Selective and Closed-Loop Recycling of Different Metals from Spent Lithium-Ion Batteries Through Phosphoric Acid Leaching: Parameter Optimization and Regulation of Reaction Kinetics

Abstract

The sustainable recycling of valuable metals from spent lithium-ion batteries (LIBs) is critical for resource conservation and environmental protection but remains challenging due to the complex coexistence of target and impurity metals. This study systematically investigates the selective leaching behaviors of metals (Co, Li, Cu, Fe, Al) in phosphoric acid media, revealing that lithium could be preferentially extracted in mild acidic conditions (0.8 mol/L H₃PO₄), while complete dissolution of both Li and Co was achieved in concentrated acid (2.0 mol/L H₃PO₄). Kinetic analysis demonstrated that metal leaching followed a chemically controlled mechanism, with distinct extraction sequences: Li > Cu~Co > Fe > Al in dilute acid and Cu > Al~Li > Fe > Co in concentrated acid. Furthermore, we developed a closed-loop process wherein oxalic acid simultaneously precipitates Co/Li while regenerating H₃PO₄, enabling acid reuse with minimal efficiency loss during cyclic leaching. These findings establish a single-step phosphoric acid leaching strategy for selective metal recovery, governed by tunable acid concentration and reaction kinetics, offering a sustainable pathway for LIBs recycling.

1. Introduction

In the 1990s, Sony corporation pioneered and developed the first commercial lithium-ion batteries (LIBs) with potential widespread application in mobile devices [1], electronic equipment, and electric/hybrid electric vehicles (EVs/HEVs) because of their preferable electrochemical performances in terms of high operating voltage and specific capacity, modest size and weight, long service life, and high energy density [2,3]. However, a huge quantity of spent LIBs will inevitably flow into waste stream with the increasing numbers of end-of-life batteries generated annually [4,5,6]. Predictions indicate that by 2027, the global market size for spent LIBs recycling will reach 11.07 billion US dollars [7]. In addition, potential risks towards ecosystem and public health may be caused by heavy metals or organic hazardous wastes contained in spent LIBs [8]. Thus, it is necessary to recycle them LIBs an environmentally sound way. Generally, a typical LIB is made up of cathode, anode, metallic shell, separator, electrolyte, and other appurtenances. The most valuable part is the cathode [5], which contains a large proportion of valuable metals like cobalt (Co), lithium (Li), copper (Cu), aluminum (Al), and iron (Fe). Therefore, sustainable and effective recovery of value-added metals from spent LIBs will be of great importance in terms of preserving metal resources and environmental protection [9].

Currently, attention and effort are put into the recovery of valuable metals from spent LIBs using pyro-metallurgical [10,11,12] and hydro-metallurgical [13,14,15] techniques, also including bio-leaching and mechanical methods [16,17,18,19]. Chen et al. [20] developed an effective pyrometallurgical process which can recover cobalt, nickel, copper (99%), and concentrating lithium (Li₂O 7.28 wt.%) from high-manganese artificial β-spodumene slags. Though highly efficient, pyro-metallurgical methods may be discouraged by high additional investments in terms of alloy smelting at high temperature and the disposal of dangerous gases [8]. Harshit Mahandra et al. [19] used Thiobacillus thioparus to selectively leach lithium at near-neutral pH, which was able to recover 65–98% lithium with insignificant iron. Although the bio-hydrometallurgical method involves a mild processing environment, it can be frustrated by the extremely low leaching efficiency required compared to chemical-based leaching, due to the long leaching period [8]. Qi et al. [10] proposed a process based on phosphate chemistry for recovering valuable metals from spent LIBs. Under optimized conditions (0.7 mol/L H₃PO₄, 7 vol.% of H₂O₂, residence time of 40 min, reaction temperature of 40 °C, and liquid–solid ratio of 30 mL/g), the lithium leaching rate can reach 98.7%. As a traditional method for effectively separating and recovering valuable metals from spent LIBs, although it involves the use of acid and brings certain processing burdens, the hydro-metallurgical technique is a popular choice due to its lower cost, less secondary pollution, and higher recovery efficiency compared with pyro-metallurgical and bio-leaching [21,22].

The literature has reported that sulfuric acid leaching liquor can be used for the hydro-metallurgical method [23].

Table 1. A brief summary of different leaching systems.

