Selective Recovery of Cobalt and Nickel from Spent Lithium-Ion Battery NMC Cathodes Using a Hydrophobic Deep Eutectic Solvent

Abstract

A hydrophobic deep eutectic solvent (HDES) composed of Aliquat 336, decanoic acid, and n-hexanol, diluted with kerosene, was investigated for the selective leaching of LiNi_0.33Mn_0.33Co_0.33O₂ (NMC-111) cathode materials. While conventional choline chloride-based DESs co-dissolve Li and transition metals almost completely, the present HDES–acid hybrid system deliberately sacrifices maximum recovery to achieve selectivity. In combination with a low concentration of H₂SO₄, the HDES enabled preferential dissolution of Co and Ni (~84% and ~80% after 6 h at 90 °C, respectively), while Li and Mn largely remained in the solid residue (>93%). Kinetic modeling indicated that the process is controlled by a surface chemical reaction with apparent activation energies of ~~49 kJ mol⁻¹ (for Ni recovery) and ~51 kJ mol⁻¹ (for Co recovery). The leaching residues were enriched in stable Li-Mn-O phases in a way that offers a basis for stepwise recovery. These findings show that hydrophobic eutectic media coupled with mild acid activation provide a sustainable pathway for the selective recycling of LIB cathodes.

1. Introduction

The exponential growth in lithium-ion battery (LIB) production and consumption has led to an urgent need for efficient and environmentally sustainable recycling methods. Among the various components of LIBs, the cathode active material (CAM) contains critical metals such as cobalt, nickel, and lithium, which are economically and strategically important [1,2,3]. Conventional pyrometallurgical approaches, while industrially mature, are energy-intensive and result in loss of lithium and degradation of material quality [4,5,6,7]. As an alternative, hydrometallurgical recycling has received significant attention due to its higher metal recovery efficiency and potential compatibility with decentralized processing [8,9,10,11,12].

A major direction in hydrometallurgical LIB recycling has been the development of deep eutectic solvents (DESs) as greener alternatives to strong mineral acids. DESs, typically formed by mixing a hydrogen bond donor (HBD) and acceptor (HBA), offer advantages such as low volatility, tunable properties, low toxicity, and recyclability. Over the last decade, DESs based on choline chloride (ChCl) have become the most studied systems for leaching cathode materials, with combinations like ChCl:urea [13,14,15], ChCl:glycerol [16,17,18], ChCl:ethylene glycol [19,20,21,22,23] or others, demonstrating metal extraction capabilities across a range of compositions and conditions.

Table 1. Reported Li and Transition Metal Extraction Efficiencies from LIB Cathodes in ChCl-Based DES Systems.

No.	DES Composition	T (°C)	Time (h)	Extracted Metals	Refs.
1	ChCl:Oxalic Acid:H2O	100	0.17	Li: 99.05%, Co: 99.21%	[24]
2	ChCl:Oxalic Acid + microwave	100	0.17	Li: 99.05%, Co: 99.21%	[24]
3	ChCl:Formic Acid + H2O	70	0.17	Co, Li ≈ 100%	[25]
4	ChCl:Lactic Acid	70	24	Li, Ni, Co, Mn ≈ 100%	[26]
5	ChCl:p-toluenesulfonic acid:H2O	90	0.25	Co, Li ≈ 100%	[27]
6	ChCl:Benzenesulfonic Acid: Ethanolamine	90	2	Co: 98%, Li: 99%	[28]
7	ChCl:Malonic Acid:PTSA	100	24	Co: 98.6%, Li: 98.8%	[29]
8	ChCl:Oxalic Acid	120	2	Li, Mn > 96%	[30]
9	ChCl:Citric Acid + 35% H2O	40	1	Co > 98%	[31]
10	ChCl:Ethylene Glycol	180	240	Li: 91.6%, Co: 92.5%, Ni: 94.9%, Mn: 95.5%	[32]

summarizes recent studies that employed DESs for leaching LIB cathodes, showing key operational parameters, and extracted metals.

