Pomegranate peel as an ecofriendly reductant for efficient recovery of cathode materials from spent lithium-ion batteries through organic acid leaching

Abstract

Nowadays, the recycling of valuable metals from spent lithium-ion batteries (LIB) using hydrometallurgical processes is on the rise. Therefore, using safer leaching media for the dissolution process and the use of environmentally friendly chemicals are essential. In this research, a feasibility study of the recovery of valuable metals, including lithium, cobalt, manganese, and nickel, from the cathode of spent LIB in organic acid using pomegranate peel (PP) as a green reductant was evaluated. To optimize the dissolution process, the effects of various parameters, such as CH₃COOH concentration, temperature, time, and the PP/LIB, were investigated using response surface methodology (RSM) experimental design. Based on ANOVA, linear models were obtained for Co, Mn, and Li recovery, except for the Ni recovery model. According to the results, temperature was identified as the most crucial factor in the recovery of Ni, Co, Mn, and Li, due to the facile hydrolysis of PP at elevated temperatures. Based on the optimization results for leaching cathodes of LIB in the presence of PP, the predicted recovery values for Ni, Co, Mn, and Li at the optimum condition of 320 min, 5.5 M CH₃COOH concentration, 92 °C temperature, and PP/LIB of 3.5 g/g were 83.3, 85.9, 84.9, and 91.2%, respectively, which showed good agreement with the calculated results from the experimental recovery. FTIR characterization of leaching samples in the presence and absence of PP indicated that glucose molecules produced during dissolution were immediately consumed by LIB compounds. Furthermore, the leaching kinetics study of LIB cathodes in the presence of PP demonstrated that the dissolution mechanisms of Ni, Co, Mn, and Li followed the Avrami model.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-04351-3.

Introduction

In recent decades, the number of spent lithium-ion batteries (LIB) in electronic devices and vehicles has significantly increased. The global LIB market was roughly $48.80 billion in 2022, and it is predicted to reach around $184.15 billion by 2030, considering an annual growth rate of 18.5%^1,2. Various types of batteries, including LCO, NMC, and LPF, consist of 25 to 41% cathode active material, 4–7% aluminum foil, 7 to 17% copper foil, 15 to 18% anode active material, 3% separator, 10 to 16% electrolyte, and 17 to 27% housing materials^3–5. The cathode material accounts for the highest weight, reaching approximately $22 billion in the global market in 2022. It is also projected to grow by 6.3% to approximately $40.6 billion by 2032⁶. Meanwhile, the demand for metals to produce LIB is also increasing. The production of Ni, Co, Mn, and Li for LIB consumption in 2023 was 878, 596, 147, and 207 thousand tons, respectively. According to forecasts by 2032, the quantities for Ni, Co, Mn, and Li will reach 2,390, 1,299, 187, and 687 tons, respectively⁷. Considering the average lifespan of LIB, typically 3 to 5 years, the recycling of metals and the sustainable supply of metals are crucial⁸. Since global environmental policy is moving towards reducing the carbon footprint in various industries⁹, it is essential to adopt green methods to transition to renewable energy and ensure the sustainable supply of raw materials for LIB.

In a common hydrometallurgical recycling process for LIB, waste components such as cathode and anode materials are typically separated from the battery housing and separator through mechanical shredding and grinding¹⁰. Acids are often used to dissolve cathode materials into metal ions¹¹. Major recycling companies, including Recuply, Umicore, and Accurec, utilize strong mineral acids such as H₂SO₄ and HCl¹². In this regard, the dissolution of Li ions is strongly associated with acidic strength. At the same time, reducing agents like H₂O₂, glucose, and NaHSO₃ are employed to assist in reducing cobalt and manganese from higher to divalent oxidation states^10,13.

During the recycling process, the use of organic acids with the assistance of green reductants can be a key to sustainability in the recycling of LIB cathodes¹². In the discussion of recycling cathodes of LIB, sulfuric acid is used more than other acids due to its low cost and availability. Approximately 200 million tons of sulfuric acid are produced annually, with a price of $120 per ton^14,15. However, it should be noted that sulfuric acid, as a mineral acid, has very low biocompatibility and can cause groundwater pollution if not properly handled. Organic acids, in comparison to inorganic acids (e.g. sulfuric acid), exhibit excellent biodegradability, lower acidity, and minimal corrosion. Various organic acids have been employed to extract metal content in LIB cathodes^16–19. Among organic acids, acetic acid is considered one of the most economical with a price of $365 per ton in 2023, with an approximate production volume of 18 million tons per year^20,21. Therefore, replacing sulfuric acid with acetic acid can be a sustainable solution in the spent LIB recycling process, considering the environmental issues associated with these acids^22,23. However, a few researches have been conducted on metal extraction from LIB cathodes in an acetic acid medium. In a study by Gao et al., optimal conditions, including 3.5 M acetic acid, S/L ratio of 40 g/L, temperature of 60 °C, and 4% vol. H₂O₂, achieved dissolution efficiencies exceeding 92% for all metals (Ni, Co, Mn, and Li) within 5 min²⁴. However, the industrial use of the costly substance H₂O₂ as a reductant poses a potential challenge that necessitates further consideration in this field. The weak dissolution power of CH₃COOH highlighted in the research by Gao et al.²⁴ has been emphasized. Ni, Co, Mn, and Li dissolution in a CH₃COOH solution, along with 2.0 M ascorbic acid and bagasse pith as a reducing agent of 3.0 g/L concentration, achieved dissolution efficiencies of 61% Ni, 60% Co, 63% Mn, and 65% Li within a 3-min duration²⁵. Using ascorbic acid alongside acetic acid has enhanced recovery, but the dissolution mechanism in acetic acid medium has not been investigated. It has also been reported that ultrasonic assistance increased the collision frequency, leading to improved efficiencies for Ni, Co, Mn, and Li in a CH₃COOH²⁵. While Liang et al.²⁶ achieved high efficiencies of over 98% for all elements (Ni, Co, Mn, and Li) in an acetic acid environment using a pressure reactor at a temperature of 200 °C, it seems that temperature acts as a crucial parameter in reductive leaching processes. However, the use of high pressures for metal extraction incurs significant costs, challenging the economic viability of such processes. In another study, a temperature change from 40 to 70 °C resulted in increased leaching efficiencies (Co from 40 to 98% and Mn from 49 to 98%), and at higher than 70 °C no significant increase in leaching recovery was observed. The research group attributed this to the instability of acetic acid at higher temperatures. However, the lack of recovery increase at temperatures higher than 60 °C may be attributed to the instability of H₂O₂ as a reductant²⁷. Kazazi et al.²⁸ systematically examined the influence of acetic acid concentration (1–6 M) on spent LIB cathode leaching under fixed conditions (4 h, 80 °C, 20 g/L tangerine peel (TP)). Their work revealed a distinct dichotomy in metal recovery behavior. While lithium extraction showed strong acid-dependence - increasing from 62.51% (1 M) to 73.15% (6 M) due to enhanced proton (H₃O⁺) availability - the leaching efficiencies for cobalt, nickel and manganese remained consistently low (< 10%) across the entire concentration range, despite the presence of TP’s organic reducing compounds. Table 1 summarizes the literature review on the effects of acetic acid in the leaching of the LIB cathodes.

Table 1.

Summary of acetic acid’s effects on lithium-ion battery cathode leaching.