Type of Acid	Leaching Reagent	Leaching Condition	Leaching Rate	Ref.
Mineral acid	Sulfuric acid (1 mol/L)	95 °C, 20 g/L, 240 min	96% Li, 91% Co, 96% Ni, and 88% Mn	[28]
	Nitric acid (1 mol/L)	75 °C, 20 g/L, 60 min	95% Li and 95% Co	[29]
	Hydrochloric acid (6 mol/L)	60 °C, 8 g/L, 120 min	95% Ni, Co, and Mn	[30]
	Phosphoric acid (0.7 mol/L)	40 °C, 50 g/L, 60 min	99% Co and Li	[31]
Organic acid	Citric acid (1.5 mol/L)	80 °C, 20 g/L, 120 min	99% Li, 92% Co, 91% Ni, and 94% Mn	[32]
	Ascorbic acid (1.25 mol/L)	70 °C, 25 g/L, 20 min	98% Li and 94% Co	[33]
	Malic acid (1.5 mol/L)	90 °C, 20 g/L, 40 min	100% Li and 90% Co	[34]
	Succinic acid (1.5 mol/L)	70 °C, 15 g/L, 40 min	96% Li and 100% Co	[35]
	Oxalic acid (1 mol/L)	95 °C, 15 g/L, 150 min	98% Li and 97% Co	[36]
	Aspartic acid (1.5 mol/L)	90 °C, 20 g/L,120 min	60% Li and 60% Co	[37]
	Glycine (0.5 mol/L)	80 °C, 20 g/L, 360 min	95% Co	[38]
	L-Tartaric acid (2 mol/L)	70 °C, 17 g/L, 30 min	99% Mn, 99% Li, 98% Co, and 99% Ni	[39]

summarizes different leaching systems for recovering valuable metals from spent LIBs using different acids. It can be concluded that both mineral and organic acids may demonstrate similar leaching performance [24]. Nevertheless, the mineral acids adopted may result in severe corrosion of equipment and secondary noxious gases like SO₂ and Cl₂ [25], and organic acids will inevitably enhance the costs during industrialized application. Furthermore, the separation and recovery of valuable metals from leaching solutions of mineral and organic acids may be challenging due to the tedious metal separation processes, and only relatively low leaching efficiencies can be achieved in some organic acid leaching systems, like 60% for Co and Li using aspartic acid [26]. In that case, for the selective recovery of different metals from waste LIBs, the hydro-metallurgical method is widely used in industry [27]. Therefore, it is necessary to simulate the acid leaching process of spent LIBs in the actual production of multi-metal ion systems.

Phosphoric acid, as a weak acid, is milder than strong acids such as sulfuric acid and hydrochloric acid in terms of both environmental pollution and the corrosion resistance of reactor materials. In our previous work, low concentration of phosphoric acid (0.7 mol/L) has been innovatively used as both leaching and precipitating reagent under mild leaching conditions, and about 99% Co and Li can be separated and recovered in a single leaching step [31]. Moreover, it can be efficiently recycled [40]. However, lack of clarity persists with regard to the leaching behaviors of valuable metals and impurity metals from spent LIBs under different phosphoric acid media (low and high concentrations), such as leaching kinetics and the leaching order of different metals. The present work was designed to uncover detailed leaching and recovery phenomena and their essence from both experimental and theoretical aspects, using different concentrations of phosphoric acid as leaching reagent. It is expected that a sustainable and efficient recovery route can be established based on the leaching behaviors of different metals and provide a technical and theoretical foundation for metal recovery in different acidic media.

2. Experimental Section

2.1. Materials and Reagents

Spent cathode materials (waste LiCoO₂, LCO) obtained after pretreatment through discharging, dismantling, peeling off Al foil, roasting, and grinding were used as feed materials. The waste LiCoO₂ powders were then completely dissolved in concentrated hydrochloric acid (10 mol/L HCl). Inductively coupled plasma–optical emission spectroscopy (ICP-OES)/inductively coupled plasma–mass spectrometry (ICP-MS) was used to determine the metal elements in the sample, in which the mass contents of valuable Co and Li metals was 58.74% and 6.02%, respectively, and the mass content for impure Cu, Al, and Fe metals was 0.028%, 2.026%, and 0.031%, respectively. The main chemical reagents used in this study included concentrated phosphoric acid (H₃PO₄, 85%, AR, Tianli Chemical Reagent Co., Ltd., Tianjin, China), hydrogen peroxide (H₂O₂, 30 v/v%, AR, Tianli Chemical Reagent Co., Ltd.), sodium sulfate (Na₂SO₄·10H₂O, 99%, AR, Tianli Chemical Reagent Co., Ltd.), ammonium hydroxide (NH₃·H₂O, 25–28%, AR, Tianli Chemical Reagent Co., Ltd.), oxalic acid (H₂C₂O₄, AR, Sinopharm Group Chemical Reagent Co., Ltd., Shanghai, China), sodium dihydrogen phosphate (NaH₂PO₄, 99% purity, AR, Fuchen Chemical Reagent Co., Ltd., Tianjin, China), cobaltous sulfate (CoSO₄, 99% purity, AR, Fuchen Chemical Reagent Co., Ltd.), N-methyl-2-pyrrolidone (NMP, 99% purity, AR, Tianjin Kemio Chemical Reagent Co., Ltd., Tianjin, China), and aluminum powders, iron powders, and copper powders (Al, Fe, and Cu, 99% purity, Aladdin Reagents Co., Ltd., Shanghai, China). All other reagents used were of analytical grade and deionized water was used for the preparation and dilution of different solutions during the whole experiment.