Regardless of the nature of the hydrogen bond donor—be it lactic acid, oxalic acid, benzenesulfonic acid, or formic acid—and regardless of reaction temperature or duration, all tested systems result in significant or nearly complete lithium recovery from the cathode matrix.

This trend is particularly evident even in mild conditions. For example, the ChCl:formic acid system achieved complete Li and Co extraction in just 10 min at 70 °C [25], and similar outcomes were reported for ChCl: p-toluenesulfonic acid:H₂O [27] and ChCl:oxalic acid [24] systems. Even water-containing systems such as ChCl:CA (citric acid) with 35% water at 40 °C showed over 98% extraction efficiency for cobalt, with lithium co-leached as confirmed [31].

From a mechanistic standpoint, this outcome is expected. The Li–O bonding in layered oxides is more labile than Co–O or Ni–O, owing to the lower lattice energy of Li-containing oxides and the strong solvation affinity of Li⁺ [33]. Acidic or protic DES media promote simultaneous lattice disruption and solvation of Li⁺ and transition metal ions. Redox-stable matrices and hydrophobic domains can reduce Co/Ni mobility by limiting redox cycling and solvation, but, under the strongly solvating conditions of acidic DES, these restrictions are overcome, explaining the good dissolution of Ni and Co. Li, although generally the most labile species, is under our conditions preferentially retained in the solid phase [34,35].

At the same time, retaining lithium in the solid residue simplifies downstream processing by avoiding complex separation from multi-metal leachates. It preserves lithium for targeted recovery under milder, selective conditions and enables stepwise processing where high-value transition metals are extracted first. This approach reduces reagent use, improves selectivity, and supports a more efficient and sustainable resource recovery pathway.

This can be achieved by using hydrophobic deep eutectic solvents (HDES) specifically designed to limit lithium solubility through low polarity, reduced hydrogen-bonding capacity, and poor Li⁺ coordination. Such media selectively solvate and mobilize transition metal ions like Co²⁺ and Ni²⁺, while lithium remains in the solid phase due to unfavorable solvation thermodynamics. HDES have recently been explored in various fields, including the leaching ofnickel laterite ore [36], ytterbium-lutetium oxides [37], and platinum group metals [38].

In a recent study, Sato et al. [39] demonstrated the application of a hydrophobic diluted deep eutectic solvent (HD-DES) composed of methyltrioctylammonium chloride (Aliquat 336), decanoic acid, and n-hexanol in a molar ratio of 1:2:5, diluted with 40 wt% kerosene. This system was successfully used for the solid–liquid extraction of nickel from laterite ores at ambient conditions. Inspired by this approach, we adopt the concept of non-aqueous, hydrophobic DES systems for direct leaching of transition metals from LIB cathodes, while aiming to suppress lithium solubility. As the model material, we used LiNi_0.33Mn_0.33Co_0.33O₂ (NMC-111) cathode powder, which offers a representative composition of transition metals and a well-defined layered structure suitable for evaluating selective leaching and lithium retention. To the best of our knowledge, this is the first study to apply such a strategy to layered NMC cathode materials.

2. Materials and Methods

2.1. Materials

Spent lithium-ion batteries (LG Chem 18,650 cells) were collected from post-consumer electronic waste. Cells were first fully discharged by immersion in a 5% NaCl solution for 24 h to eliminate residual voltage. After drying, the batteries were manually dismantled in a fume hood. The steel casing, separator, and anode were removed, and the cathode foil (Al coated with NMC-based active material) was separated. The cathode material was gently scraped from the aluminum foil using a plastic blade, and then sieved through a 100 µm mesh. The resulting powder was washed with ethanol and deionized water to remove electrolyte residues, and dried at 80 °C under vacuum for 12 h prior to leaching experiments.

All other reagents were of analytical grade and used as received: methyltrioctylammonium chloride (Aliquat 336, ≥95%), decanoic acid (≥98%), n-hexanol (≥99%), sulfuric acid (H₂SO₄, 98 wt%), and kerosene (technical grade).

