Maximizing energy utilization and lithium leaching efficiency via sequential electrochemical dual-oxidation and soaking-relaxation

TL;DR

A new two-stage electrochemical method efficiently recovers lithium from spent batteries with high efficiency and reduced energy use.

Contribution

A two-stage electrochemical process for lithium leaching with reduced energy consumption and high efficiency.

Findings

The two-stage method achieves 99.87% lithium leaching efficiency from commercial NCM111 cathodes.

Electric energy consumption is reduced by 49.78% through lattice oxygen transformation during the process.

Abstract

Given the escalating global demand for lithium resources, optimizing electric energy consumption in the electrochemical dual-oxidation (EDO) process, which includes both electrode oxidation and electrocatalytic oxidation, for lithium leaching from spent lithium-ion cathodes, is imperative. Herein, we propose an energy-effective two-stage continuous oxidation method for lithium leaching from various composition spent ternary lithium-ion batteries (NCM) cathodes. Coupling EDO (stage I) with soaking relaxation (stage II) enables both commercial and spent LiNi1/3Co1/3Mn1/3O2 (NCM111) cathodes to achieve optimal electric energy efficiencies, with lithium leaching efficiencies of 99.87% and 98.12%, respectively. A comprehensive mechanism study reveals that the EDO not only drives lithium leaching from NCM111 lattice at stage I, but also effectively induces the transformation of lattice oxygen (O2-) into oxidized lattice oxygen (On-, n < 2), thus driving the continuous lithium leaching at stage II with 49.78% reduction in electric energy consumption. This work unravels the electrical energy profile and structural dynamics during the EDO lithium leaching process, which paves the way for implementing the EDO method for the industrial sustainable recovery of critical metals.

This study introduces a two-stage electrochemical method for lithium recovery from spent LiNi1/3Co1/3Mn1/3O2 (NCM) batteries, achieving over 98% efficiency with nearly 50% less electric energy use via lattice oxygen oxidation and ion exchange.

Figures

Relationship between electric energy input and lithium leaching efficiency.a Lithium leaching efficiency variation under constant EDO at 2.5 V. b Time-dependent current variation. c Schematic illustration of the two-stage continuous oxidation lithium leaching process. The error bars in (a) are presented as the mean ± standard deviation of three independent experiments. Source data for the figure are provided as a Source data file.

Time-dependent leaching efficiency and electric energy input.a Lithium leaching during stage I. b Lithium leaching during stage II. c Overall metal leaching during stage II. The abbreviation EDO in (b) denotes electrochemical dual-oxidation. Error bars are presented as the mean ± standard deviation of three independent experiments. Source data for the figure are provided as a Source data file.

Active oxidation species evolution and alkaline ion effects within electrolyte.a EPR spectra. b, c Changes of free chlorine content during stage I and stage II. d Lithium leaching in fresh 0.5 M NaCl solution. e Effects of different cations. f Effect of Na^+^ concentration. The error bars in (d–f) are presented as the mean ± standard deviation of three independent experiments. Source data for the figure are provided as a Source data file.

a, during the first 20 min of the EDO process, the Li content in the electrode material decreased, and the O3 phase electrode material transformed into P2 phase. After 20 min, the lithium content decreased, while the sodium content increased in the electrode material due to the insertion of Na^+^. The transition from the P2 to O2 phase began, and this process was completed by the end of stage I as Na^+^ continued to be inserted. These results suggest that during the period of 0–20 min, EDO-driven lithium leaching is dominant, where O3-phase Li_0.7_Ni_1/3_Co_1/3_Mn_1/3_O_2_ transforms into P2-p

Full text

Introduction

The global transition towards a low-carbon and cleaner energy framework has elevated lithium as a critical element for electric vehicle (EV) power batteries. The increasing demand and the uneven distribution of lithium resources pose significant challenges to the security of the lithium supply. Traditional lithium mining practices result in substantial environmental damage. Simultaneously, the widespread use of electric vehicles leads to a rapid increase in spent lithium-ion batteries (LIBs). By 2030, the global volume of spent lithium-ion batteries is projected to reach 11 million tons^1^. Sustainable recycling of these batteries is crucial to alleviating the supply-demand imbalance of lithium resources, reducing environmental pollution, and promoting a circular economy for lithium resources. Among various lithium-ion batteries, ternary lithium-ion batteries (NCM) are gaining prominence due to their higher energy density and the rich critical metals contained within the cathode compared to lithium iron phosphate (LiFePO_4_) batteries^2^. The rich economic value within the spent NCM batteries exacerbates the issue of spent battery disposal and underscores the importance of efficient lithium recovery.

Over the past few decades, hydrometallurgy has been recognized as the primary industrial technology for recovering lithium from spent batteries^3^. Non-selective leaching, involving hazardous acidic reagents and redox agents, can achieve high lithium leaching efficiency; however, it often results in complex and critical metal separation procedures during subsequent redundant separation processes^4–8^. In this regard, the selective leaching method is highly desirable for improving the selective lithium leaching and recovery efficiency.