2.2. Leaching Experiments

All of the leaching experiments were conducted in a three-necked and round-bottom flask, which served as a reactor. The specific experimental steps are listed as follows. (a) A precisely weighed quantity of waste LiCoO₂ powder was introduced into the reactor. A leaching solution containing a specific concentration of H₃PO₄ was subsequently added, while H₂O₂ was introduced dropwise. The leaching reaction was then conducted for a predetermined duration. (b) Upon the reaction’s completion, vacuum filtration was employed to achieve solid–liquid phase separation. (c) The filtered residues were then dried and ground for material characterization, while the filtrate was diluted for subsequent ICP-OES analysis. In addition, the filtrate obtained in the circulating leaching test without dilution was reused as leaching reagent for the leaching of waste LiCoO₂ powders.

The reaction described above was a liquid–solid non-catalytic reaction. Previous studies have shown that solid particles in the liquid–solid non-catalytic reaction are unreacted cores, abbreviated as the “shrinking core model” [41]. Therefore, the shrinking-core model was adopted for the investigation of leaching kinetics mechanism for relevant leaching reactions in different acidic media with different reaction temperatures and retention times [42]. According to this model, when the leaching of Li, Co, Al, Cu, and Fe from spent LIBs is controlled by chemical reaction or diffusion-controlled process through inner layer diffusion, the leaching process can be described by Equations (1) and (2), respectively:

1 − (1 − X)^1/3 = k_ct

(1)

1 − 2/3X − (1 − X)^2/3 = k_dt

(2)

where X is the reaction fraction (i.e., leaching efficiency), k_c and k_d are the reaction rate constants (min⁻¹), and t is the leaching time (min).

In addition, the Arrhenius equation was applied for the determination of apparent activation energy (Ea) and pre-exponential factor (lnA) under fixed leaching conditions in different acidic media in a retention time range of 10–60 min and a temperature range of 30–70 °C. The Arrhenius equation can reveal the dependence of the rate of chemical reaction on the temperature, and it can be described as Equation (3) or Equation (4):

k = Ae^−Ea/RT

(3)

ln(k) = ln(A) − Ea/RT

(4)

where k and A are the rate constant and frequency factor of relevant reaction (mol/(L·h)), Ea is the apparent activation energy (kJ/mol) of corresponding reaction, R is the gas constant 8.314 J/(mol·K), and T is the temperature (K).

To explore the leaching kinetics mechanism, single-factor experiments were conducted focusing on residence time and acid concentration. In the cyclic leaching experiments, identical leaching conditions were applied as those used for the leaching of different metals. During the leaching process, 2 mL leachate was withdrawn every 10 min, with an equal 2 mL volume of fresh acid solution added to maintain the reaction conditions. The concentrations of Li, Co, Al, Cu, and Fe in all leachate samples were subsequently analyzed by ICP-OES or ICP/MS.

2.3. Analytical Methods

The concentrations of all metals in leaching solutions were determined by inductively coupled plasma–optical emission spectroscopy or inductively coupled plasma source–mass spectrometry (ICP-OES or ICP/MS, Therm, Waltham, MA, USA). The morphology of solid components before and after leaching can be characterized by environmental scanning electron microscopy (SEM, FEI Q45, Hillsboro, OR, USA), and surface element content was detected with an Energy Dispersive Spectrometer (EDS, EDAX Octane Prime, Mahwah, NJ, USA). In addition, an X-ray Diffractometer (XRD, D8 Advance, Bruker Company, Karlsruhe, Germany) was used for the identification of solid materials before and after leaching. Finally, Fourier-transform infrared spectrometry (FT-IR, VECTOR-22, Bruker Company, Germany) and X-ray photoelectron spectroscopy (XPS, Thermo) were used for characterization of waste LiCoO₂ powders, leaching product, and pure Co₃(PO₄)₂ powders. Detailed information of above characterization equipment can be found in the Supporting Materials (Table S1). Leaching efficiencies of different metals were calculated according to following equation:

$$y={\displaystyle \frac{m_l}{m_s}}\times 100\%$$

(5)

where y is leaching efficiencies of different metals, m_l is mass content of metal in leaching solution (g), m_s is the total metal content in waste LiCoO₂ powders or mixtures (g).