2.2. Preparation of Hydrophobic Deep Eutectic Solvent

The HDES was prepared by mixing Aliquat 336, decanoic acid, and n-hexanol in a molar ratio of 1:2:5 under constant stirring at 60 °C until a homogeneous clear liquid was obtained. The resulting eutectic was cooled to room temperature, and then diluted with 40 wt% kerosene.

2.3. Leaching Experiments

Leaching experiments were conducted in glass reactors(GG-17, Yuhua Instrument Co., Ltd., Gongyi, China) equipped with reflux condensers and continuous magnetic stirring. For each run, 0.1 g of cathode powder was introduced into a fixed volume (1, 2, 3, or 4 mL) of the prepared HDES diluted with 1 wt% of de-ionized water. A measured amount of H₂SO₄ (0.2, 0.8, and 1.2 M) was added as a proton source to activate metal dissolution. The temperature was controlled at 60, 80, or 100 °C using a thermostatic oil bath. The leaching duration was 6 h; aliquots of the leachate were taken at regular intervals (1 h) for analysis using a micropipette.

2.4. Analysis of Metal Content

The elemental composition of both the leachates and solid residues was analyzed using atomic absorption spectrometry (AAS) with a GBC Savant system (GBC Scientific Equipment Pty Ltd., Melbourne, Australia). Prior to analysis, solid samples were subjected to complete microwave-assisted digestion in a hydrochloric acid and hydrogen peroxide mixture (HCl:H₂O₂) at 80 °C for 2 h to ensure full dissolution. All analyses were carried out in triplicate, with relative standard deviations maintained within ±3%.

2.5. Physicochemical Characterization of HDES

FTIR spectra were recorded using an InfraLUM FT-08 FTIR spectrometer (Lumex, St. Petersburg, Russia) in the range of 4500–500 cm⁻¹. The viscosity of the HDES was measured using a VPZh-4M capillary viscometer (ZIP, Ivanovo, Russia).

3. Results and Discussion

3.1. Characterization of NMC Powder Sample

The XRD pattern of the initial powder sample is presented in Figure 1.

The powder XRD pattern exhibits characteristic reflections at approximately 18.7°, 36.5°, 44.7°, 59.0°, and 65.3° 2θ, which correspond to the (003), (101), (104), (018)/(110), and (113) planes of the layered LiNi_0.33Mn_0.33Co_0.33O₂ phase with a hexagonal structure (space group R$\bar{3}$m). These features confirm the presence of a well-formed layered oxide without significant cation mixing. A peak around 26.5° is also observed, consistent with the (002) reflection of turbostratic graphite.

The chemical composition of the initial powder is presented in

Table 2. Chemical composition of the cathode.

Li (wt%)	Co (wt%)	Ni (wt%)	Mn (wt%)
6.79	19.21	19.13	17.65

.

Thermal treatment of the cathode powder at 620–680 °C for 3 h in a nitrogen atmosphere resulted in a mass loss of approximately 6–7% due to the removal of organic binders and conductive carbon additives. After this treatment, the remaining solid residue is consistent with the stoichiometry of the active phase LiNi_0.33Co_0.33Mn_0.33O₂. Leaching procedures were performed by using the non-treated powder.

3.2. HDES Characterization

The FTIR spectrum (Figure 2) of the freshly prepared HDES (1:2:5 MTAC:decanoic acid:n-hexanol) shows key features confirming hydrogen bonding and component integrity. A broad O-H stretching band at ~3350 cm⁻¹ indicates strong hydrogen bonding. Peaks at ~2925 and ~2854 cm⁻¹ correspond to aliphatic C–H stretching. The sharp band at ~1730 cm⁻¹ is assigned to the C=O group of decanoic acid, confirming its molecular form. Bands at ~1465, ~1385, and ~1160 cm⁻¹ reflect CH₂ bending and C-O/C-N stretching, characteristic of the eutectic structure.