Recently, electrochemical methods have emerged as a promising alternative approach for recycling spent lithium-ion batteries due to their environmentally friendly nature and selective lithium leaching performance^9–12^. The electrochemical methods utilize electrical energy as a driving force for lithium-ion migration, thereby eliminating the need for harmful chemicals and reducing the environmental impact associated with traditional hydrometallurgical processes^13,14^. While the reported electrochemical methods, including direct electrode oxidation^9–11,15^, and indirect electrocatalytic oxidation^16^, typically achieve a moderate lithium leaching efficiency. Electrochemical dual-oxidation (EDO), an electrochemical method that combines electrode oxidation with electrocatalytic oxidation^17^, has been proposed to improve lithium leaching efficiency and kinetics. However, a significant challenge remains regarding optimizing the energy input profile and lithium leaching efficiency in an energy-effective and sustainable manner.

Herein, by fine-tuning the electric energy input and lithium leaching efficiency, a continuous oxidation lithium leaching method is proposed by combining an EDO stage (Stage I) with a soaking relaxation stage (Stage II). Comprehensive mechanistic studies reveal that EDO is the dominant factor in driving lithium leaching from the LiNi_1/3_Co_1/3_Mn_1/3_O_2_ lattice at stage I, while the oxidized lattice oxygen further drives ion exchange and significantly contributes to lithium leaching at stage II. Unlike the single-stage EDO method, which consumes high electric energy, our method maximizes both the electric energy conversion efficiency and the lithium leaching efficiency through the synergistic effects of stages I and II. This work elucidates the optimization of the electric energy profile and structural dynamics in the sustainable two-stage EDO lithium leaching process, paving the way for the industrial implementation of the electrochemical method for sustainable and environmentally friendly recovery of critical metals.

Results

Concept within this work

To evaluate the necessity of optimizing energy input for efficient lithium leaching, an initial EDO lithium leaching experiment was conducted using commercial LiNi_1/3_Co_1/3_Mn_1/3_O_2_ (NCM111) cathode material as a representative anode. As shown in Fig. 1a, although >99% lithium leaching efficiency has been achieved within 150-min of EDO, the leaching speed significantly decreased at the later stage, accompanied by a decline in both faradaic efficiency (see Supplementary Note 3 for calculation details) and current (Fig. 1b). Specifically, after 60 min, the oxygen evolution reaction (OER) and other side reactions at the electrode-electrolyte interface dominant in electric energy consumption, leading to a continuous decrease in electric energy utilization efficiency for lithium leaching and an increase in overall energy consumption. To mitigate overall energy consumption and enhance electric energy conversion efficiency for lithium leaching in the electrolyzer, we classified the overall electrochemical lithium leaching process into two stages. We proposed a two-stage continuous oxidation lithium leaching method. As shown in Fig. 1c, in the EDO stage (stage I), a constant potential of 2.5 V is applied to the electrolyzer for effective EDO of lithium leaching. In stage II, no external electric energy is input, and in-depth lithium leaching is achieved through crystal structural relaxation when the electrode is continuously soaking within the oxidative electrolyte.Fig. 1. Relationship between electric energy input and lithium leaching efficiency.a Lithium leaching efficiency variation under constant EDO at 2.5 V. b Time-dependent current variation. c Schematic illustration of the two-stage continuous oxidation lithium leaching process. The error bars in (a) are presented as the mean ± standard deviation of three independent experiments. Source data for the figure are provided as a Source data file.

Maximizing electric energy efficiency within the two-stage continuous oxidation lithium leaching process

The influence parameters within the two-stage continuous oxidation lithium leaching process were systematically evaluated to maximize the electric energy utilization and lithium leaching efficiency. During stage I, we explored the relationship between electric energy input and lithium leaching efficiency from 0 to 60 min (significantly shorter than the 150 min previously used). As shown in Fig. 2a, when the EDO time was extended to 50 min, 55 min, and 60 min, the corresponding energy input ratios increased to 112.71%, 119.97%, and 126.86%, respectively, resulting in EDO lithium leaching efficiencies of 80.73%, 82.89%, and 84.43%. This indicates that although 100% of the theoretical electric energy is input into the electrolyzer, the faradic efficiency for lithium leaching is far less than 100%. The unavoidable occurrence of side reactions, including lattice oxygen oxidation, chloride oxidation, and oxygen evolution, results in an additional consumption of the input electric energy. Subsequently, we interrupted the electric energy input and investigated lithium leaching efficiency via soaking relaxation in stage II. Interestingly, for the electrode after EDO of 50 min, 55 min, and 60 min, as shown in Fig. 2b, the lithium leaching efficiency continuously increased with the extension of soaking relaxation time. The leaching equilibria arrived after 3 h, with maximum lithium leaching efficiencies of 98.58%, 99.87%, and 99.90%, respectively. Among the three electrodes, those with 55-min EDO and 60-min EDO achieved optimal lithium leaching efficiencies of greater than 99.80%. Therefore, for the subsequent experiments, 55 min is the optimal EDO time at stage I, as it achieved 99.87% lithium leaching efficiency with minimal electric energy consumption. To evaluate the leaching selectivity, the time-dependent leaching efficiencies of all critical metal cations were monitored at stage II (Fig. 2c). It can be seen that, in addition to the significant increase in lithium leaching efficiency, the leaching efficiency of Ni, Co, and Mn remained below 0.4%, which suggests remarkable selectivity for lithium leaching.Fig. 2. Time-dependent leaching efficiency and electric energy input.a Lithium leaching during stage I. b Lithium leaching during stage II. c Overall metal leaching during stage II. The abbreviation EDO in (b) denotes electrochemical dual-oxidation. Error bars are presented as the mean ± standard deviation of three independent experiments. Source data for the figure are provided as a Source data file.