Study	Conditions	Recovery Efficiency	Key Findings
Gao et al.24	3.5 M CH3COOH, 40 g/L S/L, 60 °C, 4% H2O2, 5 min	> 92% (Ni, Co, Mn, Li)	High efficiency but relies on costly H2O2; Limited efficiency of CH3COOH without reductants.
Yan et al.25	CH3COOH + 2.0 M ascorbic acid + 3.0 g/L bagasse pith, 3 min	61% Ni, 60% Co, 63% Mn, 65% Li	Ascorbic acid improves recovery; mechanism of CH3COOH leaching was not studied.
Liang et al.26	CH3COOH, pressure reactor, 200 °C	> 98% (Ni, Co, Mn, Li)
Natarajan et al.27	CH3COOH, 40–70 °C (H2O2 as reductant)	Co: 40→98%, Mn: 49→98%	Instability of H2O2 at high temps limits recovery.
Kazazi et al.28	1–6 CH3COOH, 80 °C, 4 h, 20 g/L TP	Li: 62.51% (1 M) → 73.15% (6 M)	Higher acidity aids Li dissolution; No effect on Ni/Co/Mn recovery.

Adding a reducing agent to convert cobalt (III) to cobalt (II) and manganese (IV) to manganese (II) in the leaching process seems to be essential^10,16,29. Various reducing agents have been employed to extract valuable metals from LIB cathodes^30–32. In a study where green reducing agents, namely glucose, sucrose, and cellulose, were employed in the leaching of LIB cathodes, evidence indicates an enhancement in the dissolution of Ni, Co, and Mn with the addition of glucose and sucrose. However, cellulose was found to have a minor effect on the dissolution of Ni, Co, and Mn, indicating that cellulose compounds are not easily decomposed at 80 °C²⁶. This result strongly supports the need for higher temperatures to hydrolyze these compounds. The addition of orange peel in the leaching process of LIB cathodes in 1.5 M citric acid at a temperature of 100 °C resulted in a recovery of approximately 80–99% for Ni, Co, Mn, and Li. Therefore, the use of fruit peels for providing reductive conditions can be considered a green solution for extracting metals from LIB cathodes. Hence, in this study, acetic acid was used as an inexpensive and environmentally friendly acid for dissolving LIB cathode materials. Pomegranate peel (PP) was utilized as a reducing agent, and the recovery of metal extraction was optimized using a response surface methodology (RSM) experimental design. On average, three million tons of pomegranates are produced annually in various countries, including Iran, Tunisia, Turkey, Spain, and the USA, with the majority of it utilized for pomegranate juice production. The waste generated from pomegranate juice production can be utilized as a sustainable source for LIB cathode dissolution³³. Finally, characterization of the leaching residue and liquor was also carried out, and the dissolution mechanism was investigated with the kinetic Avrami model.

Experimental

Materials and reagents

3 kg of different cell phone batteries were obtained from a battery collector. Then, different parts of the anode, cathode, separator, and shell were separated manually. Regarding quantity, 33.9% was related to metal housing, 10.1% to plastic, 21.4% to anode and 34.6% to cathode material (containing Nickel Cobalt Manganese Oxides (NCM)). Acetic acid (CH₃COOH) and sodium hydroxide (NaOH) used for leaching were analytical grade chemicals from Merck Company, with a purity of 98.0 and 99.5%, respectively. Following a 24-hour drying period at 70 °C, the pomegranate peels (PP) were ground using a grinder and utilized in the experiments in their original state.

Characterization and analysis

The phase analysis of the spent LIB and leaching residue was characterized via X-ray diffraction (Philips PW-3710 XRD equipment with CuK_α beam and an accelerating voltage of 400 kV). Also, X’pert HighScore Plus software version 2.2b (2.2.2) was used to determine the phases. The composition of solutions resulting from the leaching process was measured using an atomic absorption spectrometer (AAS, Varian 240) and ICP-OES (Varian, Zarazma Co., Tehran, Iran). The chemical characteristics of the leaching solution in the presence and absence of PP were analyzed by Fourier transform infrared spectroscopy (FTIR, PerkinElmer Spectrum version 10.02). The particle size distribution of LIB and PP powder was determined by a laser particle size analyzer (Fritsch, Germany).

Preparation of LIB cathode and PP powder

To prevent short-circuiting between the anode and cathode of the spent batteries during disassembly, all batteries were immersed in 10% NaCl solution for 24 h. After the electrical discharge of the spent batteries, the cathode, anode, and plastic parts were manually separated from each other. In the next step, a 7.0 M NaOH solution was used to remove the aluminum foil in the LIB cathode according to reaction 1. The alkaline leaching of aluminum was carried out for 30 min. To reduce the dissolution of other elements (Ni, Co, Mn, and Li), the alkaline dissolution process of aluminum was performed at room temperature. According to the solution analysis, a negligible amount of metals in the cathode powder dissolved in the NaOH solution, which was previously observed as well³⁴.

1

After the completion of the alkaline leaching process, the solid powder was separated using vacuum filtration from the liquid. The powder of LIB cathode underwent multiple washes with deionized water to ensure no soluble aluminum compounds remained. In the next step, the remaining LIB cathode from the alkaline leaching step was placed at 600 °C for 6 h to remove the moisture content and its volatile substances as well as carbon introduced from the anode section³⁵. To get the chemical analysis of the LIB cathode, 1 g black powder was digested in 100 mL aqua regia (HCl + HNO₃). It was analyzed using Inductively Coupled Plasma mass spectrometry (ICP-MS, Vista-Pro, Varian Co.). The LIB cathode contains 26.6% C, 18.1% M, 7.20% N, and 6.1% L. It is worth noting that the aluminum content in the cathode powder is less than 0.7%. The preparation of PP involved initially drying pomegranate peels in the oven for 24 h at a temperature of 110 °C, followed by crushing and finely milling them. The particle size of the LIB cathode and PP was in the range of 0.01–53.20 μm (with D80 = 11.16 μm) and 80–350 μm (with D80 = 195 μm).

Leaching experiments based on RSM

The optimization of leaching of the LIB cathode was carried out using the RSM based on the Central Composite Design (CCD) method. Generally, the CCD method involves conducting 2^k + 2k + N₀ experiments, where k is the number of parameters under investigation. The number 2^k corresponds to factorial design experiments, 2k to center-point experiments, and N₀ to the number of center-point experiments^36,37. Repetition of center-point experiments aims to determine laboratory error. In the leaching of the LIB cathode, four parameters were investigated at five levels. To minimize systematic errors, experiments were conducted randomly. The selected parameters and their levels are presented in Table 2, based on previous research^24,25,27. According to preliminary tests, it has been observed that stirring speed did not significantly affect the dissolution of LIB cathode in acetic acid, and therefore, it was excluded from the RSM design. The DX 13.0.5.0 software was used for RSM experimental design calculations. The value of α, equal to 2.0, was employed to determine the parameter levels outside the range. Consequently, the minimum levels for time, CH₃COOH, temperature, and PP/LIB were obtained as 20 min, 0.5 M, 35 °C, and 0.13 g/g, respectively. In comparison, the maximum levels were 420 min, 10.5 M, 95 °C, and 2.63 g/g, respectively.

Table 2.

Selected parameters and their levels.

Factor	Name	Units	–α	-1	0	1	+α
A	time	min	20	120	220	320	420
B	CH3COOH	M	0.5	3	5.5	8	10.5
C	Temperature	°C	35	50	65	80	95
D	PP/LIB	g/g	0.13	0.75	1.38	2	2.63

To conduct the leaching of the LIB cathode in CH₃COOH, initially, 50 mL of acetic acid as the specified concentration according to the RSM experimental design was prepared using deionized water. Subsequently, the solution was transferred into a flask. A cold-water condenser was employed to minimize solution evaporation. At this point, the solution temperature was adjusted to the desired level according to the RSM experiment design using an oil bath for precise temperature control. Temperature was controlled using a thermometer with an accuracy of 0.2 °C. Following this step, 1 g of LIB cathode powder was introduced into the 50 mL solution. The addition of the pomegranate peel as reducing agent to the solution was made based on the desired PP/LIB ratio based RSM design. During the leaching process, the solution was stirred at a constant speed of 700 rpm by a magnetic stirrer. Once the leaching process duration was completed, the solution was separated from the leaching residue using vacuum filtration. Subsequently, the solution was diluted and analyzed to determine metal recovery. The leaching residue from the experiment conducted under optimal conditions was washed multiple times with deionized water to entirely remove soluble materials. The recovery of Ni, Co, Mn, and Li was determined using the following Eq. 