The thermal behavior of the initial HDES was characterized by combined TGA-DSC analysis (Figure 3).

The TGA curve shows a progressive mass loss up to ~300 °C, which corresponds to the volatilization of the organic components of the HDES. Above this temperature no liquid phase remains, and the mass stabilizes. The mass decrease between 250 and 300 °C produced no distinct DSC peak because evaporation occurred gradually and the heat flow was distributed over a broad range. A small exothermic signal is observed below 100 °C, which may be related to residual moisture release or structural rearrangements in the eutectic mixture. A pronounced endothermic peak appears at ~140 °C, associated with accelerated volatilization, while a broader exothermic event above 200 °C is attributed to partial decomposition of organic constituents. The pronounced increase in weight loss occurs at ~200–220 °C.

The variation of density and viscosity of the HDES as a function of temperature is presented in Figure 4.

The density decreases from 0.92 g cm⁻³ at 25 °C to 0.86 g cm⁻³ at 80 °C, exhibiting a slight curvature typical for organic eutectic mixtures. The viscosity demonstrates a strong Arrhenius-type decline, dropping from ~150 mPa·s at 25 °C to ~20 mPa·s at 80 °C, which reflects the pronounced sensitivity of molecular mobility to thermal energy in this solvent system. The modest scatter of experimental points indicates good reproducibility of the measurements. This profile confirms that the pure HDES maintains relatively high viscosity at ambient conditions, but becomes substantially more fluid at elevated temperatures, which can facilitate enhanced mass transfer during leaching.

For the practical leaching tests, the HDES was diluted with 40 wt% kerosene to reduce viscosity and improve mass transfer. The temperature dependence of density and viscosity for this diluted system is shown in Figure 5.

The density decreases from ~0.87 g cm⁻³ at 20 °C to 0.76 g cm⁻³ at 80 °C, while the viscosity drops from ~54 mPa·s to ~10 mPa·s over the same range.

Compared to the undiluted HDES (Figure 4), which exhibits higher density (0.92 → 0.86 g cm⁻³) and significantly higher viscosity (150 → 20 mPa·s), kerosene dilution yields a ~3–4-fold reduction in viscosity across the studied temperature range, without dramatically altering the density–temperature profile. This reduction in viscosity is expected to enhance diffusion of dissolved species, and shorten leaching times, particularly at moderate temperatures where mass-transfer limitations are most pronounced.

3.3. Leaching Results

An initial series of leaching experiments was carried out using the prepared HDES without any added mineral acid, in order to assess its intrinsic ability to dissolve metals from the cathode material. In each run, 0.1 g of cathode powder was treated with 4 mL of the HDES solution, and all experiments were conducted at 80 °C for 4 h, with periodic sampling and subsequent ICP-OES analysis of Li, Co, Ni, and Mn. The results (

Table 3. Time-dependent extraction of metals from NCM cathode in HDES at 80 °C.

Time (h)	Li (%)	Mn (%)	Co (%)	Ni (%)
1	1.2	2.5	15.4	17.8
2	2.5	4.1	26.7	29.3
3	3.6	5.6	33.5	37.2
4	4.5	6.8	38.2	42.7

) show that lithium and manganese exhibited very low dissolution under these conditions, while cobalt and nickel were leached to a greater extent but still insufficiently for practical recovery.

The notably low extraction yields of lithium and manganese in HDES stem from their intrinsic solvation and redox chemistry. Li⁺ ions are strongly solvated and stabilized in highly polar, protic environments. In contrast, hydrophobic DES offer low polarity and lack sufficient proton activity, making Li⁺ transfer into the organic phase highly energetically unfavorable [40]. Manganese (primarily Mn(IV) in NCM materials) is chemically robust and requires proton-mediated oxidative dissolution to break the Mn–O bonds; in neutral or slightly reducing, acid-free DES conditions, such dissolution is negligible [31]. Meanwhile, cobalt and nickel ions, especially in their lower oxidation states, can form weak complexes with components of the DES or undergo partial reduction (e.g., Co³⁺ → Co²⁺), enhancing their solubility even without acid assistance [41].