Leaching kinetics during the two-stage continuous oxidation lithium leaching process

Previous studies reveal that lithium leaching is a solid-liquid heterogeneous process controlled by a combination of diffusion and chemical reactions. The quantitative leaching rates can be controlled by three steps: (1) liquid film diffusion (Eq. 1); (2) chemical reaction (Eq. 2); and (3) product layer diffusion (Eq. 3)^18,19^.

\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$X=k{{\cdot }}t$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$1-{\left(1-X\right)}^{\tfrac{1}{3}}=k{{\cdot }}t$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$1-3{\left(1-X\right)}^{\tfrac{2}{3}}+2\left(1-X\right)=k{{\cdot }}t$$\end{document}

Where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$X$$\end{document} is the fraction of the leaching efficiency, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$k$$\end{document} is the rate constant of the controlling step, and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$t$$\end{document} is the leaching time (min). As shown in Supplementary Fig. 3, under different rate-control assumptions, the lithium leaching process can be divided into three stages based on the changes in tendency, corresponding to 0–20 min, 25–55 min, and 85–235 min. These variations in the trend indicate that the controlling mechanism underwent three distinct shifts during the lithium leaching process^20,21^. The fitted data and corresponding data analysis for each equation are presented in Supplementary Table 3 and Supplementary Note 4, respectively. Based on the study, the lithium leaching process in stage I is predominantly controlled by chemical reactions during the initial phase, gradually evolving into a combined control mechanism involving chemical reactions and product layer diffusion. In stage II, the process is primarily controlled by diffusion through the product layer, so a long relaxation time is required to achieve near 100% lithium leaching efficiency.

Morphology and structural dynamics during the two-stage continuous oxidation lithium leaching process