2

In which C_i (g/L) is the concentration of metal i (i.e., Ni, Co, Mn, and Li) in the leaching solution, m (g) is the mass of LIB cathode, V_f (L) is the volume of leaching solution, x_i (%) is the mass fraction of metal i in the LIB cathode, and X (%) is the leaching efficiency.

Results and discussion

Results of LIB leaching with PP using RSM

The RSM method was employed to optimize the leaching parameters of the LIB cathode, in the presence of PP, to enhance the recovery of Ni, Co, Mn, and Li. Table 3 presents the recovery of Ni, Co, Mn, and Li in acetic acid medium in 28 experiments.

Table 3.

Recoveries of Ni, Co, Mn, and Li in the leaching process of LIB cathode based RSM design.

Run	Time (min)	CH3COOH (M)	Temp. (°C)	PP/LIB (g/g)	Ni ( %)	Co ( %)	Mn ( %)	Li ( %)
1	220	5.50	65	2.63	70.2	44.5	64	82.8
2	320	8	50	0.75	54.8	34.2	45.8	56.6
3	120	8	80	2	83.4	55.7	76.7	73.8
4	320	3	80	0.75	71.8	52.1	61	84.4
5	220	5.5	65	1.38	75.6	47.5	66.9	70.3
6	120	8	50	0.75	48.4	32.4	39.2	50.8
7	320	3	50	2	69.2	54	60.3	71.2
8	320	8	80	0.75	70.6	51.4	61.1	79
9	220	5.5	65	1.38	74.5	55.7	68	74.1
10	120	3	50	2	65.2	51	61	61.9
11	120	3	80	0.75	66.3	50.2	67.4	76.2
12	220	5.5	65	0.13	44.2	28.7	30.2	72.1
13	320	3	80	2	76.3	62.7	66.5	86.9
14	220	10.5	65	1.38	60.3	43.4	53.6	74.6
15	320	3	50	0.75	71.9	43	66.9	68.9
16	20	5.5	65	1.38	55.4	24.7	16	44.3
17	120	8	50	2	58.4	38.8	57.2	56.6
18	220	5.5	65	1.38	70.6	50.3	67.2	75.4
19	220	5.5	65	1.38	82.2	50.9	57.9	78.2
20	420	5.5	65	1.38	76.4	63.6	71.9	80.3
21	320	8	80	2	91.4	83.7	81.9	83.6
22	120	3	80	2	69.3	48.9	65.7	79.8
23	120	8	80	0.75	66.1	47.1	58.4	71.3
24	320	8	50	2	64.2	47.6	62.5	59.8
25	120	3	50	0.75	60.5	38.7	52	59.8
26	220	5.5	35	1.38	28	24.8	33.2	59.4
27	220	5.5	95	1.38	51.1	66.4	43	88.5
28	220	0.5	65	1.38	50.4	41.4	45.7	56.6

A Box-Cox transformation is an accepted tool that improves for normality of the data and reduces risks for model factor bias³⁸. Box-Cox plots for RSM design without any transformation are illustrated in Fig. 1 for Ni, Co, Mn, and Li. According to Figs. 1a,b,d, which correspond to Ni, Co, and Li data, no transformation is required, and the original data can be used. However, the Box-Cox plot for Mn, a power of 2.13 is recommended for increasing the normality of the recovery data. For simplicity, a power of 2 was used in the transformation.

Fig. 1.

Cox-Box plots for the recovery of (a) Ni, (b) Co, (c) Mn, and (d) Li.

Model determination and analysis of variance (ANOVA)

To model and predict the recovery of each of the Ni, Co, Mn, and Li elements using the data obtained from 28 RSM-designed experiments, a first-order polynomial equation for Co, and a second-order equation for Mn, Ni, and Li were proposed as functions of independent variables. Equations (3) to (6) represent the dissolution efficiencies of Ni, Co, Mn, and Li based on the values of parameter codes A, B, C, and D, corresponding to time, CH₃COOH concentration, temperature, and PP/LIB, respectively. Equations (3) to (6) were modeled using DX 13.0.5.0 software and were iteratively simplified. During the simplification process, parameters with no significant effect and had an F-value less than the standard F were ignored. The goal was to obtain more straightforward mathematical relationships whenever possible. The F-value can be considered a measure of its impact on the recovery level, calculated as the ratio of the variance of each parameter to the residual variance. A higher F-value for a parameter implies a more significant impact^36,39. To convert the actual parameter values to coded form, Eq. (7) can be utilized.

3

4

5

6

7

In Eq. (7), X_i represents the actual values of leaching parameters (A, C, D), X₁ is the actual value at the higher level (+ 1), and X₀ is the actual value of each parameter at the middle point level. The actual values of each parameter at different levels are provided in Table 2. In Eqs. (3) to (6), the larger the coefficient of a parameter, the more significant its impact will be, and the positive or negative sign of the coefficient indicates its enhancing or reducing effect on metal recovery, respectively. According to the models for Ni, Co, Mn, and Li, temperature is one of the most influential parameters affecting the recovery of these elements from LIB cathodes. While acetic acid concentration does not have a significant impact on metal extraction in the range of 1 to 6 M, it was not considered in the analysis of variance (ANOVA). In studies on leaching LIB cathodes in CH₃COOH, the negligible effect of acid concentration within various ranges has been emphasized^24,25. For example, in the research by Liang et al.²⁶, the CH₃COOH concentration was varied in the range of 1 to 2.5 M, and the concentration of CH₃COOH has a minor effect on the leaching rate of Li. The Li leaching recovery can be maintained above 90% at low acetic acid concentrations. However, the dissolution of Ni, Co, and Mn has slightly improved with the increase in CH₃COOH concentration from 0.1 to 2.5 M, indicating a minor influence of this parameter²⁶. The presence of CH₃COOH can act as a proton buffer to provide a steady source of H⁺ for dissolution facilitation. It can coordinate with metal ions to enhance the diffusion from the solid to the aqueous phase.

Tables 4, 5, 6, and 7 present the results of the analysis of variance for Ni, Co, Mn, and Li, respectively. The p-values for all leaching parameters in this study are considerably less than the threshold significance level (P < 0.05 or lower than 5%). The analysis of variance (ANOVA) of experimental data indicates that the parameters of time, temperature, and PP/LIB are statistically significant for Ni, Co and Mn recovery. In the case of Li recovery, time and temperature are the significant parameters. Additionally, the squared term of time is also a significant parameter for Li with a p-value lower than 0.05. In the recovery of Ni, in addition to the first-order parameters of time and PP/LIB, temperature has a second-order effect on extraction (Table 4). The p-value in the last column for Ni recovery is 0.0002, meaning the model error probability is below 0.05.

Table 4.

ANOVA results for optimization of Ni dissolution in the leaching process of LIB cathode.

Source	Sum of squares	df	Mean square	F-value	p-value
Model	2836.90	4	709.23	8.40	0.0002	Significant
A-time	372.88	1	372.88	4.42	0.0468
C-Temprature	922.56	1	922.56	10.93	0.0031
D-PP/LIB	590.04	1	590.04	6.99	0.0145
C2	951.42	1	951.42	11.27	0.0027
Residual	1942.10	23	84.44
Lack of fit	1872.40	20	93.62	4.03	0.1384	Not significant
Pure error	69.71	3	23.24
Cor total	4779.01	27

Table 5.