Although Co and Ni exhibited higher extraction yields than Li and Mn, their final recoveries after 4 h (≈38–43%) were still far from the levels required for efficient downstream recovery. To intensify metal dissolution, mineral acid was introduced as a proton source to disrupt the metal–oxygen framework more effectively and to promote complexation of the released metal ions.

Sulfuric acid was selected as the proton source, due to its low cost, wide industrial availability, and high proton activity in aqueous–organic systems. In addition, sulfate anions can facilitate cobalt and nickel dissolution through the formation of soluble MeSO₄²⁻ complexes, while avoiding chloride-induced corrosion issues associated with HCl. A small amount of de-ionized water (1 wt%) added to activate proton transfer and facilitate metal–oxygen bond cleavage while maintaining low polarity to restrict Li⁺ solvation [42].

Time-dependent leaching behavior of Li, Mn, Co, and Ni from NCM cathode material in HDES containing 0.2 mol L⁻¹ H₂SO₄ at 100 °C and L:S = 40:1 mL:g is presented in Figure 6.

Li and Mn exhibited very limited dissolution, reaching only ~7% and ~12% extraction, respectively, after 6 h. In contrast, Co and Ni showed a progressive increase in recovery over time, attaining ~58% and ~52% after 6 h. The pronounced gap between Co/Ni and Li/Mn extraction confirms the high selectivity of the system under low-acid conditions, while highlighting the need for higher acid concentrations or adjusted parameters to achieve complete transition-metal recovery.

The effect of sulfuric acid concentration on metal leaching was investigated at 100 °C and a liquid-to-solid ratio of 40:1 mL:g, using acid concentrations of 0.2, 0.8, and 1.2 mol L⁻¹ (Figure 7).

For Ni (Figure 7a), the extraction increased markedly when the acid concentration was raised from 0.2 to 0.8 mol L⁻¹, with the final recovery after 6 h rising from about 52% to approximately 80%. A further increase in acid concentration to 1.2 mol L⁻¹ resulted in only a marginal improvement, reaching ~82%, indicating that 0.8 mol L⁻¹ H₂SO₄ was sufficient to achieve near-maximum Ni dissolution under the tested conditions.

For Co (Figure 7b), increasing the acid concentration from 0.2 to 0.8 mol L⁻¹ led to a pronounced rise in extraction, with the 6 h recovery improving from about 58% to approximately 85%. Further acid increase to 1.2 mol L⁻¹ produced only a negligible gain, reaching ~86%. This confirms that 0.8 mol L⁻¹ H₂SO₄ was effectively sufficient to achieve near-complete Co dissolution under the given conditions.

Li and Mn exhibited only minor dissolution under all tested acid concentrations, with final recoveries not exceeding 4% and 6%, respectively, after 6 h at 100 °C.

Subsequently, the effect of temperature on metal leaching was investigated at an acid concentration of 0.8 mol L⁻¹ and a liquid-to-solid ratio of 40:1 mL:g, using temperatures of 60, 80, 90, and 100 °C (Figure 8 and Figure 9).

For Ni, raising the temperature from 60 to 80 °C substantially increased extraction, with the 6 h recovery rising from ~60% to ~79% (Figure 8).

A further increase to 90 °C brought the recovery to ~81%, and heating to 100 °C resulted in only a marginal gain (~82.5%), indicating that near-maximum dissolution was already achieved at 90 °C.

For Co, extraction improved from ~65% at 60 °C to ~74% at 80 °C, reaching ~85% at 90 °C (Figure 9). Increasing the temperature to 100 °C yielded only a slight further increase (~86.5%), confirming that 90 °C was sufficient to approach the maximum achievable recovery under these conditions.

