From Waste to Cathode: A Comparative Evaluation of Sol–Gel and Co-Precipitation Routes for Closed-Loop Recycling of Lithium-Ion Battery Cathodes

Abstract

The exponential growth of lithium-ion batteries (LIBs) in electric vehicles and energy storage systems has amplified the urgent need for sustainable recycling strategies. Conventional pyrometallurgical and hydrometallurgical methods for LIB recycling are energy-intensive, chemically demanding, and fail to preserve the structural integrity of cath-ode materials. Closed-loop recycling, in contrast, enables the recovery of layered oxides with minimal processing steps, reducing environmental footprint and supporting a circular economy. This study provides a systematic comparison of two regeneration approaches—sol–gel synthesis and hydroxide co-precipitation—for closed-loop recycling of layered NCM (LiNi_xCo_yMn_zO₂) cathode materials recovered from spent LIBs. Spent cells were mechani-cally processed and leached using malic acid to recover Ni, Co, Mn, which were subsequently used to synthesize NCM622 cathode powders. The regenerated materials were characterized using SEM/EDX, XRD, and electrochemical testing in CR2032 coin cells. Both methods successfully produced phase-pure layered oxides with the R-3m structure, with distinct differences in structural ordering and electrochemical behavior. The sol–gel-derived NCM622 displayed higher crystallinity and reduced cation mixing, evidenced by an I(003)/I(104) ratio of 1.896 compared to 1.720 for the co-precipitated sample, and delivered a high initial discharge capacity of 170 mAh/g at 0.1 C. However, it exhibited significant capacity fade, retaining only 60 mAh/g after 40 cycles. In contrast, the co-precipitation route produced hierarchical porous spherical agglomerates that offered superior cycling stability, maintaining ~150 mAh/g after 40 cycles with lower polarization (ΔE_p = 0.16 V). Both materials demonstrated electrochemical performance comparable to commercial NCM. Overall, hydroxide co-precipitation emerged as the most industrially viable method due to scalable processing, compositional robustness, and improved long-term stability of regenerated cathodes. This work highlights the critical influence of synthesis route selection in LIB closed-loop recycling and provides a technological framework for industrial recovery of high-value NCM cathode materials.

1. Introduction

The rapid expansion of the global market for lithium-ion batteries (LIBs) is driven by their widespread use in portable electronics, electric vehicles (EVs), and stationary energy storage systems. Owing to their high energy density, long cycle life, and relatively low self-discharge, LIBs have become the dominant rechargeable power source of the 21st century [1]. Global LIB demand is projected to increase several-fold in the next decade, largely fueled by the accelerating transition to e-mobility and the deployment of renewable energy integration technologies [2].

However, the growing volume of spent LIBs poses critical challenges. At the end of their service life—typically after 1000–3000 charge–discharge cycles—LIBs lose their capacity to efficiently power devices and vehicles. It is estimated that millions of tons of spent batteries will reach disposal annually by 2030 [3]. Uncontrolled landfilling or incineration of such waste results in severe environmental and health hazards due to the leaching of toxic transition metals, electrolytes, and organic solvents into soil and groundwater [4]. At the same time, spent LIBs contain valuable and strategically important metals such as cobalt, nickel, manganese, and lithium, which are finite and geographically concentrated resources. Recovering these metals is therefore essential both from an environmental and economic perspective [5].

Current industrial recycling technologies for LIBs are dominated by pyrometallurgical and hydrometallurgical processes [6]. Pyrometallurgy involves high-temperature smelting of battery components, leading to recovery of cobalt, nickel, and copper, whereas lithium is typically lost in the slag phase [7]. Although pyrometallurgical routes are relatively mature and straightforward to implement, they suffer from severe drawbacks: high energy consumption, substantial CO₂ emissions, loss of lithium, and secondary pollution requiring additional waste treatment [8].

Hydrometallurgy, on the other hand, relies on leaching processes using mineral acids, followed by solvent extraction, precipitation, or ion-exchange to separate and purify individual metals. This approach enables recovery of lithium alongside cobalt, nickel, and manganese, but requires complex multi-step separation, extensive use of chemical reagents, and generates large volumes of wastewater. Both pyrometallurgical and hydrometallurgical methods treat spent LIBs primarily as sources of individual metals, which are then used to synthesize new cathode materials from scratch [9]. While effective, this approach disrupts the original layered structure of cathode powders and often leads to resource inefficiency [10].

To overcome these limitations, closed-loop recycling of cathode materials has emerged as a promising strategy [11,12]. Instead of decomposing cathode powders into constituent metals, closed-loop recycling restores or resynthesizes the functional cathode material from the spent material itself. In particular, layered cathode oxides such as LiNixCoyMnzO₂ (NCM) and LiCoO₂ (LCO) can be regenerated via controlled leaching, re-lithiation, and thermal treatment [13]. This approach offers several key advantages:

• Fewer processing steps compared to hydrometallurgy.
• Reduced chemical consumption and wastewater generation.
• Preservation of layered crystal structures.
• Higher economic value, since regenerated NCM powders are directly suitable for battery manufacturing.
• Potential to close the materials loop, aligning with circular economic principles [14].