ANOVA results for optimization of Co dissolution in the leaching process of LIB cathode.

Source	Sum of squares	df	Mean square	F-value	p-value
Model	3099.66	3	1033.22	18.48	< 0.0001	Significant
A-time	860.40	1	860.40	15.39	0.0006
C-Temprature	1589.25	1	1589.25	28.42	< 0.0001
D-PP/LIB	650.00	1	650.00	11.63	0.0023
Residual	1341.87	24	55.91
Lack of Fit	1307.07	21	62.24	5.37	0.0955	Not significant
Pure error	34.80	3	11.60
Cor total	4441.53	27

Table 6.

ANOVA results for optimization of Mn dissolution in the leaching process of LIB cathode.

Source	Sum of squares	df	Mean square	F-value	p-value
Model	2.479 × 107	3	8.263 × 106	5.50	0.0051	Significant
A-time	7.195 × 106	1	7.195 × 106	4.79	0.0386
C-Temperature	6.914 × 106	1	6.914 × 106	4.60	0.0423
D-PP/LIB	1.068 × 107	1	1.068 × 107	7.11	0.0135
Residual	3.606 × 107	24	1.503 × 106
Lack of fit	3.500 × 107	21	1.667 × 106	4.69	0.1140	Not significant
Pure error	1.067 × 106	3	3.556 × 105
Cor total	6.085 × 107	27

Table 7.

ANOVA results for optimization of Li dissolution in the leaching process of LIB cathode.

Source	Sum of squares	df	Mean square	F-value	p-value
Model	2733.10	3	911.03	25.03	< 0.0001	Significant
A-time	728.20	1	728.20	20.01	0.0002
C-Temprature	1795.74	1	1795.74	49.34	< 0.0001
A2	209.16	1	209.16	5.75	0.0247
Residual	873.43	24	36.39
Lack of fit	841.13	21	40.05	3.72	0.1529	Not significant
Pure error	32.30	3	10.77
Cor total	3606.53	27

The ANOVA table demonstrates that the model for predicting Ni recovery is statistically significant (p-value = 0.0002), indicating a high confidence level in the overall predictions. The residual mean square (84.44) represents variability not explained by the model, contributing to prediction error margins. Individual factors like temperature (p = 0.0031) and PP/LIB (p = 0.0145) show strong significance, with confidence levels exceeding 98%. The lack of fit is not significant (p = 0.1384), suggesting that the model fits the data well, minimizing systematic errors. Overall, the low p-values and well-fitted nature of the model indicate reliable predictions with acceptable error margins.

The analysis of variance (ANOVA) results for cobalt recovery demonstrate that the regression model is statistically highly significant, with an extremely low p-value (< 0.0001), indicating a high degree of reliability in the model’s predictive capability. The residual mean square (55.91) quantifies the unexplained variability, contributing to prediction error margins. Key factors such as time (p = 0.0006), temperature (p < 0.0001), and PP/LIB (p = 0.0023) are all highly significant, with confidence levels exceeding 99%. The lack of fit is not significant (p = 0.0955), suggesting the model fits the data well and minimizes systematic errors. Overall, the low p-values and robust model fit indicate reliable predictions with acceptable error margins for Co recovery.

Statistical evaluation of the Mn recovery data through ANOVA revealed significant model adequacy (p-value = 0.0051), suggesting acceptable reliability in the generated predictions. The residual mean square (1.503 × 10⁶) represents the unexplained variability, contributing to prediction error margins. Key factors such as time (p = 0.0386), temperature (p = 0.0423), and PP/LIB (p = 0.0135) are all significant, with confidence levels exceeding 95%. The lack of fit is not significant (p = 0.1140), suggesting the model fits the data well and minimizes systematic errors. Overall, the model’s significance, combined with its well-fitted nature, indicates reliable predictions with acceptable error margins for Mn recovery.

ANOVA results for Li recovery indicate exceptional model significance (p < 0.0001), demonstrating strong predictive validity. The residual mean square (36.39) quantifies the unexplained variability, contributing to prediction error margins. Key factors such as time (p = 0.0002), temperature (p < 0.0001), and the quadratic effect of time (A², p = 0.0247) are all significant, with confidence levels exceeding 97%. The lack of fit is not significant (p = 0.1529), suggesting the model fits the data well and minimizes systematic errors. Overall, the low p-values and robust model fit indicate reliable predictions with acceptable error margins for Li recovery.

Effect of parameters

Effect of time and acetic acid

Figure 2 illustrates the effect of the parameters time and acetic acid concentration on the recovery of Ni, Co, Mn, and Li under leaching conditions with a temperature of 80 °C, and PP/LIB of 2.0 g/g. As the results indicate, the recovery of Ni increased from 71.6 to 79.4% with an increase in time from 120 to 320 min at 3 M CH₃COOH concentration (Fig. 2a). The relatively modest 7.8% increase in Ni recovery at 200 min suggests that it is readily soluble under leaching conditions with an acetic acid concentration of 3.0 M, a temperature of 80 °C, and PP/LIB of 2.0 g/g within the first 120 min. At higher concentrations of CH₃COOH, similar conditions for Ni recovery are observed. It can also be concluded that the concentration of acetic acid in the range of 3 to 8 M does not significantly affect Ni recovery, confirming the results of the ANOVA table (see Table 4). From Fig. 2b, Co recovery reached 55.0%, and with an increase in time to 320 min, this value increased to 66.9% (conditions: 5.5 M CH₃COOH, 80 °C and PP/LIB of 2.0 g/g). With the increase in CH₃COOH concentration from 3 to 8 M, Co recovery has only increased by only 0.6%, indicating the lack of influence of this parameter in this concentration range on Co dissolution from the LIB cathode. A similar effect of CH₃COOH concentration on the recovery of Mn and Li is observed in Fig. 2c,d. While the Mn recovery increased from 63.6 to 71.8%, with an increase in time from 120 to 320 min at 5.5 M CH₃COOH concentration. Regarding Li recovery, the time varied between 120 and 320 min, resulting in an increase in recovery from 73.4 to 83.4%. Similar conditions regarding the effect of time on the recovery of metals present in the LIB cathode have been reported by Liang et al.²⁶ and Yan et al.²⁵. It has been reported that there is a direct relationship between increasing time and enhanced metal recovery in the recycling of LIB cathodes. Therefore, the increase could be attributed to enhanced collisions between LIB cathode particles and acid molecules due to the extended leaching time²⁷.

Fig. 2.

The impact of time and acetic acid concentration on the recovery of Ni, Co, Mn, and Li under leaching conditions at a temperature of 80 °C and PP/LIB of 2.0 g/g (software DX13).

Acetic acid concentration has a minor effect on LIB cathode leaching primarily because it is a weak acid with limited proton (H⁺) availability, making it less effective for dissolving transition metals (Ni, Co, Mn) that require both acidic and reductive conditions for efficient extraction. While higher acetic acid concentrations show negligible impact on transition metals without additional reductants like H₂O₂ or ascorbic acid. In general, organic acids, like acetic acid, exhibit several weaknesses compared to inorganic acids in leaching processes primarily due to their lower acidity (higher pKa values), slower reaction kinetics, and dependency on oxidants for effective metal recovery⁴⁰. Despite their lower acidity, some organic acids like methanesulfonic and citric acids have shown comparable performance to mineral acids in dissolving metals from mine tailings⁴¹. Acetic acid is classified as a weak organic acid due to its limited dissociation into ions when dissolved in water. Unlike strong acids, it doesn’t fully release protons (H⁺) in solution. Among organic acids, it occupies a middle ground in terms of strength, with some acids being weaker and others stronger. This behavior stems from its carboxyl group (-COOH), a defining feature of carboxylic acids, which generally exhibit weak acidic properties. It appears that these factors contribute to acetic acid having only a minor impact on the transition metals recovery during the leaching of lithium-ion battery (LIB) cathodes. Its weak acidic nature and limited dissociation likely reduce its effectiveness in extracting transition metals compared to stronger acids.