To investigate the morphological and structural evolution of the cathode material during the two-stage continuous oxidation lithium leaching process, the particle morphology of NCM111 at different stages was examined using scanning electron microscopy (SEM) images (Supplementary Fig. 4 and Fig. 3a). Initially, the pristine NCM111 particle preserved a smooth surface. In contrast, after the EDO lithium leaching in stage Ⅰ, noticeable cracks appeared on the particle surface, which are related to changes in the material’s structure. Following the soaking relaxation leaching in stage Ⅱ, the cracks on the particle surface increased further, and the overall crystal split into a thin nanosheet morphology. Furthermore, the crystallographic structural evolution of the NCM111 during stages I and II was systematically studied by in-situ X-ray diffraction (XRD) with a homemade in-situ electrochemical powder XRD cell (Supplementary Fig. 5). As shown in Fig. 3b, at stage I, with the EDO proceeding, the diffraction peak of the pristine O3 phase (003) at two-theta of 18.7° began to split into two peaks, with the new diffraction peak at two theta of 18.3° corresponding to the monoclinic O'3 phase (001)^22^, indicating a decrease in crystal symmetry due to lattice distortion under EDO process. Subsequently, a P2 phase (002) diffraction peak appeared at 15.7° with EDO proceeding. The intensity of the O'3 phase peak initially increased and then decreased, disappearing simultaneously with the O3 phase diffraction peak, indicating the complete transformation of the O3 phase into P2 phase. This transformation was attributed to the leaching of lithium. After 20 minutes, an O2 phase with (002) diffraction peak appeared at 17.7° and continued to intensify, while the P2 phase diffraction peaks weakened. By the end of stage I, the material was fully transformed into the O2 phase. The transition from P2 to O2 phase could be attributed to the sliding of adjacent TMO_2_ (TM = Ni, Co, Mn) layers in regions with low Li content, which balanced O-O repulsion and Li-O attraction, leading to the phase transition^23^. To explore the relationship between the alkali metal content in the electrode material and phase transitions, we employed inductively coupled plasma mass spectrometry (ICP-MS) to analyze the elemental composition of the electrode material during the EDO process. Our results confirmed that the leaching efficiency of transition metals is extremely low (Fig. 2c), indicating that the concentration changes in these metals can be neglected. Changes in the composition of alkali metals (Li and Na) can be expressed by the atomic ratio between alkali metals (Li, Na) and the summation of transition metals (Ni, Co, Mn) within the solid-state electrode material. As shown in Supplementary Fig. 6a, during the first 20 min of the EDO process, the Li content in the electrode material decreased, and the O3 phase electrode material transformed into P2 phase. After 20 min, the lithium content decreased, while the sodium content increased in the electrode material due to the insertion of Na^+^. The transition from the P2 to O2 phase began, and this process was completed by the end of stage I as Na^+^ continued to be inserted. These results suggest that during the period of 0–20 min, EDO-driven lithium leaching is dominant, where O3-phase Li_0.7_Ni_1/3_Co_1/3_Mn_1/3_O_2_ transforms into P2-phase Na_0.01_Li_0.47_Ni_1/3_Co_1/3_Mn_1/3_O_2_; during 20–55 min, both EDO leaching and electrochemical driven Li^+^/Na^+^ exchange occurred^24^, yielding O2-phase Na_0.16_Li_0.17_Ni_1/3_Co_1/3_Mn_1/3_O_2_. In stage II, in-situ powder XRD (Fig. 3c) showed diffraction peaks shifting to higher angles. Figure 3d magnifies the (002) peak of O2-Na_0.16_Li_0.17_Ni_1/3_Co_1/3_Mn_1/3_O_2_, showing rapid shifts in the first 30 min, followed by a slowing. This indicated a reduction in interlayer spacing, consistent with the HRTEM images (Supplementary Fig. 7), where the spacing decreased from 5.05 to 4.97 Å. Additionally, aberration-corrected scanning transmission electron microscopy with energy-dispersive X-ray spectroscopy (STEM-EDS) (Supplementary Fig. 8) confirmed increased Na content during soaking, consistent with ICP-MS observation (Supplementary Fig. 6b). As soaking relaxation proceeded, the contents of Li and Na exhibited opposite tendencies, and the material transformed from O2-Na_0.16_Li_0.17_Ni_1/3_Co_1/3_Mn_1/3_O_2_ to O2-Na_0.31_Ni_1/3_Co_1/3_Mn_1/3_O_2_. Notably, the decrease in the Li/(Ni+Co+Mn) atomic ratio was almost equal to the increase in the Na/(Ni+Co+Mn) atomic ratio, suggesting that ion exchange dominates lithium leaching during stage II. The minor difference in Na and Li content changes could result from oxidation-induced lithium leaching in the electrolyte, driven by residual active oxidation species, including free radicals and free chlorine. These active oxidation species are generated during stage I through electrocatalytic oxidation of the electrolyte. The powder XRD diffraction results of pristine NCM111 (Supplementary Fig. 9) indicate a distinct α-NaFeO_2_ structure with a space group of R-3m, where Li occupies the Wyckoff 3a site of the O3 phase structure. To obtain the accurate atomicity of the evolution structures, Rietveld refinement has been employed for the refinement of crystallographic parameters on P2-Na_0.01_Li_0.47_Ni_1/3_Co_1/3_Mn_1/3_O_2_ and O2-Na_0.16_Li_0.47_Ni_1/3_Co_1/3_Mn_1/3_O_2._ As presented in Supplementary Tables 4–5 and Fig. 3e, f, in the P2 phase structure, Li occupies the Wyckoff 2b and Wyckoff 2d site, while in O2 phase, Li occupies the Wyckoff 3a site. The schematic of the crystal structure evolution of NCM111 during the two-stage continuous oxidation lithium leaching process could be summarized in Fig. 3g. This analysis reveals that the intricate crystal structure has evolved from the initial O3 phase (LiNi_1/3_Co_1/3_Mn_1/3_O_2_) to the P2 phase (Na_0.01_Li_0.47_Ni_1/3_Co_1/3_Mn_1/3_O_2_) and ultimately stabilized in the O2 phase (Na_0.31_Ni_1/3_Co_1/3_Mn_1/3_O_2_), where Na cations occupy the interlayer spacing. These observations suggest that ion exchange is the driving force behind structural evolution, while the subsequent structural transitions facilitate efficient lithium leaching. The ion exchange dominance during stage II ultimately maximizes lithium leaching and reduces energy consumption during the leaching process. These findings contribute to a deeper understanding of the structure-performance relationship, offering critical insights for optimizing the recycling process of spent NCM111 batteries.Fig. 3. Structural evolution of NCM111 during lithium leaching.a High magnification ex-situ SEM images. b In-situ XRD during stage I. c In-situ XRD during stage II. d Magnified view of the evolution of (002) XRD diffraction peak during stage II. e, f Rietveld refinement for the P2-Na_0.01_Li_0.47_Ni_1/3_Co_1/3_Mn_1/3_O_2_ and O2-Na_0.16_Li_0.17_Ni_1/3_Co_1/3_Mn_1/3_O_2_. g Overall crystal structural evolution during the two-stage continuous oxidation lithium leaching process. Source data for the figure are provided as a Source data file.

