in the D346-R3.
Table 7.
ICP-OES data with recovery for Nickel-cobalt compound D346-R3 and Manganese compound D346-R4 manufactured by addition of NaOH to the solution containing manganese ion (taking 250 g leaching solution)

Fig. 3.
The comparison of XRD pattern between the final dried sample, (D346-R3) and Siegenite (Ni2CoS4)23).
Similarly, the XRD pattern is fully consistent with Siegenite (Ni2CoS4)22) compound confirming the presence of only nickel and cobalt ions. The small additional the peaks suggest the possibility of moisture contamination or other impurities such as copper sulfide, which can be precipitated easily due to its lower solubility product. Nonetheless, main peaks consistently align with those of Siegenie.
3.3. Recovery of manganese ions in the solid form from Mn ion solution
The filtrate obtained after the sulfide precipitation at pH 5.5 is mixed with another filtrate obtained after the treatment of metal sulfide compounds with 3.5% HCl solution (wt/g). Sodium hydroxide was then added until pH reaches 11.2, to precipitating manganese hydroxide. The solution was filtered, and manganese hydroxide was recovered as the residue while filtrate still contains significant amount of lithium ions. However, no attempt was made to recover lithium ions, as the primary objective was the removal of manganese ions. The light brown color of residue as Mn(OH)2 was also dried in oven at 105°C for 5 hours. After this period, light brown color of the sample turned dark brown as shown in Fig. 4. The sample was analyzed by ICP-OES and XRD to confirm the loss of nickel and cobalt ions in the sample. ICP-OES result and XRD pattern of the dried sample are shown in Table 7 and Fig. 4 respectively.
The ICP-OES data indicate that the final compound (D346-R4), obtained after the removal of manganese ions contains only trace amount of nickel and cobalt ions, confirming the negligible loss these ions during manganese ions removal process. The XRD patterns confirm the manganese compound Mn3O4 (Hausmannite)23,24). The mechanism of Mn3O4formation is explicitly explained in the research paper23). Mn3O4 is essentially a mixture of two manganese oxide MnO and Mn2O3. It is considered that amorphous Mn(OH)2, after the hydrolysis, undergo oxidation in presence of air to form Mn(OH)3. Finally Mn3O4 crystal is formed when these manganese hydroxide go further oxidation. The reaction mechanism is shown in reactions7, 8, and 9.
The XRD result in D346-R4 compound confirmed the absence of any peaks of nickel and cobalt compounds, indicating no significant loss of nickel and cobalt ions.
Based on the ICP- OES data for samples D346-R3 and D346-R4, the recovery percentages of Ni and Co ions in D346-R3, as well as the recovery percentage of Mn ions in D346-R4, were calculated in Table 7 using the following equations:
Where a1, b1, and c1 represent the weights (in grams) of Ni, Co, and Mn ions respectively in sample D346-R3, and a2, b2, and c2 represent the weights (in grams) of Ni, Co, and Mn ions respectively present in sample D346-R4. The data in Table 7 showed that manganese ion removal from the leaching solution is above 99% with negligible loss of nickel and cobalt ion because the recovery rate of both Ni and Co ions were above 99% in the D346-R3 sample.
4. Conclusion
The black mass containing nickel, cobalt, and a significant amount of manganese was purified to remove manganese ions from sulfuric acid leach solutions by exploiting the different solubility properties of various metal sulfides at different pH levels. The optimal pH for this process was found to be 5.5, where nickel and cobalt ions precipitate as insoluble metal sulfides, leaving 84% of the manganese ions in solution due to lower solubility product of manganese sulfide compared to that of nickel and cobalt sulfide. The resulting insoluble metal sulfide, containing nickel, cobalt, and a small amount of manganese, was then treated with hydrochloric acid to remove the remaining manganese ions. Experimental results demonstrated that employing 3.5% hydrochloric acid effectively removes all manganese ions while minimizing the loss of cobalt and nickel ions. This is attributed to the lower lattice energy and higher basicity of manganese sulfide compared to nickel and cobalt sulfides.
The final solid compound, after manganese removal, was analyzed using ICP-OES and XRD. Both analyses confirm a 99.8% removal of manganese ions from the leach solution. The XRD pattern is consistent with Seigenite, containing only NiS and CoS.
Furthermore, the manganese ions in the solution were hydrolyzed with NaOH to form an insoluble hydroxide compound, which was then converted to its oxide form (Mn3O4) by drying in an oven at 105°C for 5 hours. This method effectively removes excessive manganese ions from the sulfuric acid leach solution, facilitating the recovery of pure nickel and cobalt compound. However, caution is necessary due to the formation of hazardous H2S gas, which raises environmental concerns.