Effect of temperature and PP/LIB

Figure 3 illustrates the effect of temperature and PP/LIB on the recovery of Ni, Co, Mn, and Li under leaching conditions with a time of 320 min and CH₃COOH of 5.5 M. According to the obtained results, the recovery of Ni, Co, Mn, and Li show 12.2% (from 57.4 to 69.6%), 16.1% (from 40.4 to 56.5%), 9.7% (from 51.9 to 61.6%), and 17.1% (from 66.2 to 83.366.2%) increase within the temperature range of 50 to 80 °C at PP/LIB = 0.75 g/g. With an increase in the PP/LIB to 2.0, the recovery of Ni, Co, and Mn experience increases of 12.4%, 16.1%, and 8.0%, respectively, within the same temperature range (50 to 80 °C). While Li recovery in the temperature range of 50 to 80 °C and a constant PP/LIB of 2.0 increases from 66.2 to 83.3%, which exactly matches the value at the temperature range and a PP/LIB ratio of 0.75, indicating the lack of influence of the PP/LIB factor on its dissolution. This was also demonstrated in the Li recovery equation (Eq. 6). Considering the absence of reductive leaching for Li, it was entirely expected that the reducing agent PP would also have no effect on its dissolution. Given that the LIB cathode leaching is an endothermic process, an increase in temperature will lead to higher efficiency according to Le Chatelier’s principle, as emphasized in previous studies^27,42. Increasing the temperature will increase the energy of the molecules, and their collisions with sufficient energy can drive the reaction forward and overcome the activation energy^26,27. Since fruit peels contain high amounts of cellulose and hemicellulose, they can be converted into reducing sugars (e.g., glucose) at high temperatures⁴³. In other words, at high temperatures, long hydrocarbon chains are broken down into shorter hydrocarbon structures like sugars. Additionally, it has been reported that fruit peels may contain antioxidant compounds such as flavonoids and phenolic acids, which can participate in reductive leaching of the LIB cathode and enhance metal recovery. Therefore, the discussion on the formation of high glucose concentration at high temperatures will be very important, as it can have a significant impact on reductive leaching according to the following reaction:

8

Fig. 3.

The impact of temperature and PP/LIB on the recovery of Ni, Co, Mn, and Li under leaching conditions with CH₃COOH of 5.5 M and time of 320 min (software DX13).

Certainly, another reaction can be considered for lithium, in which part of the lithium can dissolve in acetic acid without the need for a reducing agent, as shown in the following reaction:

9

It has been reported that during the LIB cathode leaching in an acetic acid environment with H₂O₂ as a reducing agent, temperature has been recognized as the key determinant²⁴. Similarly, in another study, it has been reported that an increase in temperature resulted in an improvement in the leaching recovery for Ni, Co, Mn, and Li²⁶. In that study, the enhancement in recovery was attributed to the increase in pressure due to temperature rise, creating harsh conditions such as in viscosity, surface tension, and a lower dielectric constant, along with an increase in diffusivity. According to the research of Natrajan et al.²⁷, an increase in temperature up to 70 °C led to an improvement in the recovery of Li, Co, and Mn.

In the case of PP/LIB effect on metal recovery, Ni, Co, and Mn experienced 12.5, 10.3, and 10.9% increase, respectively, with the rise in the ratio from 0.75 to 2 g/g at a temperature of 65 °C, CH₃COOH concentration of 5.5 M and time of 320 min. The reason is the reductive leaching of the LIB cathode, where Co(III) and Mn(IV) are reduced to Co(II) and Mn(II), respectively, to become soluble^24,44. As mentioned previously, there is no meaningful effect of PP/LIB on Li dissolution, indicating that the reducing agent has minimal effect on the leaching of Li. In most studies conducted on reductive leaching of the LIB cathodes in a CH₃COOH environment, the impact of the reducing agent on Li recovery has been relatively minor^26,43, which finding is also confirmed in the ANOVA related to lithium (Table 7). In general, the dissolution reaction of the LIB cathode in the presence of the PP can be considered as a reaction (8). It should be noted that the hydrolysis of PP leads to the formation of glucose-containing derivatives, which can accelerate the dissolution process^45,46.

In the study by Liang et al.²⁶, various reducing agents were used to dissolve the LIB cathode in acetic acid, with two significant substances being glucose and cellulose. According to the results of this study, adding glucose resulted in a recovery of almost 95% for all elements, including Ni, Co, Mn, and Li. Conversely, by adding cellulose under similar conditions, less than 25% dissolution was achieved for Ni, Co, and Mn, with only Li dissolving at around 70%. They mentioned that the reason for the most significant effect of glucose on the dissolution of LIB cathode is the presence of the aldehyde functional group in the structure of glucose. This functional group facilitates the reduction of Co and Mn ions, resulting in enhanced dissolution²⁶.

Therefore, the presence of glucose compounds during dissolution is crucial and can significantly contribute to dissolution. Nevertheless, in the current research, high dissolution of Ni, Co, Mn, and Li was achieved with the assistance of compounds, mainly cellulose, which are cost-effective compared to other reducing agents. This achievement is considered one of the novelties of this research group. Additionally, in the dissolution reaction of LIB cathode, glucose undergoes oxidation, giving rise to compounds such as polyhydroxy acids, aldonic acid, tartaric acid, oxalic acid, and formic acid, and ultimately decomposes into CO₂ and H₂O^26,47. According to the results of Yan et al.²⁵, the components present in bagasse pith include cellulose, hemicellulose, and lignin. This composition is similar to the pomegranate peel used in this study. Among these substances, cellulose is the only one that decomposes into glucose, further breaking down into compounds such as 2-glucaric acid, tartaric acid, glyoxylic acid, tartronic acid, and formic acid, which can accelerate the LIB cathode dissolution in acetic acid media.

In addition, the reductant effect of banana bagasse powder significantly impacts the recovery of Li, Ni, Co, and Mn from lithium-ion battery cathode materials in acetic acid media⁴⁸. At 80 °C, metal recovery efficiencies reach 50–70% for all Li, Ni, Co, and Mn after 4 h, while at 100 °C, recoveries exceed 90% after 8 h. Cobalt shows a more gradual increase compared to the other metals due to the reduction of Co(III) to more soluble Co(II). The mechanism involving cellulose, hemicellulose, and lignin in the banana bagasse creates a reducing environment. This environment facilitates the reduction of Co(III) to Co(II) and Mn(IV) to Mn(II), with reducing sugars like glucose, produced by hydrolysis of banana bagasse components, acting as the primary reductants. The obtained results are in good agreement with our experimental findings. Furthermore, the study demonstrated that tangerine peel (TP) effectively enhanced the recovery of Ni, Co, Mn, and Li from spent LIB cathodes through reductive leaching, with optimal performance at 100 °C, 3 M acetic acid, and a TP/NCM ratio of 2 g/g²⁸. While Li recovery was less dependent on TP concentration, Ni, Co, and Mn showed strong reductant dependence, with TP’s effectiveness attributed to its hydrolysis into glucose, which was entirely consumed during leaching.