Active oxidation species evolution and alkaline ion effects within the electrolyte during the two-stage continuous oxidation lithium leaching process

Previous research suggests that the active oxidation species within the electrolyte also play a critical role in the lithium leaching process^17^. To probe the presence of active oxidation species in the electrolyte, we employed electron paramagnetic resonance (EPR) spectroscopy and the N, N-diethyl-1,4-phenylenediamine (DPD) method^25–28^ for detecting active oxidation species. As shown in Fig. 4a, •OH and ClO• radicals were detected in stage I, but not in stage II, indicating that these radicals are generated exclusively during the EDO process and are rapidly consumed. The free chlorine content (Fig. 4b, c) increased during stage I, but decreased to zero after 3 h of soaking, confirming its consumption. To further investigate the role of free chlorine in the lithium leaching process during stage II, the soaking solution was replaced with fresh 0.5 M NaCl after stage I was completed. As shown in Fig. 4d, after replacing the soaking solution, the lithium leaching efficiency after 3 h dropped to 98.32%, ~1.55% lower than without replacement (99.87%), indicating that free chlorine contributed modestly to stage II leaching. Subsequently, we explored the effects of metal cations and concentration on lithium leaching. As shown in Fig. 4e, deionized water, fresh 0.5 M NaCl, and 0.5 M KCl were used as soaking solutions in stage II experiments. Lithium leaching did not occur in deionized water, whereas the KCl solution yielded higher leaching rates than the NaCl solution. This may account for a weaker hydration shell in K^+^ compared to Na^+^, making the desolvation process during ion exchange much easier^29^. Figure 4f illustrates the effect of varying Na^+^ concentrations on lithium leaching efficiency. Increased Na^+^ concentrations accelerated the lithium leaching rate, as the concentration gradient of the reactive species governs the ion exchange process. This behavior is consistent with the principles of diffusion-controlled ion exchange kinetics, where the concentration of available cations directly influences the exchange rate^30^.Fig. 4. Active oxidation species evolution and alkaline ion effects within electrolyte.a EPR spectra. b, c Changes of free chlorine content during stage I and stage II. d Lithium leaching in fresh 0.5 M NaCl solution. e Effects of different cations. f Effect of Na^+^ concentration. The error bars in (d–f) are presented as the mean ± standard deviation of three independent experiments. Source data for the figure are provided as a Source data file.

Role of lattice oxygen and high valence transition metal during the two-stage continuous oxidation lithium leaching process

To further investigate the complex leaching mechanisms and major driving forces involved in the two-stage continuous oxidation lithium leaching process, we employed ex-situ electron energy loss spectroscopy (EELS) and X-ray photoelectron spectroscopy (XPS) to analyze the oxidation states of various elements in the electrode. The EELS analysis results are summarized in Fig. 5a–c and Supplementary Fig. 10. Throughout the entire two-stage continuous oxidation lithium leaching process, the L_3_ and L_2_ peaks of Ni blue shifted, with peak intensity ratio decreased from 3.33 to 2.46, confirming continuous Ni oxidation (Fig. 5a). Co, however, exhibited a more complex behavior (Fig. 5b). The L_3_ and L_2_ peaks showed significant blue shifts in Stage I but then red shifts in Stage II. When viewed alongside the Co XPS data (Fig. 5e), this suggests that Co undergoes an initial reduction followed by subsequent oxidation. The initial reduction is linked to the oxidation of lattice oxygen by high-valence Co under the high voltage of Stage I^31,32^. Notably, the L_3_/L_2_ ratio did not follow a simple trend (decreasing from 2.45 to 2.21, then increasing to 2.33). This complex pattern is likely due to the simultaneous occurrence of sequence redox reaction and local structure evolution for Co, as has been reported in battery research community by Jiang^33^ and Guo^34^. Finally, Mn remained highly stable. Its L_3_ and L_2_ peak positions and L_3_/L_2_ ratio barely changed throughout the process (Fig. 5c). This stability can be attributed to the role of Mn in maintaining the structural integrity of the material, with minimal contribution to the charge storage^35^. Ex-situ XPS double confirms the valance state evolution (Fig. 5d–f). Scanning Kelvin probe (SKP) mapping (Supplementary Fig. 11) reveals a significantly more negative surface potential after soaking, indicating a notable change in the surface chemical potential/corrosion behavior at stage II. These SKP observations are consistent with the oxidation state changes inferred from EELS and XPS. It is worth noting that the O 1 s XPS spectra (Fig. 5g) exhibit significant differences after stages I and II. In pristine NCM111, the peak at 529.3 eV corresponds to lattice oxygen (O^2-^), while the peak at 530.9 eV corresponds to ionic OH^-^^36^, which may be due to minor residual LiOH during the synthesis of the pristine NCM111. After the EDO in stage I, the peak at 530.9 eV disappears, indicating that LiOH has been redissolved within the electrolyte at stage Ⅰ. Additionally, two new peaks appear at 530.5 eV and 531.3 eV, corresponding to oxidized lattice oxygen (O^n-^, n < 2) and oxygenated deposited species, respectively^37^. After stage II, O^n-^ disappears, indicating that O^n-^ was reduced during the soaking relaxation process, and the electron might be transferred from Ni^2+^ and Co^2+^ to O^n-^ via chemical reactions of Ni^2+^+O^n-^→Ni^3+^+O^2-^ and Co^2+^+O^n-^→Co^3+^+O^2-^^37,38^, which is also consistent with the elevated valence states of Ni and Co. The O K-edge EELS (Fig. 5h) revealed that the pre-peak initially shifted to lower energy loss and then to higher energy loss during lithium leaching, accompanied by a corresponding increase and subsequent decrease in its relative intensity to the main peak, further supporting a sequence of lattice-oxygen (O^2-^) oxidation in stage I followed by reduction in stage II. Raman spectroscopy (Fig. 5i) reveals key bond changes. In stage I, the Ni-O A1g mode blue-shifts by 29.4 cm^-1^, confirming O^2-^ oxidation (→O^n-^), which elongates the Ni-O bond by electron loss (dominant over Ni oxidation effects)^11^. In stage II, the peak red-shifts by 32.5 cm^-1^ as O^n-^ oxidizes Ni, shortening the Ni-O bond via enhanced Ni-O electrostatic attraction. To further verify that O^n-^ drives the lithium leaching reaction in stage II, an extra experiment was conducted by introducing the reducing agent KI into the soaking solution. No lithium leaching occurred after adding KI (Supplementary Fig. 12a), indicating that the strong reducing power of I^-^ eliminated the reaction’s driving force. XRD analysis confirmed that soaking in the KI solution did not alter the electrode crystal structure (Supplementary Fig. 12b). The solution turned yellow due to the oxidation of I^-^ to I_2_ (Supplementary Fig. 12c). Furthermore, when KI was added after the completion of stage II, no significant color change was observed (Supplementary Fig. 12d), confirming that O^n-^, rather than high-valent Ni, Co, or Mn, was responsible for the oxidation of I^-^. These findings further validate that O^n-^ plays a key role in driving lithium leaching during stage II.Fig. 5. Chemical and electronic state evolution of Ni, Co, Mn, and O during two-stage continuous oxidation lithium leaching process.a–c Ex-situ electron energy loss spectra of Ni, Co, and Mn. d–f Ex-situ X-ray photoelectron spectra of Ni 2p, Co 2p, and Mn 2p. g XPS spectra of O 1s. h O K-edge of EELS spectra. i Raman spectra. Arrows in the EELS spectra indicate spectral changes: left/right arrows denote peak shifts to lower/higher energy loss, respectively; upward/downward arrows denote an increase/decrease in peak intensity. Source data for the figure are provided as a Source data file.