Within closed-loop recycling, sol–gel synthesis and co-precipitation methods have attracted significant attention due to their ability to produce high-purity, well-ordered layered oxides with tunable compositions [15,16]. However, systematic comparative studies of these two routes in the context of recycling spent LIBs remain limited.

The sol–gel method involves the formation of a homogeneous gel from precursor salts and organic acids acting as chelating agents. Upon calcination, this gel transforms into a crystalline oxide with fine particle size and uniform elemental distribution. In recycling contexts, organic acids such as citric, lactic, or malic acid serve dual roles as leaching and complexing agents, enabling efficient dissolution of metals from black mass and subsequent gel formation [17]. Sol–gel synthesis has been shown to yield NCM powders with high purity, small particle size (100–300 nm), and electrochemical performance comparable to commercial materials [18]. Nevertheless, sol–gel routes are time-intensive, require expensive reagents, and present scalability challenges, which limit their industrial applicability [19].

The co-precipitation method, in contrast, is based on the controlled precipitation of mixed hydroxide or oxalate precursors from aqueous solutions of Ni, Co, and Mn salts. Process parameters such as pH, temperature, stirring rate, and aging time critically influence precursor morphology and composition [20]. After mixing with lithium sources and calcination, the precursors are converted into layered NCM oxides. Co-precipitation offers superior scalability, cost-effectiveness, and reproducibility compared to sol–gel. Moreover, particle morphology can be tailored to achieve desirable spherical secondary particles that improve electrode packing density and electrochemical stability [21]. However, maintaining uniform composition and controlling impurity incorporation remain challenges, particularly when using mixed black mass leachates [22].

While both sol–gel and co-precipitation methods have been individually investigated for regenerating cathode materials from spent LIBs, direct comparisons of these two approaches within the same experimental framework are scarce. Understanding their relative strengths and weaknesses is essential for determining which method holds greater promise for industrial-scale recycling. In this work, the process is classified as closed-loop recycling rather than direct regeneration, as it involves partial dissolution of spent cathode materials and subsequent re-synthesis of layered NCM oxides. Unlike direct re-lithiation methods, this approach temporarily disrupts the original crystal structure but retains the elemental composition within the recycling loop.

The novelty of this work lies in the first direct side-by-side comparison of sol–gel and hydroxide co-precipitation synthesis routes integrated into a closed-loop recycling process using actual black-mass-derived mixed-metal leachate. The study provides a comprehensive evaluation that correlates synthesis route with precursor composition, phase purity, microstructure evolution, and electrochemical performance. Unlike direct re-lithiation processes, the approach investigated here involves temporary dissolution and subsequent re-synthesis of layered NCM and is therefore more appropriately classified as closed-loop recycling rather than direct regeneration.

The present study addresses this gap by systematically comparing sol–gel and co-precipitation approach for regenerating NCM cathodes from spent LIBs. Through this comparative analysis, the study contributes to the development of sustainable, industrially viable recycling strategies that can alleviate resource scarcity, reduce environmental impacts, and support the transition toward a circular battery economy.

2. Materials and Methods

Spent lithium-ion batteries (consumer-grade LiCoO₂ cells) from LLC “NPO” “LENENERGOMASH”, Saint Petersburg, Russia were discharged in 1 M NaCl solution until it reached 0 V potential, then were disassembled under inert conditions. All chemical reagents used in the future operations were supplied by LLC “Nevareactive”, Saint Petersburg, Russia. Spent LIBs were ground into powder using cutting mill, sifted using vibrating screens to obtain black mass (mixture of anode and cathode materials, the average chemical composition obtained by EDX-method is presented in

Table 1. Partial elemental composition of the black mass after mechanical pretreatment, obtained by EDX, with standard deviation, atomic %.

Sample	O	SDO	F	SDF	Al	SDAl	P	SDP	Co	SDCo	Cu	SDCu	Total
Black mass	47.67	0.61	20.90	0.98	12.08	0.32	1.12	0.15	18.03	0.85	0.20	0.23	100.00

) and subjected to leaching using organic acid 1.5 М malic acid and 3% H₂O₂ for both sol-gel and hydroxide co-precipitation synthesis approach. The elemental composition of the black mass (Table 1) and for the following hydroxide precursor (

Table 2. The chemical ratio of transition metals and standard deviations in synthesized samples in atomic %.