3. Results and Discussion

3.1. Selective Leaching of Valuable Metals

To investigate the kinetic mechanism under suitable reaction conditions, we first optimized the leaching conditions. Under a constant concentration of 0.8 mol/L phosphoric acid, the effects of H₂O₂ volume fraction, leaching temperature, leaching time, and solid-to-liquid ratio on the leaching efficiency were explored. As shown in Figure 1A, with fixed conditions of 20 mL/g pulp density, 60 min retention time and 60 °C, the volume fraction of H₂O₂ was varied in the range of 0–8 vol.%. The leaching efficiency of Li gradually increased within the range of 0–4 vol.% and remained nearly constant after adding 4 vol.% H₂O₂. However, the volume fraction of H₂O₂ had no significant effect on the leaching of Co. Based on these observations, the optimal H₂O₂ addition was determined to be 4 vol.%. The optimization of temperature, retention time, and pulp density are shown in Figure 1B–D, respectively. The final optimized leaching conditions were 4 vol.% H₂O₂ addition, a leaching temperature of 60 °C, a retention time of 60 min, and a solid-to-liquid ratio of 20 mL/g.

Relatively pure waste cathode materials (waste LiCoO₂) were obtained after the pretreatment procedure. Figure 2 shows that the leaching of Li and Co involved different leaching behaviors with the variation of phosphoric acid concentration (under the optimized leaching conditions of reductant dosage 4 vol.% H₂O₂, leaching temperature 60 °C, retention time 60 min, and pulp density 20 mL/g, see Table S2 in Supplementary Materials). Under mild phosphoric acid concentrations (i.e., 0.2–1.0 mol/L, Figure 2A), the leaching efficiency of Li underwent a continuous increase from 16% to 98%. Co could hardly be dissolved in mild phosphoric acidic solutions with the increase of acid concentration from 0.2 mol/L to 0.8 mol/L. Afterwards, the leaching efficiency of Li almost remained at the same level and the leaching efficiency of Co experienced a slight rise from 1% to 12% with the increase in acid concentration from 0.8 mol/L to 1.0 mol/L. This indicates that Co and Li can be effectively separated and recovered in a single leaching step under mild acidic medium.

The leaching behavior of Co and Li, however, indicates a quite different mode according to leaching results illustrated in Figure 2B. The leaching efficiency of Co underwent a steady enhancement from 12% to 99% and the leaching efficiency of Li remained unchanged with the increase in acid concentration from 1.0 mol/L to 2.0 mol/L, indicating that Co can be gradually dissolved in phosphoric acid solution under higher acid concentrations. Afterwards, the leaching efficiencies of Co and Li remained unchanged with further increases in acid concentration from 2.0 mol/L to 3.0 mol/L, which may be attributed to complete dissolving of waste LiCoO₂ under 2.0 mol/L phosphoric acid solution.

In conclusion, 0.8 mol/L and 2.0 mol/L were identified as optimized acid concentrations for the selective separation and recovery of valuable metals from waste LiCoO₂. It was also observed that the leaching behaviors of Co and Li were determined by acid concentration. Therefore, it is necessary to explore the reaction mechanism during the leaching process. Phosphoric acid (H₃PO₄) is a tribasic acid which can be ionized to H₂PO₄⁻, HPO₄²⁻, PO₄³⁻, and H⁺ with ionization constants of 2.12 (pKa₁), 7.21 (pKa₂), and 12.67 (pKa₃) (see Equations (6)–(8)) [43]. Salts of Li⁺/Co²⁺ with HPO₄²⁻/PO₄³⁻ present low solubility while salts of Li⁺/Co²⁺ with H₂PO₄⁻ can be dissolved in aqueous solution. Thus, the selective leaching of Li and Co may be attributed to characteristics of H₃PO₄ and chemical reactions under different concentrations of acidic media (Equations (9)–(11)). Leaching reactions of Li and Co can be controlled based on dosage and concentration of phosphoric acid. Waste LiCoO₂ can react with H₃PO₄ in the presence of H₂O₂, and Co₃(PO₄)₂ and Li₃PO₄ precipitates are initially generated with a low molar ratio of n(H₃PO₄):n(LiCoO₂) = 1:1 (0.52 mol/L H₃PO₄ required, Equation (9)). Then, Li₃PO₄ precipitate is converted to soluble LiH₂PO₄ with further addition of H₃PO₄ at a molar ratio of n(H₃PO₄):n(LiCoO₂) = 1.67:1 (0.86 mol/L H₃PO₄ required, Equation (10)). Finally, Co₃(PO₄)₂ precipitate is dissolved with a molar ratio of n(H₃PO₄):n(LiCoO₂) = 3:1 (1.56 mol/L H₃PO₄ required, Equation (11)). It can be concluded that the above theoretical results are consistent with the experimental results demonstrating that Li can be selectively extracted under low acid concentration of 0.8 mol/L, and the further promotion of acid concentration from 0.8 to 2.0 mol/L results in the further dissolution of Co₃(PO₄)₂ precipitate to soluble Co(H₂PO₄)₂.