Quantitative contributions of stage-dependent lithium leaching mechanisms

Overall, the lithium leaching mechanism can be summarized as follows: during stage Ⅰ, lithium leaching is driven by (i) direct electrode oxidation (Fig. 6a(I)); (ii) electrocatalytic oxidation (Fig. 6a(II)); and (iii) electrochemical-driven ion exchange (Fig. 6a(Ⅲ)); during stage Ⅱ, lithium leaching proceeds via (iv) O^n-^-driven ion exchange (Fig. 6a(IV))) free chlorine oxidation (Fig. 6a(V)) with the calculation details for the contribution of each driving forces provided in Supplementary Note 5. The quantified contributions of these mechanisms are illustrated in Fig. 6b. Specifically, electrochemical dual oxidation, which consists of both direct electrode oxidation and electrocatalytic oxidation, accounts for 66.92% of the total lithium leaching efficiency. In comparison, electrochemical-driven ion exchange contributes 15.97%. In stage Ⅱ, O^n-^-driven ion exchange accounts for 15.43%, and the residual free chlorine oxidation contributes 1.55%. These quantitative insights elucidate the critical roles of distinct oxidative and ion exchange pathways across the two-stage continuous oxidation lithium leaching process.Fig. 6. Two-stage continuous oxidation lithium leaching mechanism and universality of the method.a Mechanistic diagram of two-stage continuous oxidation lithium leaching. b Contribution of different driving forces to overall lithium leaching. c–e Lithium leaching efficiency of NCM523, NCM622, and NCM811. The error bars in (c–e) are presented as the mean ± standard deviation of three independent experiments. Source data for the figure are provided as a Source data file.

Universality of the two-stage continuous oxidation lithium leaching strategy

To validate the universality and practical applicability of this two-stage strategy beyond model materials, leaching experiments were performed on spent NCM111 cathodes from practical spent lithium-ion batteries (Supplementary Fig. 13a–d). After 70 min of EDO followed by a 2.5-h soaking relaxation, the lithium leaching efficiency reached 98.43%, equivalent to the 98.49% achieved after 160 min of EDO alone. However, the energy input ratio was significantly reduced from 273.12% to 142.59% (Supplementary Fig. 14a, b), demonstrating that the two-stage process can effectively lower electric energy consumption while maintaining high lithium leaching efficiency. Supplementary Fig. 14c shows that the leaching efficiencies of Ni, Co, and Mn remained below 0.2%, confirming that the method retains high lithium selectivity when processing spent materials. To further establish the generality of the proposed method, leaching experiments were performed on various spent NCM cathodes, including LiNi_0.5_Co_0.2_Mn_0.3_O_2_ (NCM523), LiNi_0.6_Co_0.2_Mn_0.2_O_2_ (NCM622), and LiNi_0.8_Co_0.1_Mn_0.1_O_2_ (NCM811). After 70-min of EDO (stage I), lithium leaching efficiencies of 84.28%, 86.47%, and 90.05% were achieved, respectively. A subsequent 2.5-h soaking relaxation (stage II) further enhanced the efficiencies to 98.08%, 98.34%, and 98.57% (Fig. 6c–e). These results underscore the method’s universality across diverse NCM cathodes and its promise for sustainable lithium recovery from spent lithium-ion batteries.

