In this study, the experiments were conducted using cathode materials provided by SungEel HiTech Co.,Ltd. The used cathode materials were mixed with ammonium sulfate ((NH₄)₂SO₄, 99%, SAMCHUN) at a specific molar ratio and placed in a crucible. The mixture was subjected to reductive roasting at 550^oC for 1 h under a nitrogen atmosphere [13]. After the reaction, the mixture was cooled to room temperature, and water leaching was performed at a liquid-to-solid ratio of 5 mL/g, with stirring at 450 rpm for 1 h. After leaching, the mixture was filtered and washed to separate the residue, which was then dried in a vacuum oven at 120^oC [14–16].

The dried residue was subjected to acid leaching without using a reducing agent, using sulfuric acid (H₂SO₄, 98%, Matsunoen), for 6 h [17]. After leaching, the product was filtered and washed to remove residual carbon and unreacted impurities. Next, sodium hydroxide (NaOH, 30%, SAMCHUN) solution was added to the leachate to increase its pH and thus enable the precipitation and removal of metallic impurities such as Al, Fe, and Cu. To achieve the target molar ratio of Ni:Co:Mn = 6:1:3, stoichiometric amounts of NiSO₄, CoSO₄, and MnSO₄ were added to the purified leachate. Subsequently, a 2 M mixed-metal sulfate solution was formulated and used for the synthesis of NCM precursors.

To recover lithium, sodium carbonate (Na₂CO₃, 99.5%, SAMCHUN) solution was added to the lithium-containing filtrate. The reaction progressed at 90^oC for 1 h, precipitating lithium in the form of Li₂CO₃. The precipitated Li₂CO₃ was filtered and washed with hot distilled water to remove residual SO42− and Na⁺ ions. The final product was dried in a vacuum oven at 105^oC for 6 h [18,19].

A hydroxide coprecipitation method was used to synthesize the recycled Ni_0.6Co_0.1Mn_0.3(OH)₂ precursor. Sodium hydroxide (NaOH, 30%, SAMCHUN) and ammonium hydroxide (NH₄OH, 25–30%, SAMCHUN) were added together with the prepared leachate in a 5 L beaker to maintain the pH of 11.5. The reaction progressed at 55^oC for 8 h under a nitrogen atmosphere. After the reaction, the precursor was filtered and washed with distilled water to remove residual NH₃⁺ and SO42− ions impurities. The washed precursor was dried in a vacuum oven at 120^oC for 12 h. The dried precursor and synthesized Li₂CO₃ were weighed at a molar ratio of 1:1.07 and thoroughly mixed. The mixture was placed in a crucible and calcined in two steps—first, at 500^oC for 5 h, and then, at 880^oC for 12 h—under an oxygen atmosphere to synthesize the LiNi_0.6Co_0.1Mn_0.3O₂ cathode material [20,21].

Phase changes after reductive roasting and the structure of the synthesized Li₂CO₃ were analyzed using X-ray diffraction (XRD, Bruker, D2 Phaser). The leachate composition and impurity content were analyzed both qualitatively and quantitatively using inductively coupled plasma optical emission spectroscopy (ICP-OES, PerkinElmer, Avio 550). The morphology and crystal structure of the synthesized cathode materials were analyzed using scanning electron microscopy (SEM, JEOL, JSM-7610F) and XRD. Cu Kα₁ radiation (λ = 1.54 Å) was used, and scanning was performed in the 2θ range of 10^o–80^o at a rate of 2^o/min.

The electrode slurry was prepared by mixing the cathode materials, a conductive additive (Super-P), and a binder (polyvinylidene fluoride) in a weight ratio of 8:1:1, using N-methyl-2-pyrrolidone as the solvent. The slurry was coated onto aluminum foil, which was used as a current collector, to a thickness of 25 μm using the doctor blade method. The coated electrodes were dried in a vacuum oven at 120^oC for 10 h and punched into circular disks, each with a diameter of 14 mm. Coin cells (CR2032) were assembled under an inert atmosphere inside a glove box using a 1 M LiPF₆ solution, in a mixed solvent of EC:DEC = 3:7, as the electrolyte. Electrochemical performance was evaluated at room temperature within a voltage range of 3.0–4.4 V. The cells were formed at 0.1 C and then subjected to charge–discharge cycling at 1C for 100 cycles [22].