H₃PO₄ (l) = H₂PO₄⁻ (aq) + H⁺ (aq)   pKa₁ = 2.12

(6)

H₂PO₄⁻ (aq) = HPO₄²⁻ (aq) + H⁺ (aq)   pKa₂ = 7.21

(7)

HPO₄²⁻ (aq) = PO₄³⁻ (aq) + H⁺ (aq)   pKa₃ = 12.67

(8)

LiCoO₂ (s) + H₃PO₄ (aq) + H₂O₂ (l) = 1/3Co₃(PO₄)₂ (s) + 1/3Li₃PO₄ (s) + 3/2H₂O (l) + 5/4O₂ (g)

(9)

Li₃PO₄ (s) + 2H₃PO₄ (aq) = 3LiH₂PO₄ (aq)

(10)

Co₃(PO₄)₂ (s) + 4H₃PO₄ (aq) = 3Co(H₂PO₄)₂

(11)

In order to further determine the leaching product, SEM-EDS was employed for the analysis of leaching product compared with pure Co₃(PO₄)₂ prepared (detailed preparation can be found in Supplementary Materials). Figure 3 shows that both leaching the product and pure Co₃(PO₄)₂ present similar morphologies to a regular sphere. In addition, the atomic percentages (Atomic%) of O, P, and Co in pure Co₃(PO₄)₂ are 73.36%, 11.39%, and 15.25%. This indicates that the atom ratio of Co and P is 15.25:11.39 (close to the stoichiometric ratio of 3:2), while the atom ratio of P and O is 11.39:73.36 (lower than the stoichiometric ratio of 1:4), which may be caused by bound or free water. However, the atomic percentages of O, P and Co in the leaching product are 60.56%, 18.15%, and 21.29%, which means that the atom ratio of Co and P is 21.29:18.15 (close to the stoichiometric ratio of 3:2) and the atom ratio of P and O is 18.15:60.56 (close to the stoichiometric ratio of 1:4). According to the above analytical results, it can preliminarily be stated that the leaching product obtained was relatively pure Co₃(PO₄)₂.

To further confirm the leaching products obtained after leaching reactions from different concentrations of phosphoric acid, XRD was employed for the exploration and analysis of crystal structure and chemical phase of waste LiCoO₂, leaching residues (2.0 mol/L H₃PO₄) and leaching product (0.8 mol/L H₃PO₄) (see Figure 4). It was discovered that intensive characteristic peaks around 003, 101, and 104 were obtained for the waste LiCoO₂ powders. However, the intensity of these characteristic peaks was hardly detected after leaching in 2.0 mol/L phosphoric acid, indicating a complete dissolution of waste LiCoO₂. Under acid concentration of 0.8 mol/L, the characteristic peaks around 003, 101, and 104 for LiCoO₂ powders almost disappeared, while new characteristic peaks with less intensity appeared after leaching, which may have been caused by the generation of new chemical compounds. In addition, different peaks with relatively weak intensities may be attributed to the amorphous crystal of the leaching product. Therefore, further characterization methods are still required for the determination of the obtained leaching product.

Pink leaching product was obtained after leaching in 0.8 mol/L acidic medium, and this precipitate was similar to pure cobaltous phosphate (Co₃(PO₄)₂). Thus, FT-IR was then applied for the final confirmation of obtained leaching product. According to Figure 5, two characteristic absorption peaks around 610 cm⁻¹ and 510 cm⁻¹ were obtained for waste LiCoO₂, indicating the bonds of Co-O and Li-O, respectively. It was discovered that new characteristic absorption peak zones around 580 cm⁻¹ (zone A) and 1040 cm⁻¹ (zone B) appeared after leaching, characteristic peaks of Co-O bond (blue shift phenomenon) and P-O bonds (four different characteristic absorption peaks). This indicates the generation of Co₃(PO₄)₂ and a complete dissolution of Li. In addition, it can be also confirmed from above characterization results that the leaching product and pure Co₃(PO₄)₂ display similar characteristic peaks, which further supports the assumption that the obtained pink precipitate was pure Co₃(PO₄)₂.