Sample	Ni	SDNi	Co	SDCo	Mn	SDMn	Total
Sol-gel	60.00	0.94	17.53	0.28	22.47	0.35	100.00
CP precursor	53.71	0.84	24.16	0.39	22.13	0.36	100.00
Co-precipitation	53.71	0.84	24.43	0.38	21.86	0.34	100.00

) was determined using SEM–EDX (EDX, Oxford Instruments X-Max 80, Oxford Instruments plc, Abingdon, UK) provides semi-quantitative atomic ratios of detectable elements; therefore, the reported values represent partial composition normalized to 100 at.% among the analyzed transition-metal-containing phases. Light elements such as Li and C are not reliably quantified by EDX and are therefore not included in the reported data. Solid/liquid ratio was 1:50 g/mL. Leaching was performed at 90 °C under continuous stirring at 300 rpm, with residence time 60 min. After the leaching stage, the suspension containing dissolved transition-metal ions and undissolved species (graphite, carbon black, and PVDF binder) was separated by gravity-assisted filtration. A glass filtration funnel equipped with an ashless filter (pore size 1–3 µm) was used to ensure complete removal of the solid carbonaceous fraction and polymeric residues. This pore size is sufficient to retain PVDF particles, graphite flakes, and carbon black agglomerates, while allowing the filtrate containing Ni²⁺, Co²⁺, Mn²⁺, and Li⁺ ions to pass through without loss of dissolved species. The collected filtrate was used directly for subsequent precursor preparation. The resulting solution contained Co²⁺ and Li⁺ in proportions suitable for following precursor synthesis. In Figure 1, the schematic flowchart of leaching process is shown.

The sol–gel synthesis approach was investigated in detail in our previous research [23]. In contrast to the previously reported process, which utilized purified spent LCO feedstock and a single sol–gel route, the present research applies an adapted malic-acid leaching protocol to chemically complex mixed black mass and introduces a comparative evaluation of two synthesis pathways—sol–gel and hydroxide co-precipitation—specifically designed for the regeneration of NCM622 from mixed-metal leachates. This synthesis adapted to integrate both leaching and precursor synthesis. For the sol–gel synthesis, Ni and Mn acetates were dissolved in the filtrate obtained after leaching (containing Co²⁺ and Li⁺ ions), followed by the addition of lithium acetate (LiAc) to obtain the transition metal (TM) solution. In contrast, for the co-precipitation route, LiAc was not added at this stage. A gel was formed under controlled evaporation at ~90 °C, dried in vacuum, was ground in agate mortar. The dried gel was subjected to a multistage heat-treatment protocol in a muffle furnace. First, the material was slowly heated at a rate of 1 °C/min to 150 °C with a residence time of 30 min to remove residual moisture and volatile organic components. After dehydration, the precursor was further heated to 500 °C and held for 4 h to complete combustion of organics and initiate phase formation. The final crystallization stage was carried out at 850 °C for 16 h to form the layered NCM oxide structure with R-3m symmetry. The scheme of the process is presented in Figure 2.

For the hydroxide co-precipitation route, mixed aqueous solutions in stoichiometric ratios of Ni, Mn acetates and Co/Li malate from the leaching process were prepared. The initial solution of 1 M ammonia water was also prepared and placed in the reactor. Precipitation was induced by dropwise addition of 2.81 M NaOH, 1.405 M transition metal (TM) solution and 3.66 M of NH₄OH under pH control (10–11.5 for various NCM types) using pH meter pH-160-MI at 50 °C with continuous stirring at 800 rpm. The addition rate of all solutions was the following: TM-1 mL/min, NaOH-1 mL/min, NH₄OH-0.3 mL/min. Argon was introduced in the reactor to prevent manganese oxidation and formation of undesirable phases. Precipitated hydroxide was aged for 24 h with continuous stirring of 800 rpm; after that, the precipitate was filtered and washed with distillated water until reaching a neutral pH value to remove residual Na⁺ ions and dried at 95 °C in vacuum. In Figure 3, the flowchart of co-precipitation route is shown.

A comprehensive set of characterization techniques was employed to investigate the structural and electrochemical properties of the regenerated NCM powders. The microstructural features and surface morphology of the samples were examined using a Tescan Mira 3 scanning electron microscope, Tescan, Brno, Czech Republic, operated in secondary electron (SE) and back-scattered electron (BSE) mode. Elemental composition was determined through energy-dispersive X-ray spectroscopy (EDX, Oxford Instruments X-Max 80). Phase analysis was conducted on a Bruker D8 Advance X-ray diffractometer (Bruker Corporation, Billerica, MA, USA) equipped with a CuKα radiation source (λ = 1.5406 Å). Data collection was performed within the 2θ range of 15–85°, with a step increment of 0.03° and an acquisition time of 0.8 s per step. Phase identification was carried out using DIFFRAC.EVA v5.0.

To evaluate the electrochemical performance, electrodes were fabricated from slurries containing the active synthesized materials, conductive carbon, and PVDF binder in N-methyl-2-pyrrolidone (NMP). For cathodes, the mixture comprised 92 wt% active material, 3 wt% Super P carbon, and 5 wt% PVDF (Solef 5130). The homogenized pastes were coated on aluminum foil collectors using a film applicator (MTI Corp., Richmond, CA, USA, MSK-AFA-III, clearance 200 μm). The coated foils were dried under vacuum at 80 °C for 24 h, calendared, and then assembled into CR2032 coin cells under an argon-filled glovebox. Metallic lithium served as the counter/reference electrode, with TC-E918 electrolyte (Tinci, Guangzhou, China), the constant volume of 40 μL per cell and double-layer Celgard 2325 separators.