Fig. 2(a) shows the phase transformations of the used cathode materials, as analyzed by XRD, depending on the amount of (NH₄)₂SO₄ added. At a molar ratio of n[(NH₄)₂SO₄] : n(2Li) = 1.2, the amount of (NH₄)₂SO₄ is insufficient to fully convert Li into Li₂SO₄, leading to the partial retention of the layered NCM phase. This result indicates that the addition of (NH₄)₂SO₄ contributes to the decomposition of the cathode’s crystalline structure [11]. As the molar ratio is increased from 1.2 to 1.4, peaks corresponding to the layered NCM phase become undetectable, and new diffraction peaks corresponding to Li₂Mn(SO₄)₂ and water-soluble Li₂- SO₄ become visible. Furthermore, as the amount of (NH₄)₂SO₄ increases, the intensity of the Li₂Mn(SO₄)₂ diffraction peaks increases.

Fig. 2(b) shows the metal leaching efficiency as a function of the molar ratio. As the molar ratio increases from 1.2 to 1.4, the Li leaching efficiency increases drastically from 24.6% to 99.2%. However, when the molar ratio exceeds 1.4, the Li leaching efficiency decreases, while that of Mn continues to increase. This result suggests that the formation of Li₂SO₄ is preferred over that of the sulfates of Ni, Co, and Mn, and that lithium can be selectively extracted using a simple water leaching process. The chemical compositions of the original used cathode materials and water-leached residue, measured using ICP-OES, are presented in Tables 1 and 2, respectively. Li in the water-leached residue was reduced to undetectable levels, while 5.3 wt% of Mn was extracted. This result can be attributed to the formation of water-soluble Li₂Mn(SO₄)₂ during the reduction roasting process, which leads to partial Mn dissolution during water leaching. Although a portion of Mn was leached, the optimal lithium leaching efficiency was achieved at a molar ratio of n[(NH₄)₂SO₄] : n(2Li) = 1.4 : 1.

Following water leaching, acid leaching was conducted using only H₂SO₄, without the addition of any reducing agent, to recover the remaining metals from the residue. As shown in Fig. 3(a), Ni and Co exhibited high leaching efficiencies of 90.13 wt% and 98.26 wt%, respectively. In contrast, Mn exhibited a relatively lower leaching efficiency of 70.34 wt% under the reductive roasting conditions. According to the report by Yang et al. [10], thermodynamic calculations indicate that the stability of metal sulfates follows the order: Li₂SO₄ > MnSO₄ > CoSO₄ > NiSO₄. In particular, Li₂SO₄ can form stably even under low SO₃ partial pressure and exists as a water-soluble compound. MnSO₄ also exhibits greater stability than CoSO₄ and NiSO₄, allowing a portion of Mn to dissolve into the solution. These properties suggest that the relatively low acid-leaching efficiency of Mn may be attributed, in part, to its thermodynamic stability. The removal efficiencies of metallic impurities from the leachate are shown in Fig. 3(b). The impurities were removed by increasing the pH with a NaOH solution to induce precipitation. As a result, Fe was completely removed (100 wt%), while Cu and Al were also effectively eliminated, with removal efficiencies of 98.34 wt% and 97.27 wt%, respectively. However, trace amounts of Cu and Al remained in the leachate.

Fig. 4(a) shows the XRD pattern of Li₂CO₃ synthesized from selectively extracted lithium. Li₂CO₃ was precipitated by adding Na₂CO₃ to the lithium-containing solution. Its characteristic peaks matched well with the standard diffraction pattern (PDF No. 01-083-1454), and no impurity peaks were observed. These results confirm the successful synthesis of Li₂CO₃ and demonstrate the efficiency of lithium extraction and purification [18]. Fig. 4(b) shows an FE-SEM image of the synthesized Li₂CO₃, revealing densely aggregated plate-like single crystals with irregular sizes. This morphology is consistent with previously reported characteristics of Li₂CO₃, where the precipitate forms large agglomerates composed of bar-shaped clusters, further confirming its identity as lithium carbonate [24].