The Optimum condition

In the optimization process, the lowest and highest levels for each parameter should be determined. Multiple responses from various combinations of four parameters, including time, CH₃COOH concentration, temperature, and PP/LIB, were utilized to achieve the highest recovery of Ni, Co, Mn, and Li during the LIB cathode leaching process. Figure 4 illustrates the optimized conditions and the corresponding element recoveries. Under the optimal conditions of 320 min leaching time, 5.5 M CH₃COOH concentration, 92 °C temperature, and a PP/LIB of 3.5 g/g, the predicted recoveries for Ni, Co, Mn, and Li were 83.3, 85.9, 84.9, and 91.2%, respectively (Fig. 4). Meanwhile, the experimentally obtained recovery values for Ni, Co, Mn, and Li under optimal conditions were 84.4, 78.3, 86.2, and 93.0%, respectively, which exhibit excellent agreement with the calculated results according to the RSM within a 10% error range, as designed in the experimental procedure.

Fig. 4.

Prediction of optimal conditions to dissolve LIB cathode using RSM design.

Characterization of leaching residue

The XRD pattern of the LIB (raw material), the leaching residue of LIB cathode with CH₃COOH in the presence and absence of PP under optimal leaching conditions (320 min leaching time, 5.5 M CH₃COOH, 92 °C temperature, and PP/LIB of 3.5 g/g) is shown in Fig. 5. The compounds, including Co₃O₄, Li_0.9Co_0.5Ni_0.5O_2−x, LiMn₂O₄, and Li_0.69Ni_1.01O₂, were the main phases in LIB material (Fig. 5a). In addition, the characteristic peak of the aluminum was not observed, indicating the complete dissolution of aluminum. The phases present in the leaching residue without PP include Co₃O₄, Li_0.9Co_0.5Ni_0.5O_2−x, LiMn₂O₄, and Li_0.69Ni_1.01O₂, showing no significant change compared to the initial LIB cathode (Fig. 5b). This is confirmed by the low recovery obtained for Ni, Co, and Mn, which are 20.1, 3.8, and 9.8%, respectively (see Fig. S1). The only element that exhibited reasonable recovery without a reducing agent is Li, at approximately 65.4% (see Fig. S1). The only peaks that have changed to around 20 and 45° compared to the initial material are likely due to the dissolution of lithium in the phases of Li_0.9Co_0.5Ni_0.5O_2−x, LiMn₂O₄, and Li_0.69Ni_1.01O₂.

Fig. 5.

XRD patterns of (a) LIB cathode, (b) residue of LIB cathode leaching in CH₃COOH medium without PP, and (c) residue of LIB cathode leaching in CH₃COOH with PP.

Adding the reducing agent PP to the leaching solution significantly altered the conditions and the characteristic peaks of Li_0.9Co_0.5Ni_0.5O_2−x, LiMn₂O₄ and Li_0.69Ni_1.01O₂ phases entirely disappeared from the leaching residue. The only remaining phase in the residue is Co₃O₄, which remains in the leaching residue even in the presence of PP. Given the higher recoveries for Ni, Mn, and Li under optimum conditions compared with Co. This result supports the notion that a portion of cobalt present in the Co₃O₄ phase does not fully dissolve in the presence of the reducing agent PP. The Co recovery under optimal conditions by adding a PP reducing agent reached approximately 78.3%.

The leaching of the LIB cathode materials (e.g., Li_0.9Co_0.5Ni_0.5O_2−x, LiMn₂O₄, and Li_0.69Ni_1.01O₂) in acetic acid media is significantly enhanced by the addition of PP as a natural reducing agent. PP contains compounds which act as electron donors, facilitating the reduction of high-valence transition metals (Co³⁺, Mn⁴⁺) to their soluble divalent states (Co²⁺, Mn²⁺). This reductive dissolution process is crucial because acetic acid alone cannot efficiently leach these metals in their higher oxidation states. While reducing metal ions (e.g., Co³⁺ + e⁻ → Co²⁺). As a result, the characteristic XRD peaks of Li_0.9Co_0.5Ni_0.5O_2−x, LiMn₂O₄, and Li_0.69Ni_1.01O₂ phases disappear from the leaching residue, confirming their complete dissolution. However, Co₃O₄ may remain in the residue due to its mixed Co²⁺/Co³⁺ valence state, which exhibits higher stability under mild reducing conditions. This eco-friendly leaching approach aligns with green chemistry principles by utilizing biodegradable reductants instead of traditional hazardous reagents like H₂O₂ or Na₂S₂O₅^49,50.

Figure 6 illustrates the SEM images of LIB cathode powder (a₁), the leaching residue of LIB cathode in CH₃COOH in the absence of a reducing agent (b₁), and the leaching residue of LIB cathode under optimal conditions in the presence of PP (c₁). Additionally, EDX elemental analysis of LIB cathode powder (a₂), the leaching residue of LIB cathode in CH₃COOH (b₂), and the leaching residue of LIB cathode under optimal conditions in the presence of PP (c₂) are shown in Fig. 6. The particles of the LIB cathode (Fig. 6a₁) consist of fine particles agglomerated together. On the other hand, the particles of the residue after 320 min in the absence of a reducing agent (Fig. 6b₁) exhibit separated particles, showing no significant change in particle size. This is consistent with the leaching results, indicating minimal changes in the recoveries for Co, Ni, and Mn during the leaching process. However, in the leaching residue with the presence of the reducing agent (PP), the particles become much finer. The morphology also changes from a regular spherical shape to an asymmetric polygonal shape under the leaching process.

Fig. 6.

SEM images and EDX analysis (a₁ and a₂) LIB cathode, (b₁ and b₂) leaching residue of LIB cathode in CH₃COOH environment, (c₁ and c₂) Leaching residue of LIB cathode in CH₃COOH environment in the presence of PP.

EDX analysis of the samples confirms the presence of Co, Ni, Mn, C, and O in all samples. Notably, significant variation in the C content is essential. In the LIB cathode (Fig. 6a₂), the carbon content is around 3.7%, resulting from its introduction from the anode during the preparation process. The increase in carbon content in the leaching residue without PP can be attributed to the dissolution of certain compounds and the reduction in the residue’s weight. However, a significant difference in the carbon content of the leaching residue becomes evident by the introduction of the reducing agent into the leaching solution. This can be attributed to the existence of the hydrocarbon structure of PP compounds. Furthermore, EDS results in the residue without PP show reducing the weight% of Ni (from 6.7 to 4.2%) and Mn (from 22.4 to 19.3%) and an increase in Co content (from 29.8 to 34.5%) when compared with the LIB cathode. This result is consistent with leaching and XRD results, indicating that a fraction of Ni and Mn dissolve in the leaching process without PP, while the very low cobalt recovery leads to an increased Co content in the final residue. Additionally, the results obtained from EDS for the Ni and Mn content (Fig. 6c₂) in the PP-containing residue validate significant recoveries, with Ni reaching 85.4% and Mn reaching 86.2%. Despite Co exhibiting a higher content than Ni and Mn in the PP-containing residue, as confirmed by a recovery of approximately 65.3% under optimal conditions, it does not undergo complete dissolution. Consequently, a 5.1% cobalt content in this residue appears entirely justified.

The FTIR profile of the leaching liquor obtained from the dissolution of the LIB cathode in the presence and absence of PP in CH₃COOH is depicted in Fig. 7. A broad peak around 3116 cm^− 1 is observed, corresponding to hydroxyl (OH) groups⁵¹. Carboxylic acids exhibit a strong and broad band for O–H in the region of 2500–3300 cm^− 1, centered around 3100 cm⁻¹. The wide distribution of the stretching band for O–H in carboxylic acids is due to the presence of hydrogen-bonded dimers⁵². In the same region, stretching bands of C–H from both alkyl and aromatic groups are identified around 2934 cm^− 153. The carbonyl stretching C=O of a carboxylic acid appears as an intense band in the range of 1690–1760 cm^− 154. The precise position of this broad band depends on whether the carboxylic acid is saturated or unsaturated, dimerized, or has internal hydrogen bonding. A peak at around 1744 cm^− 1 has been identified for an acetic acid solution in the presence of metal ions⁵⁵. Bonds in the range of 1519–1638 cm^− 1 indicate vibrations of C=C. In a concentrated solution of acetic acid containing Ni, Co, Mn, and Li (Fig. 7b), a strong peak in the region of 1563 cm^− 1 suggests the probable formation of metal acetates⁵⁵.