Galvanostatic charge–discharge measurements were performed using a Neware Battery Testing System (5 V, 10 mA) (Neware Technology Limited, Shenzhen, China) within the voltage range of 2.8–4.2 V. Cyclic voltammetry (CV) experiments were conducted between 2.5 and 4.3 V at a scan rate of 0.1 mV s⁻¹. All electrochemical tests were performed at 25 ± 1 °C.

3. Results and Discussion

3.1. Morphology and Microstructure, Phase Analysis

The morphology of regenerated NCM materials is strongly dependent on synthesis route, precursor chemistry, and calcination parameters. SEM analysis revealed clear distinctions between sol–gel and co-precipitation methods.

Figure 4 presents SEM micrographs of the NCM622 powder synthesized via the sol–gel method, shown at different magnification levels. In Figure 4a (10 μm scale), the material exhibits a highly porous, continuous network composed of irregular secondary agglomerates. The morphology demonstrates a sponge-like interconnected structure formed during organic binder decomposition and subsequent calcination, which typically accompanies sol–gel processing and leads to extensive internal porosity.

The higher-magnification image (Figure 4b, 1 μm scale) reveals that these secondary structures consist of densely assembled primary particles with sub-micrometer dimensions. The annotated measurements indicate characteristic grain sizes in the range of 0.40–0.49 μm. The particle contacts exhibit neck-type junctions, suggesting partial coalescence during calcination. The fine particle size and hierarchical porosity may enhance electrolyte infiltration and shorten lithium-ion diffusion pathways.

Overall, the obtained powders demonstrate the characteristic spherical-granular morphology typical of sol–gel materials, combining uniform faceted particle surfaces and well-defined intergranular boundaries, indicating advanced grain growth. While the SEM micrographs provide information primarily on particle size and packing behavior, the crystallinity and structural ordering were confirmed separately via X-ray diffraction analysis, which verified the formation of a well-developed layered phase. Such structural compactness and uniform microstructure are favorable for achieving high electrochemical performance.

Hydroxide co-precipitated precursor and lithiated NCM622 powder displayed in Figure 5. In Figure 5a, as can be seen, the precursor material consists of fine, spherical or near-spherical primary particles that are densely packed into agglomerates. The particle size distribution appears relatively narrow, with individual particle diameters ranging from approximately 3.12 µm to 4.93 µm. The overall structure is granular and uniform, which is characteristic of materials produced via co-precipitation, where controlled nucleation and growth lead to homogenous particle formation. The surface texture of the precursor particles is smooth, suggesting good crystallinity and minimal surface defects at this stage. This morphology is crucial for ensuring uniform lithium diffusion and structural stability in the final cathode material. In Figure 5b,c, the microstructure of the NCM622 material after high-temperature lithiation is shown. The material exhibits a highly porous, sponge-like morphology composed of interconnected secondary particles. These secondary particles are aggregates of smaller primary crystallites, forming a network with visible voids and channels. The annotated dimensions reveal a range of aggregate sizes: from 11.6 µm to 27.3 µm, indicating significant particle growth and sintering during the high-temperature calcination process. The higher-magnification view provides a closer look at the primary crystallites within the calcined NCM622 material. The particles now exhibit more defined, angular surfaces, which are characteristic features typically associated with faceted grain growth. Such morphology may suggest the development of a more ordered layered oxide structure; however, confirmation of crystallization is provided separately by XRD analysis. Particle sizes are significantly reduced compared to the precursor, ranging from 0.77 µm to 8.04 µm. Notably, many particles fall below 2 µm, indicating fragmentation or recrystallization during calcination.

Compared to NCM622 synthesized via the sol–gel method—which typically yields fine, irregularly shaped primary particles (<2 µm) with limited agglomeration and relatively low intrinsic porosity—the co-precipitation-derived material demonstrates well-defined spherical secondary agglomerates composed of interconnected nanocrystallites. Such hierarchical morphology, clearly observed in the SEM micrographs, generates an open porous network that facilitates electrolyte infiltration and significantly improves Li⁺ diffusion pathways throughout the particle interior. Enhanced porosity reduces ionic transport resistance and mitigates mechanical stress generated during repeated lithiation/delithiation, thereby contributing to improved cycling stability and reduced polarization during electrochemical operation.

While the sol–gel route offers superior molecular-level homogeneity and more precise stoichiometric control, the resulting loosely aggregated particles lack the engineered secondary-particle architecture required for high tap density, uniform current distribution, and structural durability under cycling. In contrast, spherical secondary particles produced via hydroxide co-precipitation provide improved electrode packing and mechanical integrity, leading to enhanced retention of electrochemical performance over prolonged cycling.