The L:S ratio strongly influences reagent availability, mass transfer, and metal concentration in the leachate. Therefore, we tested L:S values of 10:1, 20:1, and 40:1 mL:g under 0.8 mol L⁻¹ H₂SO₄ at 90 °C to balance recovery and solution concentration (

Table 4. Recovery of Co and Ni (%) at different L:S ratios over time (0.8 mol L⁻¹ H₂SO₄, 90 °C).

Time (h)	L:S = 10:1	L:S = 20:1	L:S = 40:1	L:S = 10:1	L:S = 20:1	L:S = 40:1
	Co	Co	Co	Ni	Ni	Ni
1	33.2	39.4	39.8	30.2	35.1	35.9
2	50.4	59	60.2	45.5	53.1	54.3
3	61.6	71.1	72.4	55.6	65.3	66.7
4	68.3	78.1	79.4	61.7	72.9	74.3
5	72.4	81.7	82.9	65.4	77.1	78.5
6	75.0	83.9	85.1	70.3	79.6	80.8

).

At 0.8 mol L⁻¹ H₂SO₄ and 90 °C, increasing the L:S ratio from 10:1 to 20:1 markedly improved cobalt and nickel extraction at all leaching times. After 6 h, Co recovery rose from 75.0% to 83.9% and Ni recovery from 70.3% to 79.6%. A further increase to 40:1 produced only marginal gains, reaching 85.1% for Co and 80.8% for Ni, indicating that 20:1 is close to optimal for balancing high recovery with a sufficiently concentrated leachate.

It should be emphasized that the observed selectivity is not solely a property of the HDES itself, but rather of its combination with a low concentration of H₂SO₄. The hydrophobic eutectic medium suppresses Li⁺ solubility, while the mineral acid provides the necessary proton activity for disrupting the NMC lattice and mobilizing Co/Ni. In contrast to conventional aqueous leaching, the acid dosage is reduced by an order of magnitude, thereby mitigating environmental impact while retaining high transition-metal recovery efficiency.

Thus, sequential variation of acid concentration, temperature, and liquid-to-solid ratio showed that high Co/Ni recovery can be achieved while keeping Li and Mn dissolution low. Increasing H₂SO₄ concentration from 0.2 to 0.8 mol L⁻¹ markedly enhanced Co/Ni leaching, whereas a further increase to 1.2 mol L⁻¹ gave negligible improvement. Raising temperature from 60 to 90 °C significantly accelerated dissolution, with near-maximum yields reached at 90 °C; heating to 100 °C offered only marginal gains. Increasing the L:S ratio from 10:1 to 20:1 mL:g substantially improved recovery, while 40:1 provided only a slight additional effect and diluted the leachate.

Under these optimal conditions—0.8 mol L⁻¹ H₂SO₄, 90 °C, L:S = 20:1 mL:g, 6 h—the process delivered about 84% cobalt and 80% nickel recovery, while lithium and manganese remained low, at under 6–7%.

Compared to conventional ChCl-based DES systems, which typically achieve nearly complete co-dissolution of Li and transition metals under mild conditions, the present HDES system delivers slightly lower absolute recovery of Co and Ni (~80–85%). However, the key advantage lies in its ability to retain Li in the solid residue, thereby enabling stepwise recovery. In this sense, the method deliberately sacrifices maximum overall extraction in favor of enhanced selectivity, reducing the burden of downstream separation and improving process sustainability.