Techno-economic evaluation

To demonstrate the scalability of the two-stage continuous oxidation lithium leaching method, a customized equipment with a single-batch processing capacity of 500 g was constructed (Fig. 7a, b, Supplementary Fig. 15a). In each electrolytic cell, cost-effective titanium plates were employed as the current collectors coated with active materials. Lithium was leached from spent NCM cathode materials and recovered via carbonate precipitation. A high lithium leaching efficiency of 98.12% was achieved after 70 min of EDO (stage I), followed by a 4-h soaking relaxation (stage II), as shown in Fig. 7c. The leaching efficiency surpasses most reported electrochemical recovery methods (Supplementary Table 6). The obtained Li_2_CO_3_ product exhibited a purity exceeding 99.5%, meeting the standards for battery-grade lithium carbonate (Supplementary Fig. 15b and Supplementary Table 7). A detailed comparative analysis of energy consumption and environmental impacts are presented in Fig. 7d, with the related details documented in Supplementary Notes 6–7. Compared to the conventional EDO method, the two-stage continuous oxidation lithium leaching method reduces energy consumption by 49.78%, resulting in a savings of $142.84 in electricity costs per ton of spent NCM111 processed. In addition, we carefully examined all relevant cost and revenue streams (Supplementary Data 1 and Supplementary Note 8). The analysis shows that recovering lithium from one ton of spent NCM111 cathode material can yield a gross profit of$643.67, with electricity cost savings accounting for 22.19% of the total gross profit. Given the substantial contribution of electricity costs to large-scale recycling operations, the two-stage continuous oxidation lithium leaching method can effectively reduce operational expenses.Fig. 7. Scalability validation and techno-economic evaluation of the two-stage continuous oxidation lithium leaching system.a General procedure for lithium leaching and recovery. b Schematic diagram of the continuous-flow lithium recovery system. c Pilot-scale lithium leaching efficiency under optimal energy input condition. d Evaluation of energy consumption and environmental-economic impacts of different methods. The error bars in (c) are presented as the mean ± standard deviation of three independent experiments. Source data for the figure are provided as a Source data file.

Discussion

The two-stage continuous oxidation lithium leaching method proposed in this study demonstrates significant efficiency advantages in electric energy utilization over conventional electrochemical methods. By precisely controlling the duration of EDO, we achieved a lithium leaching efficiency of 98.12% from spent LiNi_1/3_Co_1/3_Mn_1/3_O_2_, with a 49.78% reduction in electric energy consumption. Mechanistic studies reveal that the two stages synergistically enhance lithium leaching through chemical coupling and energy complementarity: In stage I, EDO not only drives lithium but also induces the transformation of lattice oxygen (O^2-^) into oxidized lattice oxygen (O^n-^, n < 2), thereby creating a driving force for subsequent lithium leaching. Additionally, we also found the occurrence of electrochemical-driven ion exchange. In stage II, O^n-^ promotes ion exchange between electrolyte cations and lithium in the material, enabling further lithium leaching. During this process, O^n-^ is reduced along with lithium leaching, while nickel and cobalt maintain charge balance through oxidation reactions. Notably, the type of electrolyte cations significantly influences leaching efficiency, with K^+^ exhibiting markedly superior facilitation to Na^+^. The mechanism proposed in this study broadens and deepens the understanding of the lithium leaching process, providing critical theoretical foundations and technological support for developing environmentally friendly and energy-efficient recycling systems for spent ternary lithium-ion batteries.

Methods

Chemicals and materials

All chemicals were used as received without further purification. Sodium chloride (NaCl, Macklin, AR, 99.5%), potassium chloride (KCl, Aladdin, ≥99.5%), potassium iodide (KI, Macklin, ≥99.0%), N, N-Diethyl-1,4-phenylenediamine sulfate (Rhawn, 98%), N-methyl-2-pyrrolidone (NMP, Macklin, 98%), polyvinylidene difluoride (PVDF, Aladdin), and carbon cloth (CeRech Co., Ltd., Wos1009) were directly utilized without additional purification. The commercial cathode material LiNi_1/3_Co_1/3_Mn_1/3_O_2_ (NCM111) (Supplementary Table 1) used in this study was supplied by Ningbo Jinhe Lithium Battery Material Co., Ltd. The different composition spent cathode materials, LiNi_0.5_Co_0.2_Mn_0.3_O_2_ (NCM523), LiNi_0.6_Co_0.2_Mn_0.2_O_2_ (NCM622), and LiNi_0.8_Co_0.8_Mn_0.8_O_2_ (NCM811) (Supplementary Table 2) were provided by Dongguan Pu Dehua Electronic Materials Co., Ltd.