EDX measurements confirmed that the regenerated powders had near-stoichiometric compositions, the results obtained by EDX method are presented in Table 2. The Ni:Co:Mn ratios were calculated as averaged values from multiple EDX point measurements after excluding oxygen content to obtain normalized transition-metal atomic proportions; standard deviations are also included. The sol–gel sample showed almost excellent compositional control, with Ni:Co:Mn ratios close to the intended 6:2:2. There is a slight cobalt deficiency and manganese excess. This near-ideal composition arises from the molecular-level homogeneity inherent to the sol-gel process. The small Co deficit may stem from volatilization of cobalt species during high-temperature calcination or from incomplete incorporation due to slightly different decomposition kinetics of cobalt-containing precursors compared to Ni and Mn. The compensatory increase in Mn content suggests that the system maintains charge balance and overall stoichiometry, albeit with a minor shift toward Mn-rich regions.

Before discussing the composition of the final co-precipitated oxide, the chemical composition of the hydroxide precursor was evaluated to identify potential metal losses occurring during the precipitation step. The obtained deviation of stoichiometry indicates preferential precipitation of Co²⁺ and Mn²⁺ relative to Ni²⁺, likely caused by competition between metal ions for complexation with residual malate ligands and ammonia, as well as differences in precipitation kinetics. Ni²⁺ may form more stable complexes, hindering its full incorporation into the hydroxide sediment or causing partial retention in the filtrate during washing. These precursor-level deviations directly influence the composition of the calcined co-precipitated NCM622 powder.

As a result, the final co-precipitation-derived oxide sample shows a significant nickel deficiency and excess cobalt relative to the intended 6:2:2 ratio. This off-stoichiometry has direct electrochemical implications: reduced Ni limits the theoretical capacity (given the dominant role of the Ni²⁺/Ni⁴⁺ redox couple), while increased Co may enhance rate performance and conductivity but potentially worsen structural stability. Moreover, deviations from ideal stoichiometry intensify cation mixing (migration of Ni²⁺ into Li layers), which increases internal resistance and accelerates capacity fading during cycling. Therefore, precise control of nickel retention during the precursor precipitation stage is essential for optimizing both initial capacity and long-term cycling stability of NCM622 cathodes.

The structural evolution of the NCM622 cathode material was monitored by X-ray diffraction from the hydroxide precursor stage to the final layered oxide phase. Two synthesis routes were compared to evaluate their impact on crystallinity, phase purity, and structural ordering. In Figure 6 there is a diffractogram of precursor of NCM obtained from co-precipitation approach. The XRD pattern exhibits broad, low-intensity reflections centered at 2θ ≈ 23.5°, 33.4°, and 60.7°, which are assigned to the (001), (100/101), and (110/111) planes of β-Ni(OH)₂ (PDF #00-014-0117). The absence of sharp peaks indicates a poorly crystalline or nanocrystalline nature, typical of hydroxide precursors formed under mild aqueous conditions. The broadening of the (001) peak suggests limited stacking along the c-axis, consistent with the layered structure of β-Ni(OH)₂, where metal hydroxide sheets are held together by weak hydrogen bonding. No impurity phases such as NiO, α-Ni(OH)_2, or mixed-metal oxides are detected, confirming the formation of a single-phase hydroxide precursor suitable for subsequent calcination into layered NCM. The phase, morphology, and crystallinity of the hydroxide precursor directly dictate the microstructure, compositional homogeneity, and electrochemical performance of the final NCM cathode material. In particular, β-Ni(OH)₂ is preferred over α-Ni(OH)₂ due to its thermodynamic stability and well-defined layered structure, which facilitates uniform Li⁺ insertion/extraction and minimizes structural degradation during cycling. Poorly controlled precipitation can lead to phase impurities, non-uniform metal distribution, or amorphous regions—all of which degrade capacity retention and rate capability. Therefore, precise control over precursor phase and morphology is not merely preparative, it is a fundamental design parameter for high-performance cathode materials.

The XRD patterns of the calcined NCM622 materials (Figure 7) reveal well-defined, sharp peaks corresponding to the layered R-3m crystal structure (PDF 00-056-0146), confirming successful formation of the desired layered oxide phase in both samples. Quantitative analysis of peak intensities and structural parameters is presented in

Table 3. The structural characteristics of synthesized layered-oxide materials.