3.4. Leaching Kinetic Analysis

In leaching kinetics studies, the shrinking core model (SCM) is among the most widely applied approaches [43,44,45]. In this model, leachable particles are idealized as uniform spheres whose radius gradually decreases as the solid phase dissolves into the solution. The relationship between the fraction of solid dissolved and the leaching time is expressed by different equations, depending on the rate-controlling step of the process [46]:

$$1-{\displaystyle \frac{2}{3}}{X}_{Me}-(1-X_{Me}{)}^{{\displaystyle \raisebox{1ex}{$2$}\!\left/ \!\raisebox{-1ex}{$3$}\right.}}=k\mathsf{\tau }$$

(1)

$$1-(1-X_{Me}{)}^{{\displaystyle \raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$3$}\right.}}=k\tau$$

(2)

$${\displaystyle \frac{1}{3}}\ln \left(1-X_{Me}\right)-1+(1-X_{Me}{)}^{{\displaystyle \raisebox{1ex}{$-1$}\!\left/ \!\raisebox{-1ex}{$3$}\right.}} =k\tau$$

(3)

where $X_{Me}$ denotes the fraction of the solid phase dissolved, $k$ is the leaching rate constant for the chemical reaction, and $\tau$ is the time required to reach the specified $X_{Me}$.

The presented equations describe leaching processes controlled by different rate-limiting mechanisms: diffusion through the product layer (Equation (1)), surface chemical reaction (Equation (2)), and a mixed control involving both diffusion and surface reaction (Equation (3)). Although an L:S ratio of 20:1 (mL/g) was identified as optimal for leaching, the kinetic evaluation was carried out at an increased ratio of 40:1 (mL/g) to ensure stable acid concentration and to minimize possible deviations from the assumptions of the SCM.

To determine the controlling step, the left-hand sides of Equations (1)–(3) were plotted against leaching time at three temperatures, using the data from Figure 8 (Ni) and Figure 9 (Co). The analysis was restricted to the initial 4 h of leaching in order to reduce the influence of acid depletion, solubility limitations. The corresponding determination coefficients are summarized in

Table 5. Determination coefficients (R²) for the linear fits based on Equations (1)–(3), calculated from the data presented in Figure 8 (Ni recovery).

Equation	Temperature, °C
	60	80	90
1	R2 = 0.7095	R2 = 0.6371	R2 = 0.6849
2	R2 = 0.9641	R2 = 0.9263	R2 = 0.9572
3	R2 = 0.6460	R2 = 0.7516	R2 = 0.6993

and

Table 6. Determination coefficients (R²) for the linear fits based on Equations (1)–(3), calculated from the data presented in Figure 9 (Co recovery).

Equation	Temperature, °C
	60	80	90
1	R2 = 0.6835	R2 = 0.7219	R2 = 0.7105
2	R2 = 0.9547	R2 = 0.9063	R2 = 0.9342
3	R2 = 0.5952	R2 = 0.6834	R2 = 0.6711

.

Based on the data in Table 4 and Table 5, it can be concluded that the surface chemical reaction model (Equation (2)) provides the highest determination coefficients across all temperatures for both Ni and Co recovery; this indicates that the leaching process is predominantly controlled by the surface reaction rather than by diffusion or mixed control mechanisms.

To evaluate the kinetic parameters of Ni and Co leaching, the data from Figure 8 and Figure 9 were analyzed using Equation (1), and the corresponding plots of $1-(1-X_{Me}{)}^{{\displaystyle \raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$3$}\right.}}$ versus leaching time at three temperatures are shown in Figure 10a,b.

The kinetic plots for Ni (Figure 10a) and Co (Figure 10b) obtained at an L:S ratio of 40:1 (mL/g) and evaluated within the initial 4 h exhibited satisfactory linearity with the surface chemical reaction model. For Ni, the apparent rate constant increased from 0.0417 h⁻¹ at 60 °C to 0.1223 h⁻¹ at 80 °C and 0.1755 h⁻¹ at 90 °C. In the case of Co, k rose from 0.0399 h⁻¹ at 60 °C to 0.1099 h⁻¹ at 80 °C and 0.1849 h⁻¹ at 90 °C. The consistently high determination coefficients (R² > 0.94) support the applicability of the surface reaction model.

Arrhenius plots for Ni and Co leaching from NCM cathodes in HDES were constructed using the calculated rate constants (Figure 11). The linear fits indicate that the dissolution of these metals is governed by a surface-controlled chemical reaction.