Fig. 7.

FTIR analysis of the leaching solution under optimal conditions (a) in the absence of reducing agent (b) in the presence of PP.

The peak at 1392 cm^− 1 in the solution without PP and 1413 cm^− 1 in the PP-containing solution likely correspond to the bending vibration of the OH group, and the shift in peak position likely occurs due to the presence of metal ions. Peaks at approximately 1272 cm^− 1 and 1015 cm^− 1 are attributed to CH₃ and C––C bonds, respectively. Peaks generated in the range of 550–900 cm^− 1 are related to out-of-plane bending vibrations of C–H bonds. However, the characteristic peak at 660 cm^− 1 in the solution containing PP is likely associated with out-of-plane bending vibrations of cellulose compounds^56,57.

Figure 8 shows the UV-vis spectroscopy of the leaching solution under optimal conditions (320 min leaching time, 5.5 M CH₃COOH, 92 °C temperature, and PP/LIB of 3.5 g/g). In an acetic acid solution containing Co(II), Ni(II), and Mn(II), the UV–Vis spectrum exhibits peaks at 338 nm, 366 nm, and 400 nm, which correspond to electronic transitions influenced by the coordination environment of these metal ions. Co(II) and Ni(II) are the primary contributors to these peaks due to their characteristic d-d transitions and ligand-to-metal charge transfer (LMCT) bands. Co(II) typically forms acetate complexes that enhance LMCT transitions, contributing to the peak at 338 nm. Ni(II), which commonly adopts octahedral or square planar geometry, exhibits strong d-d transitions responsible for the peaks at 366 nm and 400 nm. Mn(II), with its d⁵ configuration, has weak spin-forbidden d-d transitions and contributes minimally to these peaks; however, it may influence the rising absorbance below 300 nm through LMCT interactions with acetate ligands. The spectrum reflects a combination of ligand-field effects and charge-transfer bands in the UV–Vis range. Acetic acid acts as both a solvent and ligand, coordinating with the metal ions to form complexes that alter their electronic structure and absorption properties. Mn(II)’s weak absorptions are overshadowed by the stronger transitions of Co(II) and Ni(II). The observed peaks highlight the importance of coordination chemistry in determining UV-Vis spectral features, with Co(II) and Ni(II) dominating due to their intense ligand-field transitions in this acidic medium.

Fig. 8.

UV–Vis spectroscopy of the leaching solution under optimal conditions (320 min leaching time, 5.5 M CH₃COOH, 92 °C temperature, and PP/LIB of 3.5 g/g).

Kinetic evaluation

The leaching process of the cathode in LIB through a reductive mechanism can be investigated using kinetic models. Given that the dissolution of metals from LIB cathodes is a heterogeneous solid-liquid process, the kinetics of the metal leaching process can be described by a shrinking core model^26,49,58. Among various kinetic models, the shrinking core model is widely applicable in leaching processes^49,58,59. In Figs. S2a-d, kinetic data at four temperatures (50, 65, 80, and 95 °C) are plotted with chemical reaction mechanisms, diffusion in the product layer, and film diffusion. The parameters of CH₃COOH concentration and PP/LIB are considered as 5.5 M and 3.5 g/g, respectively, based on the results of optimal conditions. A kinetic model should be chosen when the correlation coefficient (R²) is near one, and the model should also be practically justifiable. For example, when a product layer does not form during leaching, choosing the diffusion model through the product layer, despite a higher R², is not justified. Therefore, both the condition of having an R² close to one and practical justifiability should be considered when selecting a model.

Considering that the diffusion through the liquid film model is valid under conditions where the liquid is very dense, and the movement of ions in it is difficult, if the diffusion through the liquid film is the controlling step, increasing the stirring speed significantly affects the dissolution rate⁶⁰. However, an increase in stirring speed has a minimal effect on the dissolution rate during the initial experiments. Therefore, for each of the four elements, Ni, Co, Mn, and Li, this mechanism cannot be considered the kinetic controlling step under the mentioned conditions. Conversely, based on the dissolution reactions of the elements within the LIB cathode, a solid product layer does not develop during the leaching process. Consequently, using the diffusion mechanism through the product layer as the primary model for elucidating the dissolution mechanism is not applicable. The only mechanism that can be justified under practical conditions, considering the dissolution reactions, is the chemical reaction model. Upon close examination of the R² values in Figs. S2a (for Ni), c (for Mn), and d (for Li), it becomes evident that there is a lack of consistency, and some values are far from 1.0. Specifically, in the case of Ni, the correlation coefficients for dissolving nickel in a CH₃COOH environment varied from 0.76 to 0.97. Consequently, the shrinking core model is not suitable for explaining the leaching mechanism of Ni, Mn, and Li. Nevertheless, examining the results of Co dissolution reveals that the R² for the chemical reaction model exceeds 0.96 (Fig. S2b). Therefore, this model can be considered for the initial investigation of the Co dissolution mechanism. However, the Avrami model offers a different perspective by focusing on nucleation and growth processes, making it suitable for systems where leaching involves complex mechanisms such as surface irregularities, heterogeneous reaction sites, or non-uniform particle dissolution. Unlike SCM, which assumes uniform particle behavior, the Avrami model accounts for time-dependent changes in reaction kinetics and can describe leaching systems with overlapping or mixed controls. This flexibility makes it applicable to non-ideal systems such as waste printed circuit boards or materials with irregular morphologies. For instance, in copper recovery studies, the Avrami model provided better fitting for leaching kinetics due to its ability to capture non-linear behavior over time⁶¹. Therefore, the Avrami model is preferred when leaching kinetics deviate from these assumptions due to complex particle-scale phenomena. However, for simplicity and to cover all four elements, another model was also examined.

Therefore, the Avrami model, one of the most common models in LIB cathode leaching, was investigated for all elements.

The Avrami model has been utilized by many researchers to describe the kinetics of transformations in crystallization and solid phase^62,63. Initially, this model was developed to describe the crystallization process. However, since the leaching reaction can be considered the reverse of the crystallization process, it can also be employed to determine the kinetics of the leaching reaction of crystals⁶⁴. The Avrami relationship is defined as follows:

10

where y is the leaching recovery, K is the reaction rate constant (min^− 1), t is the reaction time (min), and n is the Avrami index.

The Avrami model was applied to fit the data for four elements, Ni, Co, Mn, and Li, at four temperatures: 50, 65, 80, and 95 °C. The curve of ln[-ln(1-y)] versus ln(t) was fitted. Figure 9 illustrates the results of fitting the Avrami model for the leaching of the LIB cathode in CH₃COOH. The correlation coefficients for each temperature are separately presented in these plots. From the corresponding results, it can be observed that the kinetic data for Ni, Co, Mn, and Li fit well with the Avrami model, with R² values exceeding 0.94, 0.95, 0.92, and 0.83, respectively. These findings indicate that the kinetic behavior of the LIB cathode dissolution in CH₃COOH can be predominantly controlled by a mechanism analogous to reverse crystallization. This model has been previously utilized in studies of LIB dissolution in CH₃COOH^25,65.

Fig. 9.

Kinetic data plot of LIB leaching based on the Avrami model at four temperatures of 50, 65, 80, and 95 °C for (a) Ni, (b) Co, (c) Mn, and (d) Li.