Sample	hkl	Intensity	R-Factor	I003/I104	Split 006/102
Sol-gel	003	14,004	0.739	1.896	yes
	101	3863
	006	1042
	102	1816
	104	7386
Co-precipitation	003	32,217	0.660	1.720	yes
	101	10,781
	006	2720
	102	4342
	104	18,732

. The refinement results show that the co-precipitated sample exhibits a lower R-factor value (0.66 vs. 0.739), which should primarily be interpreted as an indicator of improved goodness-of-fit between the experimental XRD data and the applied structural model rather than a direct measure of cation mixing or structural disorder. However, the more relevant metric for evaluating cation mixing in layered oxide cathode materials is the I(003)/I(104) intensity ratio. This ratio is commonly used as a qualitative descriptor of the degree of cation disorder between Li⁺ sites and Ni²⁺ ions in layered NCM oxides. A lower value of this ratio typically reflects greater migration of Ni²⁺ into the lithium layer, which disrupts the ideal layered arrangement. Such antisite defects not only reduce the amount of electrochemically active lithium but also hinder Li⁺ diffusion pathways, ultimately leading to diminished capacity and overall poorer electrochemical performance [24]. A significantly higher ratio obtained for the sol–gel sample (1.896 vs. 1.720 for the co-precipitated sample) suggests reduced Ni²⁺ migration into Li⁺ layers and improved layered structural ordering. This indicates that, despite the slightly higher R-factor, the sol–gel-derived material possesses superior intrinsic structural arrangement within the R-3m framework, which is beneficial for maintaining fast Li⁺ diffusion kinetics and structural stability during electrochemical cycling. Both samples exhibit clear splitting of the (006)/(102) doublet, confirming the development of a well-ordered layered structure without significant rock-salt impurities [20,22,25].

Notably, despite the slightly higher R-factor, the sol-gel-derived material displays sharper peaks and higher absolute intensities for major reflections, suggesting superior crystallinity and grain growth during calcination. This enhanced structural quality likely stems from the molecular-level homogeneity inherent to the sol-gel process, which minimizes local compositional fluctuations that can promote defect formation.

In summary, while both synthesis routes yield phase-pure NCM622 with the desired R-3m symmetry, the sol-gel method produces a more structurally ordered material with reduced cation mixing, as evidenced by the higher I(003)/I(104) ratio and clearer peak splitting. These structural advantages are expected to translate into improved electrochemical performance, particularly in terms of initial capacity, rate capability, and cycle life—even if the co-precipitated material exhibits marginally lower R-factor values.

3.2. Electrochemical Properties

Figure 8 presents the galvanostatic charge–discharge voltage profiles of NCM622 cathode materials synthesized via sol-gel and co-precipitation methods, recorded during the 1st, 5th, and 10th cycles within the voltage window of 2.8–4.3 V vs. Li⁺/Li at a constant current rate 0.1 C. All curves exhibit the typical S-shaped voltage profile characteristic of layered cathodes, with two distinct regions of charge process, with gradual voltage rise from 3.6 V to 4.3 V, reflecting progressive delithiation of the layered structure and associated redox reactions of Ni²⁺/Ni⁴⁺ and Co³⁺/Co⁴⁺, and discharge process with a smoother, slightly sloping curve from 4.3 V down to 2.8 V, indicating reversible lithium reinsertion.

The presence of a small plateau near 3.8–3.9 V during discharge is more pronounced in the sol-gel material. This fact suggests minor phase transitions or kinetic hysteresis, consistent with the layered-to-spinel or layered-to-rock-salt transformations that can occur in Ni-rich materials under cycling stress. The sol-gel material exhibits higher initial specific capacity, reaching 170 mAh/g in the first cycle, which is close to the theoretical capacity of NCM622 (~180 mAh/g). The Coulombic efficiency on the 1st cycle for this material—99%. The voltage profiles remain relatively stable between the 1st and 10th cycles, with only minor voltage decay and capacity fade—indicating good structural integrity and minimal irreversible side reactions. The overlap of curves across cycles suggests high reversibility and low polarization growth, consistent with the superior crystallinity and reduced cation mixing observed in XRD analysis (I(003)/I(104) = 1.896).

The co-precipitated material shows an initial charge capacity of 181 mAh/g and a discharge capacity of 149 mAh/g, corresponding to a Coulombic efficiency of 79%. The reduced initial efficiency does not arise solely from structural and morphological factors, although the compositional deviation from the target stoichiometry and smaller crystallite domain size may limit the accessible active material volume and increase polarization resistance. Additional contributions include interfacial processes such as electrolyte decomposition and the initial formation of CEI layer, which consume lithium irreversibly during early cycles and are typically more significant in materials with higher specific surface area and open porous structures. A more pronounced voltage fade between the 1st and 10th cycles indicates increasing internal resistance and structural degradation, while the downward shift of discharge curves suggests stronger polarization and possible microcrack development due to mechanical stress during repeated Li⁺ insertion/extraction. The discharge curves shift downward more significantly than those of the sol-gel sample, indicating greater polarization and possible accumulation of interfacial impedance or microcracking due to mechanical stress during repeated Li⁺ extraction/insertion.

To better illustrate the dynamic evolution of reversibility of lithium-ion transport processes over cycling, the Coulombic efficiency values for both samples over the first 10 cycles are presented in Figure 9. As shown, the sol–gel material maintains a relatively stable Coulombic efficiency near 98–100% throughout cycling, indicating excellent rechargeability and suppressed side reactions. Conversely, the co-precipitation-derived sample exhibits a substantial improvement in Coulombic efficiency after the initial cycle, rapidly increasing from 79% to approximately 98% by the 10th cycle. This behavior suggests that initial structural stabilization and formation of surface passivation layers contribute to performance improvement. The convergence of Coulombic efficiency values for both materials after the first few cycles reveals that although initial losses are higher in the co-precipitated sample, long-term electrochemical reversibility can be effectively stabilized.