Both metals show linear correlations with high coefficients of determination (R² = 0.9992 for Ni and 0.9953 for Co), indicating that the process follows Arrhenius behavior. The apparent activation energies, derived from the slopes (−Ea/R), are approximately 48.9 kJ mol⁻¹ for Ni and 51.1 kJ mol⁻¹ for Co. These values fall within the range typical of chemically controlled leaching reactions and are in agreement with the kinetic analysis based on the shrinking core model. The slightly higher activation energy for cobalt reflects the stronger Co–O bonding compared with Ni–O in the layered oxide structure, which requires more energy to disrupt during dissolution.

3.5. Leaching Residue Analysis

Elemental composition of leaching residue, obtained under the optimized conditions of 0.8 mol L⁻¹ H₂SO₄ in HDES at 90 °C, L:S = 40:1 mL:g, and 6 h leaching time, was, wt%: Li—10.3; Mn—26.5; Co—4.9; Ni—6.1.

The XRD pattern of the leaching residue (Figure 12) indicates the presence of Li-Mn-O phases, including Li₂MnO₃ and MnO₂, as well as minor peaks corresponding to Li₂CO₃.

Notably, the characteristic graphite (002) reflection at ~26.5° 2θ is absent, which may be attributed either to the removal of graphite from the solid residue or to its significant modification, resulting in a highly disordered and dispersed carbon phase, likely amorphous in nature after extended treatment in the HDES medium. No distinct crystalline phases of Co or Ni are observed, despite their confirmed presence in the residue by elemental analysis. This discrepancy is likely due to the formation of poorly crystalline hydroxides or amorphous oxyhydroxides of Ni and Co, which do not give rise to well-defined Bragg reflections in XRD patterns, and may also be explained by the incorporation of Ni and Co into manganese phases as solid solutions due to the chemical similarity of these ions.

The persistence of Mn in the residue should not be considered a drawback but rather a feature of the stepwise recovery strategy. Concentrating Mn together with Li in stable Li-Mn-O phases enables their targeted extraction in a subsequent leaching stage under more oxidative or alkaline conditions. This sequential approach reduces reagent consumption in the primary step while ensuring that Mn can be recovered efficiently without interfering with Co/Ni separation.

4. Conclusions

A hydrophobic deep eutectic solvent (HDES) composed of Aliquat 336, decanoic acid, and n-hexanol, diluted with kerosene, was applied for the selective leaching of spent LiNi_0.33Mn_0.33Co_0.33O₂ cathodes. Selectivity arose from the combined action of HDES and a low concentration of H₂SO₄: the acid provided protons for lattice disruption, while the hydrophobic medium suppressed Li⁺ solubility. Under optimized conditions (0.8 mol L⁻¹ H₂SO₄, 90 °C, L:S = 20:1), ~84% Co and ~80% Ni were recovered within 6 h, whereas Li and Mn extraction remained below 7%.

Compared with conventional ChCl-based DESs, which dissolve all metals nearly completely, this approach deliberately sacrifices maximum recovery in favor of higher selectivity, simplifying downstream processing and reducing acid use. Kinetic analysis confirmed surface reaction-controlled dissolution with moderate activation energies (~33 kJ mol⁻¹). The residue was enriched in Li-Mn-O phases, enabling their targeted extraction in a subsequent step.

To our knowledge, this is the first application of the Aliquat 336/decanoic acid/n-hexanol system, diluted with kerosene, for the selective leaching of Co and Ni from spent LIB cathodes. In a wider perspective, the approach can be scaled up because the organic components are low-cost, commercially available, and the solvent can be regenerated after solid–liquid separation by removing the dissolved metal ions, enabling its reuse in subsequent leaching steps. Moreover, the use of only moderate additions of H₂SO₄ and the reduced discharge of acidic effluents indicate a smaller environmental burden compared with conventional hydrometallurgy, which also translates into lower neutralization and wastewater treatment costs. These combined factors make the method a practical candidate for sustainable large-scale LIB recycling.