In this study, the leachate was directly utilized without the individual separation or recovery of metal ions for cathode recycling, and the recycled cathode was compared to a cathode synthesized from pure raw materials.

Fig. 4(a) and 4(b) show the FE-SEM images of the virgin and recycled LiNi_0.6Co_0.1Mn_0.3O₂ cathode materials. Both samples display typical NCM-type microstructures, in which primary particles aggregate into secondary particles. Notably, the recycled cathode exhibits a more porous surface compared to the virgin sample, which may contribute to reduced electrochemical performance [25].

Fig. 5 compares the XRD patterns of the two cathode materials. All samples exhibited a well-developed hexagonal α-NaFeO₂ structure (R-3m space group) with no detectable secondary phases or structural distortions. The clear separation of the (108)/(110) peaks indicates high crystallinity in both samples. As summarized in Table 4, the recycled NCM shows a slight decrease in lattice constants (a = 2.8742 Å, c = 14.2468 Å) and unit cell volume (from 101.896 Å to 101.862 Å) compared to the virgin material. Additionally, the (003)/(104) peak intensity ratio decreased from 1.23 to 1.21, indicating increased cation mixing. This structural change is likely due to residual Cu in the recycled cathode, which may substitute for Ni or Mn, reducing structural stability and increasing cation disorder [26].

Fig. 6(a) compares the initial charge–discharge profiles of the virgin and recycled cathode materials in the voltage range of 3.0–4.4 V at room temperature with a current density of 0.1 C. The recycled cathode exhibited a slightly lower initial discharge capacity than the virgin one, with initial Coulombic efficiencies of 96.2% and 96.9%, respectively. This performance degradation is attributed to the residual Al impurity, which is electrochemically inactive. The presence of Al in cathode materials is well-known to reduce both the initial charge-discharge capacity and Coulombic efficiency, which is consistent with the observations in this study [27,28].

Fig. 6(b) presents the cycling performance of the two cathode materials at 1 C. After 100 cycles, the capacity retention of the virgin and recycled samples was 94.8% and 90.1%, respectively. These results imply that residual Al could have negatively affected secondary particle formation during co-precipitation, and Cu may have induced severe cation mixing, as indicated by lattice parameter changes [26,28].

Fig. 6(c) shows the Nyquist plots of the virgin and recycled cathode materials after the initial charge and discharge cycle. The impedance spectra were fitted using an equivalent circuit model comprising R_sf, R_ct, and Z_w, which correspond to surface film resistance, charge transfer resistance at the electrode–electrolyte interface, and Warburg impedance associated with lithium-ion diffusion within the active material, respectively. This model was selected to accurately reflect the electrochemical behavior of layered oxide cathodes after cycling. EIS measurements were conducted using a two-electrode cell configuration, with lithium metal serving as both the counter and reference electrodes. Consequently, the reported R_ct and D_Li⁺ values may include contributions from the lithium electrode. Previous studies [28,29] have reported that this configuration merges the impedance responses of both electrodes, thereby complicating the isolation of the working electrode’s individual behavior. This limitation is acknowledged, and a three-electrode configuration is planned for future studies to enable more accurate analysis. As summarized in Table 4, the virgin cathode exhibited a lower total resistance, including surface film resistance (R_sf) and charge transfer resistance (R_ct), compared to the recycled cathode, which is considered a key factor contributing to its superior electrochemical performance. Moreover, the increase in surface film and charge transfer resistances suggests the formation of a greater amount of non-conductive interfacial films resulting from side reactions with the electrolyte [30,31]. The lithium ion diffusion coefficient (D_Li) was calculated from the low-frequency impedance region using the relationship between Zre and the reciprocal square root of angular frequency, as shown in Fig. 6(d) and Equation (1) [31]. After the 0.1 C formation cycle, the D_Li was determined to be 1.43 × 10^-15 cm²/s for the virgin cathode and 1.29 × 10^-15 cm²/s for the recycled cathode, indicating slightly reduced lithium-ion mobility, likely due to the presence of residual impurities.