Figure 10 presents the evolution of discharge-specific capacity as a function of cycle number, revealing starkly different degradation behaviors. The sol-gel sample (black squares) exhibits a rapid initial capacity loss from 180 mAh/g in cycle 1 to 140 mAh/g by cycle 10—followed by a slower, steady decline to ~60 mAh/g by cycle 40, thus capacity loss is 75%. This behavior is typical of Ni-rich cathodes undergoing surface reconstruction, CEI growth, and possible microcracking due to anisotropic lattice strain during repeated Li⁺ extraction/insertion. The steep early decay may also reflect incomplete activation or irreversible side reactions at the electrode-electrolyte interface. In contrast, the co-precipitation sample (red circles) demonstrates remarkable capacity stability: after an initial rise (from 145 to 152 mAh/g in cycles 2–5), it maintains a nearly flat plateau (~150–152 mAh/g) until cycle 25, followed by only a moderate decline to ~145 mAh/g by cycle 40, the capacity loss is 3%. This suggests that despite its lower initial capacity and compositional deviation, the co-precipitated material possesses enhanced structural resilience during prolonged cycling—possibly due to its finer particle size and porous agglomerate morphology (observed in SEM), which mitigates mechanical stress and accommodates volume changes more effectively than the denser, larger-grained sol-gel particles.

Figure 11 presents the second-cycle cyclic voltammograms (CV) of NCM622 cathode materials synthesized via co-precipitation (red curve) and sol-gel (black curve) methods, recorded at a scan rate of 0.1 mV/s between 2.6–4.3 V vs. Li⁺/Li. The co-precipitated sample exhibits a significantly lower peak separation ΔE_p of 0.16 V (160 mV), compared to 0.36 V (360 mV) for the sol–gel sample, confirming reduced polarization and faster Li⁺ kinetics due to the more porous secondary particle architecture. The CV profiles reveal distinct redox signatures that reflect fundamental differences in electronic structure, reaction kinetics, and compositional homogeneity—all of which are directly linked to the materials’ synthesis-dependent microstructure and electrochemical behavior.

Both curves exhibit a single prominent anodic peak during charge (oxidation) and a corresponding cathodic peak during discharge (reduction), characteristic of Ni²⁺/Ni⁴⁺ redox activity in layered NCM oxides. Minor contributions from Co³⁺/Co⁴⁺ may also be present but are not resolved as separate peaks under these conditions.

The smaller ΔE_p observed for the co-precipitated sample indicates lower electrochemical polarization and potentially faster reaction kinetics, which is consistent with its porous, spherical secondary particle morphology (SEM), which facilitates ion transport and reduces interfacial resistance. In contrast, the larger ΔE_p for the sol-gel material suggests higher activation barriers, possibly due to denser particle packing or minor surface passivation.

The anodic peak shift from 3.58 V (sol-gel) to 3.68 V (co-precipitation) reflects a thermodynamic change in the redox process. This shift is directly attributable to the Ni deficiency in the co-precipitated sample. Nickel is the primary redox-active element in NCM622; reducing its content lowers the average oxidation state of transition metals, thereby shifting the Ni²⁺/Ni⁴⁺ redox couple to slightly higher potentials during oxidation (charge) and lower potentials during reduction (discharge). This explains why the cathodic peak also shifts downward from 3.94 V (sol-gel) to 3.84 V (co-precipitation), indicating easier reduction of Ni⁴⁺ back to Ni²⁺ in the Ni-deficient matrix, likely due to reduced electrostatic repulsion within the lattice or altered local coordination environments Similar anodic and cathodic peak shifts in Ni-rich NCM materials have been reported in the literature, where Ni content variation affects redox potential and structural stability [25,26].

The sol-gel sample exhibits sharper, more symmetric peaks, reflecting its high crystallinity and reduced cation mixing. This structural perfection enables well-defined, reversible phase transitions during Li⁺ extraction/insertion. The co-precipitation sample, while showing broader peaks, maintains excellent reversibility (low ΔE_p). Its porous agglomerate morphology enhances electrolyte access and mitigates kinetic limitations, compensating for its lower crystallinity and compositional deviation.

These CV observations align perfectly with the long-term cycling behavior. The sol-gel material’s high initial capacity correlates with its sharp, intense redox peaks, which are indicative of efficient utilization of active material. However, its rapid capacity fade (to 60 mAh/g by 40th cycle) is mirrored in the larger ΔE_p, suggesting increasing internal resistance or structural degradation during repeated cycling, possibly due to microcracking in its dense, irregular particles. The co-precipitated material’s stable capacity retention (~150 mAh/g after 40 cycles) is supported by its low ΔE_p and stable peak positions, implying minimal accumulation of irreversible side reactions or impedance growth, enabled by its favorable microstructure that buffers mechanical stress.