Based on the XPS analysis, the changes in the surface elements of the leach residue were determined. Figure 6A,B show the XPS high-resolution spectra and peak fittings of Li 1s in the LiCoO₂ leach residue before and after leaching. The binding energy of the Li 1s main peak shifts significantly towards higher energy after leaching, indicating that during the leaching process, the valence electrons of lithium undergo electrochemical displacement, and the bonding interactions change. Lithium in the leachate reacts with H₃PO₄, forming LiH₂PO₄, resulting in a gradual reduction of Li content in the leach residue until it disappears. Figure 6C,D display the high-resolution XPS spectra of Co 2p before and after leaching. It can be observed that the Co 2p spectrum consists of two major parts: the main peaks 1/2 and the satellite peak 1, as well as the main peaks 3/4 and satellite peak 2. The half-peak width of the Co²⁺ characteristic peak is greater than that of Co³⁺, so according to the diagram, peaks 1 and 3 correspond to Co³⁺, while peaks 2 and 4 correspond to Co²⁺. The satellite peaks S1 and S2 are associated with the Co²⁺ characteristic peaks. Before leaching, the Co³⁺:Co²⁺ ratio in the waste LiCoO₂ was 0.306, and after leaching, this ratio increased to 0.705. This indicates that the Co³⁺/Co²⁺ ratio in the leach residue increased after leaching, and Co in the LiCoO₂ was better retained in the solid phase. Thus, during the leaching process with a low concentration of H₃PO₄, selective leaching of Li from the LiCoO₂ can be achieved. The oxygen species on the surface of the LiCoO₂ crystal mainly consisted of lattice oxygen (O_latt) and surface hydroxyl oxygen (O_sur). Lattice oxygen originates from the internal structure of the LiCoO₂ crystal, while surface hydroxyl oxygen is introduced during the hydrothermal synthesis of LiCoO₂. As can be seen from the diagram, when LiCoO₂ participates in the leaching reaction, the oxygen species on the surface of the leach residue crystal undergo a change. The lattice oxygen peak shifts to the right and transforms into surface hydroxyl oxygen. The decrease in lattice oxygen indicates an increase in oxygen vacancies within the lattice, leading to the destruction of the crystal structure. The increase in surface hydroxyl oxygen enhances the reactivity of LiCoO₂, accelerating the leaching process of the sample in the phosphoric acid solution.

3.2. Metal Recovery from Real Waste Streams

Impurity metal ions (i.e., Cu, Al, and Fe) will inevitably flow into the waste stream during industrialization recovery process, and it is necessary to explore leaching behaviors of different metals in phosphoric acidic media. In this study, fixed amounts of Cu, Al, and Fe powders were mixed with waste LiCoO₂ powders to simulate real waste after mechanical crushing and screening. Detailed leaching results are listed in the Supplementary Materials (Table S2 and Figure S3), illustrating the leaching rates of metals under optimized experimental conditions in different acidic media. It can be observed that different metals demonstrated different leaching behaviors in different acid concentrations. Under mild acidic medium (0.8 mol/L), 98.8% Li and most of the Cu (76.7%) were leached and only small proportions of Al (38.2%), Co (3.6%), and Fe (9.1%) could be dissolved, which may indicate that Li in waste LiCoO₂ powders can be selectively extracted, followed by Cu, Al, Fe, and Co. However, these metals presented a complete dissolution phenomenon in a higher concentration of 2.0 mol/L phosphoric acid, indicating that different metals can be gradually leached with an increase in concentration. To achieve deeper understanding of these leaching reactions, the reaction kinetics were investigated for different metals.

Reaction kinetics: Leaching kinetics were investigated to explore the leaching behaviors of Co, Li, Al, Cu, and Fe from real waste at different retention times (20–60 min) and reaction temperatures (30–70 °C) under conditions of reductant dosage 4 vol.%, liquid-to-solid ratio 20 mL/g, and acid concentration 0.8 mol/L or 2.0 mol/L. It can be observed from Figure 7 that the correlation coefficients for chemical reaction controlled model (R²) for the leaching of Co, Li, Al, Cu, and Fe were over 0.9 in both a high concentration of acidic medium (2.0 mol/L) and a low concentration of acidic medium (0.8 mol/L), indicating that the leaching of different metals in phosphoric acid media is controlled by chemical reactions and both reactants and products can be rapidly diffused during the leaching process.

Apparent activation energies (Ea) were calculated based on the slope of the lnk vs. 1/T curve (based on the Arrhenius equations) in specific leaching conditions, which can be also adopted to evaluate difficulties of relevant leaching reactions in a low acid concentration of 0.8 mol/L (see Figure 8A). Frequency factors (lnA), and apparent activation energies (Ea) were calculated according to the Arrhenius equations (Equations (3) and (4)), and the following results were obtained: Li: lnA = 4.4689, Ea = 29.66 kJ/mol; Co: lnA = 3.3927, Ea = 11.31 kJ/mol; Fe: lnA = 2.7715, Ea = 10.69 kJ/mol; Al: lnA = 0.3275, Ea = 15.78 kJ/mol and Cu: lnA = 3.5981, Ea = 24.16 kJ/mol. The Ea values for different metals were over 10 kJ/mol, which indicates that the reactions were controlled by chemical reactions; this is in accordance with the leaching kinetics results. In addition, the lnk vs. 1/T curve also presents an excellent fit (R² > 0.98) for different metals in a high acid concentration of 2.0 mol/L according to Figure 8B. The following fitting results were obtained: Li: lnA = 8.9935, Ea = 40.08 kJ/mol. Co: lnA = 2.4621, Ea = 21.31 kJ/mol. Fe: lnA = 4.1055, Ea = 28.38 kJ/mol. Al: lnA = 9.1865, Ea = 41.27 kJ/mol. Cu: lnA = 10.0721, Ea = 45.52 kJ/mol.