Activation energy (∆G_A) is the minimum amount of energy required for a reaction to occur. Activation energy is calculated based on the Arrhenius equation, which is formulated as follows and depicted in Fig. 10.

11

Fig. 10.

Energy activation for (a) Ni, (b) Co, (c) Mn, and (d) Li based on the Avrami model.

In this equation, A is the frequency factor in min^− 1, ∆G_A is the activation energy in J.mol^− 1, R is the gas constant in J·K^− 1·mol^− 1, and T is the leaching solution temperature in Kelvin. If Ln(K) is plotted against 1/T, the activation energy can be calculated using the slope of the line (−∆G_A/R). The activation energies for Ni, Co, Mn, and Li were determined to be 86.35, 41.32, 67.33, and 10.38 kJ/mol, respectively. In the studies by Yan et al.²⁵, the activation energies for Li, Ni, Co, and Mn were found to be 40.6, 42.2, 42.8, and 43.8 kJ/mol, respectively. The activation energy indicates that the surface chemical reactions of Li, Ni, Co, and Mn control the reaction rates during leaching^66,67.

The obtained activation energy values provide crucial insights into the rate-determining mechanisms governing metal dissolution in acetic acid media with pomegranate peel (PP) as a reductant. For nickel (86.35 kJ/mol) and manganese (67.33 kJ/mol), the relatively high activation energies clearly indicate that their dissolution is chemically controlled, with the electron transfer and bond-breaking reactions representing the rate-limiting steps. In nickel’s case, this suggests the breakdown of stable Ni–O bonds in the crystalline lattice requires substantial energy input, while for manganese, the energy barrier likely reflects the redox shuffling between Mn(II) and Mn(III) oxidation states during dissolution. Cobalt’s intermediate activation energy (41.32 kJ/mol) reveals a more complex, mixed-control mechanism. This value falls within the transitional range (20–40 kJ/mol) where both chemical reaction and diffusion processes contribute to rate limitation^36,68. The data suggest most cobalt-containing phases dissolve through chemically controlled reduction, while the dissolution of cobalt oxide (Co₃O₄) phases appears to involve both chemical reactions at the surface and subsequent diffusion through the oxide layers⁶⁹. This dual behavior explains why cobalt’s activation energy is significantly lower than nickel’s but still higher than diffusion-controlled processes, and corroborates with the observed slower dissolution kinetics of the Co₃O₄ structure compared to other cobalt phases. Lithium’s remarkably low activation energy (10.38 kJ/mol) presents a stark contrast, clearly indicating a diffusion-controlled dissolution mechanism. In this case, the rate-limiting factor is simply the transport of protons through lithium oxide layers rather than any chemical reaction barrier. The absence of redox requirements for lithium (which remains as Li⁺ throughout the process) and the weak bonding in its oxide matrix allow for this exceptionally low energy pathway. This fundamental difference in dissolution mechanism explains lithium’s characteristically rapid leaching behavior compared to the transition metals in the system.

Economic and technical considerations

The industrial feasibility of using pomegranate peel (PP) as a green reductant for lithium-ion battery recycling hinges on a nuanced cost-benefit analysis that weighs its environmental advantages against technical and economic limitations. While PP offers compelling sustainability benefits, being a low-cost (potentially waste-derived) material that avoids toxic emissions (e.g., SO₂, Cl₂) and hazardous waste streams associated with conventional reductants like H₂O₂, its practical implementation faces several challenges: (1) Leaching kinetics are slower for transition metals (Ni, Co, and Mn) due to higher activation energies, potentially increasing operational costs through extended processing times or elevated temperature requirements; (2) The seasonal and geographically variable supply of PP requires robust collection and pretreatment (e.g., drying, grinding) infrastructure, unlike synthetic reductants with stable supply chains; and (3) While PP’s biodegradability reduces waste treatment costs, its lower leaching efficiency for critical metals like cobalt could offset gains from reduced chemical expenses. A hybrid approach, using PP for initial leaching followed by minimal acid refining, may optimize costs, but a full techno-economic analysis must account for site-specific factors such as local PP availability, energy prices, and regulatory incentives for green chemistry. Further research should quantify the trade-offs between PP’s potentially higher process costs and its potential for carbon credits or reduced environmental compliance burdens to determine its viability at scale.

Conclusion

1. RSM was employed to optimize the leaching parameters of LIB cathodes in the presence of PP. Time (20 to 420 min), CH₃COOH concentration (0.5 to 10.5 M), temperature (35 to 95 °C), and PP/LIB (0.13 to 2.63 g/g) were considered. According to ANOVA, temperature was identified as the most significant parameter affecting the recovery of Ni, Co, Mn and Li, while the CH₃COOH concentration had the most negligible impact on element recovery.

2. According to the results of the study, it was observed that increasing the observation time has a positive effect on the recovery of Ni, Co, Mn, and Li. Under conditions where the temperature is 80 °C, the acid concentration is 5.5 M, and the PP/LIB ratio is 2.0, with an increase in time from 120 to 320 min, an increase of approximately 8% for Ni, 12% for Co, 8% for Mn, and 10% for Li was observed.

3. The impact of temperature on the recovery of Ni, Co, Mn, and Li under leaching conditions of 5.5 M CH₃COOH concentration, 320 min, and PP/LIB of 2.0 g/g, indicated that raising the temperature from 50 to 80 °C showed 12% for Ni, 16% Co, 10% for Mn and 17% for Li.

4. The positive effect of increasing PP/LIB on the recovery of Ni, Co, and Mn has been significant, leading to an approximately 11% increase in all elements. However, it had no significant effect on Li recovery.

5. The leaching recovery under optimal conditions (320 min leaching time, 5.5 M CH₃COOH concentration, 92 °C temperature, and PP/LIB of 3.5 g/g) was predicted to be 83.3% for Ni, 85.9% for Co, 84.9% for Mn, and 91.2% for Li, respectively. The actual recovery values under these optimal conditions were 84.4, 78.3, 86.2, and 93.0%, showing excellent agreement with the calculated results based on the response surface methodology within a 10% error range.

6. XRD patterns of LIB cathodes indicated the presence of Co₃O₄, Li_0.9Co_0.5Ni_0.5O_2−x, LiMn₂O₄ and Li_0.69Ni_1.01O₂. XRD patterns of leaching residues of LIB cathodes in CH₃COOH medium without PP showed minimal changes in characteristic XRD peaks when compared with the LIB cathode, and only lithium-containing phases underwent alterations, indicating high recovery of this metal in conditions without PP.

7. The XRD pattern of leaching residues demonstrated that adding the reducing agent PP resulted in the almost complete removal of characteristic peaks of the Li_0.09Ni_1.01O₂ and Li₂MnO₃ phases from the leached structure. However, Co₃O₄ phase peaks were still observable, indicating the challenging dissolution of this phase.

8. FTIR characterization of leaching samples in the presence and absence of PP revealed almost similar peaks related to acetic acid bonds, with the shift in peaks due to the presence of Ni, Co, Mn, and Li elements. The absence of peaks related to glucose indicates that glucose molecules produced during dissolution are immediately consumed by LIB cathode material.

9. Kinetic analysis of LIB leaching in the presence of PP indicated that the dissolution mechanism of Ni, Co, Mn, and Li elements follows the Avrami model. The activation energies for Ni, Co, Mn, and Li were determined to be 86.35, 41.32, 67.33, and 10.38 kJ/mol, respectively.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

S.K. wrote and reviewed the main manuscript text and prepared figures and tables. M.K. did the experiments. P.A. reviewed the manuscript.

Data availability

All data generated or analysed during this study are included in this published article and its supplementary information files.

Declarations

Competing interests

The authors declare no competing interests.