While the sol-gel sample’s CV profile reflects excellent thermodynamics and structural order, the co-precipitated sample’s profile reveals practical advantages: lower polarization, better kinetics, and resilience to degradation—making it more suitable for real-world battery operation despite its off-stoichiometry.

4. Conclusions

This study presents a systematic comparison of sol–gel and hydroxide co-precipitation routes for the closed-loop recycling of NCM622 cathode materials from spent lithium-ion batteries using organic malic acid leaching. Both methods successfully produced phase-pure layered oxides with electrochemical activity comparable to commercial NCM, demonstrating the viability of closed-loop recycling as a sustainable alternative to conventional pyrometallurgical and hydrometallurgical approaches. A comparative summary of the structural, morphological, compositional, and electrochemical performance indicators for both synthesis routes is presented in

Table 4. Comparative summary.

Parameter	Sol-Gel	Co-Precipitation
Target TM stoichiometry	60:20:20
Measured TM stoichiometry	60:17.53:22.47	53.71:24.43:21.86
Precursor stoichiometry	-	53.71:24.16:22.13
Morphology	Spherical–granular secondary particles composed of well-faceted primary crystallites; smooth surfaces; medium porosity	Spherical agglomerates with hierarchical porous structure; interconnected nanocrystallites; high accessible surface area
Primary particle size	1–9 μm	0.77–8.04 μm
Secondary particle size	-	11.6–27.3 μm
XRD phase purity	Single phase R-3m layered oxide	Single phase R-3m layered oxide
Initial discharge capacity	170 mAh/g	149 mAh/g
CV ΔEp	0.22 V	0.16 V
Capacity at 40 cycles	60 mAh/g	145–150 mAh/g
Coulombic efficiency (1st cycle)	99%	79%
Strengths	High crystallinity, low cation mixing, high initial capacity	Excellent cycling stability, lower polarization, better process scalability
Limitations	Faster degradation due to dense morphology and mechanical fracture	Off-stoichiometry (Ni deficiency), peak broadening
Industrial scalability	Moderate—complex precursor chemistry	High—compatible with commercial processing
Overall assessment	High theoretical performance but limited practical durability	Most promising for industrial large-scale recycling and closed-loop reuse

.

The sol–gel material demonstrated near-target stoichiometry (Ni:Co:Mn = 60.00:17.53:22.47 at.%), reduced cation mixing (I(003)/I(104) = 1.896), and a high initial specific capacity of ~170 mAh/g. However, its dense and irregular granular morphology led to rapid capacity decay—retaining only ~35% of the initial capacity after 40 cycles (~60 mAh/g). This behavior is consistent with microcracking, surface reconstruction to spinel/rock-salt phases, and CEI growth, which have been widely reported for Ni-rich layered cathodes with limited structural compliance.

In contrast, the hydroxide co-precipitation method, despite a slight Ni deficiency (53.71 at.%) and corresponding Co/Mn excess relative to the target 60:20:20 composition, produced spherical secondary agglomerates with a hierarchical porous structure that effectively accommodated mechanical stress during cycling. This morphology translated into exceptional capacity retention—maintaining ~150 mAh g⁻¹ for 40 cycles with only gradual degradation thereafter. The lower polarization (ΔE_p = 0.16 V in CV) and stable redox behavior further confirm its kinetic and structural robustness.

Importantly, the results indicate that precise control of stoichiometry during precursor synthesis is essential for optimizing the electrochemical performance of co-precipitated materials. The study identified several critical synthesis parameters that strongly influence the incorporation of Ni, Co, and Mn: pH stability (10.6–10.7), ammonia concentration (≤1 M to avoid excessive Ni-ammine complexation), metal-ion feed ratios (to compensate for preferential Co/Mn precipitation), mixing intensity (800 rpm), and controlled washing to avoid Ni dissolution losses. Optimization of these factors can mitigate compositional deviation and enable attainment of the target stoichiometry in the final oxide. Future work will include ICP-OES mass-balance analysis of filtrates and washings and implementation of compensatory Ni-feed strategies to improve metal recovery precision.

Critically, co-precipitation offers significant advantages in scalability, cost-efficiency, and compatibility with existing industrial cathode manufacturing infrastructure. While minor compositional variations can be minimized through process optimization, the inherent morphological benefits of co-precipitation make it the more industrially viable pathway for large-scale battery recycling.

These findings underscore a key principle in sustainable battery design: functional performance and process scalability often outweigh theoretical perfection. For a circular battery economy to succeed, recycling strategies must not only recover value but also integrate seamlessly into manufacturing workflows. Hydroxide co-precipitation, when coupled with green leaching using organic acids, emerges as a promising route to achieve this balance—enabling the transformation of waste into high-performance cathodes with minimal environmental footprint.