As illustrated in Figure 9, the leaching reactivity followed the order Li > Cu~Co > Fe > Al, with corresponding final products identified as LiH₂PO₄, Cu(H₂PO₄)₂, Co₃(PO₄)₂, FePO₄ and Al(H₂PO₄)₃. The remarkably high lnA value for Li⁺ indicates superior reaction frequency, while its small ionic radius (0.76 Å) and substantial hydration energy (−515 kJ/mol) facilitate efficient lattice detachment and subsequent formation of soluble LiH₂PO₄. Notably, Cu²⁺ (0.73 Å, −2100 kJ/mol) and Co²⁺ (0.74 Å, −2050 kJ/mol) exhibit comparable leaching behaviors due to their nearly identical ionic radii and hydration energies. In contrast, the high hydration energy and strong electrostatic interaction by Fe³⁺ (0.64 Å, −4430 kJ/mol) and Al³⁺ (0.54 Å, −4690 kJ/mol) make it difficult for them to break away from the lattice, and insoluble phosphate (FePO₄, Al(H₂PO₄)₃) easily forms, as reflected by their low lnA and high Ea values. Li⁺ rapid leaching from its high lnA value; Cu²⁺/Co²⁺ leaching was modulated by both lnA and redox pathways (particularly Co³⁺ reduction), while Fe³⁺/Al³⁺ retardation arose from synergistic hydration energy and precipitation effects. Therefore, through sequential leaching control enables precise lithium-selective extraction in 0.8 mol/L H₃PO₄ mild acid medium.

In the 2.0 mol/L strong acid medium, the leaching sequence of metal ions followed the order Cu > Al~Li > Fe > Co (Figure 10), collectively determined by Ea, lnA, ionic properties (hydration energy, radius), and product characteristics. Although Cu²⁺ exhibited the highest Ea, it leached first due to its exceptionally high lnA, indicating an extremely effective collision frequency in phosphoric acid medium. The similar leaching rates of Al³⁺ and Li⁺ resulted from their comparable Ea and lnA values. Notably, despite its higher charge, Al³⁺ demonstrated considerable mobility in phosphoric acid owing to its smaller ionic radius. The relatively low Ea values for Fe³⁺ and Co²⁺ correlate with their formation of insoluble phosphates (FePO₄ and Co₃(PO₄)₂), which create dense passivation layers on particle surfaces. These layers introduce significant diffusion resistance, ultimately limiting leaching rates despite the lower activation energies. Our findings reveal a critical structure–activity relationship: metal ions forming soluble phosphate salts exhibit rapid leaching kinetics because the soluble products can readily depart from the reaction interface, thereby preventing passivation and maintaining high interfacial reaction driving forces. However, Li⁺ does not demonstrate priority leaching in this strongly acidic system, necessitating pre-removal of impurity ions to enhance Li⁺ selectivity in practical applications.

3.3. Proposed Recovery Process

After leaching, both leaching solution and leaching residues/products were obtained, and the Li⁺ enriched leaching solution and Li⁺ and Co²⁺ enriched solution were obtained in mild acidic medium (0.8 mol/L) and higher acid concentration medium (2.0 mol/L), respectively. Then, oxalic acid (H₂C₂O₄) was added to above leaching solutions for the precipitation of valuable metals and regeneration of phosphoric acid. As a typical organic acid, H₂C₂O₄ can be ionized into H⁺, HC₂O₄⁻, and C₂O₄²⁻ with stronger acidity than H₃PO₄ in aqueous. Therefore, H₃PO₄ can be simultaneously regenerated during the precipitating of Li⁺ and Co²⁺ as Li₂C₂O₄ and CoC₂O₄. Relevant chemical reactions are listed as follows:

H₂C₂O₄ (aq) = H⁺ (aq) + HC₂O₄⁻ (aq)  pKa₁ = 1.27

(12)

HC₂O₄⁻ (aq) = H⁺(aq) + C₂O₄²⁻ (aq)  pKa₂ = 4.27

(13)

2LiH₂PO₄ (aq) + H₂C₂O₄ (aq) = Li₂C₂O₄ (s) + 2H₃PO₄ (aq)

(14)

Co(H₂PO₄)₂ (aq) + H₂C₂O₄ (aq) = CoC₂O₄ (s) + 2H₃PO₄ (aq)

(15)