Electrode fabrication

Before the two-stage continuous oxidation lithium leaching experiment, NCM111 electrodes for lithium leaching studies were prepared as shown in Supplementary Fig. 1. Initially, the NCM111 cathode material was mixed with a PVDF binder in a mass ratio of 19:1. An appropriate amount of NMP was added. The mixture was stirred to form a uniform slurry. This slurry was then evenly coated on both sides of a carbon cloth (2 cm × 2.5 cm). Subsequently, the coated carbon cloth was dried in a vacuum oven at 80 °C for 10 h to obtain the electrode sheets. The effective coating area on each side of the resulting electrode sheet was approximately 4 cm^2^, with a mass loading of about 15 mg cm^-2^ on each side.

Two-stage continuous oxidation lithium leaching experiments

Supplementary Fig. 2 illustrates a schematic diagram of the two-stage continuous oxidation lithium leaching system. The prepared NCM111 electrode was used as the anode in the experiments. An identical-sized carbon cloth was used as the cathode. A 100 mL 0.5 mol L^-1^ NaCl solution served as the electrolyte to construct the electrolysis cell. First, we established a quantitative relationship between lithium leaching efficiency and EDO time. By conducting EDO experiments using an electrochemical workstation, we could control the time and record current changes to investigate the effects of different periods on lithium leaching efficiency. Subsequently, we studied the two-stage continuous oxidation lithium leaching process. In stage I, the EDO time was controlled at 50 min, 55 min, and 60 min, respectively, followed by an exploration of the subsequent effects of the soaking relaxation process in stage II to further improve the lithium leaching efficiency. This experimental design enables us to assess the impact of varying EDO time on lithium leaching efficiency in the two-stage continuous oxidation lithium leaching process, aiming to reduce the EDO time to optimize electric energy consumption while achieving high lithium leaching efficiency.

In stage I, the EDO lithium leaching was carried out under magnetic stirring at a rotation speed of 500 rpm^9^, while a constant voltage of 2.5 V^11^ was applied. The lithium leaching process in stage II was conducted without any energy input. The metal leaching efficiency was calculated via Eq. (4). The practical electric energy input was calculated via Eq. (5). A detailed calculation of theoretical electric energy is provided in Supplementary Note 1. In addition, the energy input ratio was introduced to assess the efficiency of electric energy utilization during the lithium leaching process. Here, the energy input ratio is expressed as practical electric input energy divided by the required theoretical electric energy.

\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${R}_{i}=\frac{{C}_{i}\times V}{m\times {\omega }_{i}}\times 100\%$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$W=Q\times U$$\end{document}

where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${C}_{i}$$\end{document} is the mass concentration of metal elements in the leachate, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$V$$\end{document} is the volume of the leachate (100 mL), \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$m$$\end{document} is the mass of the active material (120 mg), and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\omega }_{i}$$\end{document} is the mass fraction of metal elements in the active material. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$U$$\end{document} is the applied voltage (2.5 V), and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Q$$\end{document} is the charge quantity, with detailed calculations for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$Q$$\end{document} provided in Supplementary Note 2.

Material characterization

Powder XRD patterns were collected using a 9 kW Rigaku SmartLab machine equipped with Cu Kα radiation (λ = 1.5406 Å). The scanning speed and step size were controlled to be 10° min^-1^ and 0.02° per step, respectively. The collected XRD data were analyzed using the Rietveld refinement program GSAS^39^. The refined parameters included the unit cell parameters, zero shift, sample displacement, peak shape parameters, microstrain, preferred orientation, and atomic positions. Once a parameter converged, it was fixed during the refinement process. SEM images of the electrodes were obtained using a Zeiss Merlin microscope. Low magnification high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) images and the element analysis were obtained by an FEI Talos F200X transmission electron microscopy (TEM) equipped with an energy-dispersive X-ray spectrometer (EDS) operating at 200 kV. Electron energy-loss spectroscopy (EELS) analysis was collected on an FEI Titan Themis G2 double-aberration corrected TEM with an X-FEG electron gun and a Gatan GIF Quantum spectrometer operating at 300 kV. A laser micro confocal Raman spectrometer (LabRAM HR Evolution) recorded the Raman spectra under 532 nm excitation. The voltage potential maps were recorded using a scanning kelvin probe (SKP, VersaSCAN). The scanned area was a square (1.5 mm × 1.5 mm), with a step size of 0.1 mm in both the X and Y directions. X-ray photoelectron spectroscopy (XPS) was recorded on a PHI 5000 Versaprobe III instrument (ULVAC-PHI, UK) with a monochromatic Al Kα source. All the spectra were calibrated, referring to the C 1s binding energy of 284.8 eV. The elemental concentration analysis of the leachates was performed using an inductively coupled plasma mass spectrometer (ICP-MS) (Agilent, A7900QMS). The concentrations of free chlorine (including Cl_2_, HClO, and ClO^-^) produced in the reaction were monitored using the N, N-diethyl-1,4-phenylenediamine (DPD) method^25–28^.

Supplementary information

Supplementary Information Description of Additional Supplementary Files Supplementary Data 1 Transparent Peer Review file

Source data

Source data