After precipitating, Li⁺ and Co²⁺ were recovered as their precipitates formed, and regenerated H₃PO₄ was used as leaching reagent for circulating leaching of waste LiCoO₂ powders under the optimized leaching conditions (4 vol.% H₂O₂, 60 °C, 60 min, 20 mL/g, and acid concentration 0.8 mol/L or 2.0 mol/L). According to the circulatory leaching results illustrated in Figure 11A, the leaching ability of regenerated phosphoric acid demonstrated a gradual attenuation from Cycle 1 to Cycle 5, and leaching efficiencies for Li and Co were 91.7% and 0.9% in Cycle 5. Despite the leaching efficiency of Li experiencing a gradual reduction across the leaching cycle, which may have been caused by the loss of phosphoric acid during the formation of Co₃(PO₄)₂ in mild acidic medium, the leaching efficiency of Co always stayed at a fairly low level. This indicates that regenerated phosphoric acid can still be treated as a leaching reagent for the selective leaching of Li from waste LiCoO₂. Calculating the pH and phosphate ion concentration, we found that supplementing with 2% fresh acid can counteract the decline in leaching ability of the recycled acid, allowing the used leaching reagent to still be considered effective. In Figure 11B, the leaching efficiencies of Co and Li show a gradual decline with the proceeded leaching cycle. Almost 100% Co and Li was leached in Cycle 1 and the leaching rates for Co and Li fell to 93.4% and 85.2% in Cycle 5. The declined leaching efficiencies of Li and Co may be attributed to incomplete precipitating of Li⁺ and Co²⁺, and insufficient leaching ability of LiH₂PO₄ and Co(H₂PO₄)₂. Therefore, further investigation for the optimization of leaching replacement reactions may be required to achieve optimal leaching efficiencies for Co and Li.

3.4. Environmental Impact and Economic Evaluation

In this work, a hydrometallurgical process for the recovery of Co and Li from spent LIBs under different acidic media (see Figure 12) is proposed. To assess the environmental impact of different recycling processes, we compared the proposed method with reported technologies for recycling LIBs (Table S4). It was observed that, compared to heat treatment (700 °C, 600 min) and organic solvent dissolution (70 °C, 90 min), this method requires a lower reaction temperature (60 °C) and a shorter retention time (60 min), which greatly reduces energy consumption. In addition, phosphoric acid is regenerated during the recovery of Li and can be reused in circulating leaching process. However, valuable metals can be completely dissolved in higher concentrations of phosphoric acid, and Co and Li can be selectively and sequentially precipitated as CoC₂O₄ and Li₂C₂O₄. Simultaneously, phosphoric acid is regenerated during the precipitating reactions and the regenerated acid can be reused as leaching reagent.

Furthermore, reagent costs in this study were evaluated and compared with other processes to confirm feasibility and potential recycling profits (Table S5). A treatment sample of 1.0 kg of spent LIBs was selected, and the types, dosages, and prices of chemicals were discussed. It was found that the costs of reagents using ionic liquid (USD 402.3), organic solvents (USD 10.37), and molten salts were much higher than that in this study (USD 4.56). It can be concluded from the whole recovery process that valuable metals can be recovered as a shortcut or efficient way with high recovery efficiencies, as well as fitting the requirement of atomic economy and circular economy.

4. Conclusions

In this study, a hydrometallurgical process is proposed for the selective recovery of metals from spent LIBs. Comprehensive work in terms of selective leaching of valuable metals, recovery of different metals from real waste stream, leaching kinetics, and circulating leaching was conducted to explore the leaching behaviors of different metals. Based on the above experimental and theoretical results, the following conclusions can be obtained:

(a) For the recovery of valuable metals, 100% Li can be selectively extracted from waste LiCoO₂ of spent LIBs in a short-cut way under mild acidic medium (0.8 mol/L H₃PO₄), while Li and Co illustrate a complete dissolution in high acid concentration of 2.0 mol/L H₃PO₄ under optimized leaching conditions. Different leaching behaviors of Co and Li in different acidic media indicate that they can be selectively recovered based on the concentration of phosphoric acid;

(b) It can be concluded from leaching results and reaction kinetics of different metals that Li, Co, Cu, Al, and Fe demonstrate different leaching behaviors during the leaching reactions in different acidic media, and the leaching of different metals is controlled by chemical reaction. The leaching reactions for different metals indicated an order of Li > Cu~Co > Fe > Al and Cu > Al~Li > Fe > Co in 0.8 mol/L H₃PO₄ and 2.0 mol/L H₃PO₄, respectively.

(c) The circulating leaching results reveal that valuable metals and phosphoric acid can be simultaneously precipitated and regenerated by the addition of oxalic acid to leaching solutions; 91.7% Li and 0.9% Co were leached in mild acidic medium, and 93.4% Li and 85.2% Co were dissolved in 2.0 mol/L H₃PO₄ after five leaching cycles, indicating that Li and Co in different acidic media can be selectively precipitated as oxalate while the regenerated acid may be used as leaching reagent.