Ternary Deep Eutectic Solvent (DES) with a Regulated Rate-Determining Step for Efficient Recycling of Lithium Cobalt Oxide.

Deep eutectic solvents (DESs) have attracted extensive research for their potential applications as leaching solvent to recycle valuable metal elements from spent lithium ion batteries (LIBs). Despite various advantages like being economical and green, the full potential of conventional binary DES has not yet been harnessed because of the kinetics during leaching. Herein, we consider the fundamental rate-determining-step (RDS) in conventional binary DES and attempt to design ternary DES, within which the chemical reaction kinetics and diffusion kinetics can be regulated to maximize the overall leaching rate. As a proof of concept, we show that the ternary choline chloride/succinic acid/ethylene glycol (ChCl/SA/EG) type ternary DES can completely dissolve LCO powder at 140 °C in 16 h. By systematically studying the leaching process at various conditions, the energy barrier during leaching can be calculated to be 11.77 kJ/mol. Furthermore, we demonstrate that the extraction of the cobalt ions from the leaching solution can be directly achieved by adding oxalic ions without neutralizing the solution. The precipitate can be used to regenerate LCO with high purity. The recycled materials show comparable electrochemical performance with commercial LCO. Our design strategy of ternary DES with regulated RDS is expected to have both scientific and technological significance in the field of hydrometallurgical recycling of LIBs.

Introduction

The massive production and upgrading of electronic devices and electric vehicles have led to the rapid growth for the production of rechargeable lithium ion batteries (LIBs),[1−8] which will translate to tremendous amounts of waste in the coming years.[9−17] Direct disposal of LIBs through landfilling or burning would cause severe environmental pollution and resource wasting,[18,19] thus calling for the development of an efficient, economical, safe, and green recycling strategy of LIBs.[20−24] Among various components in LIBs, most efforts to improve recycling have focused on the extraction of valuable metal elements from cathodes, especially cobalt for its high value and environmental consideration.[21,25−28] Currently, pyrometallurgy or mixed pyro- and hydrometallury is the dominating approach used in industry (companies including Accure GmbH, Sony. Umicore, etc.).[29,30] The pyrometallurgy process generally involves a set of high-temperature processes for the extraction of metal elements, including calcination, roasting, oxide reduction, or smelting.[31] It is relatively efficient for recycling various transition metal elements. However, the combustion treatment for the battery components would generate huge energy consumption and environmental pollution. Hydrometallurgy is considered to be a promising alternative strategy due to its potential to extract metal elements with high purity.[32,33] Hydrometallurgy involves the leaching process, which converts the metal elements from electrodes to ions in solutions, followed by a series of chemical methods including precipitation, solvent extraction, or electrolytic deposition to recover the metal resources. Various solvents have been studied for the leaching process.[34] Still, the simultaneous realization of high efficiency and environmental friendliness is challenging. Inorganic acid leaching approach involves caustic reagents, which poses potential danger to workers and the environment.[35] Other mild solvents generally have their own problems. For example, organic acids like malic acid or oxalic acid are less corrosive than strong inorganic acids, but additional additives are required in the leaching process to reduce the valence of metal ions and facilitate the leaching kinetics.[34,36,37] Despite much progress, a relatively high temperature and high concentration of caustic leaching reagent are necessary for efficient leaching. Therefore, searching for new green solvents and understanding the mechanism to optimize the reaction kinetics are of both scientific and technological significance.[38−40]

Recently, deep eutectic solvents (DESs) have emerged as a new type of solvents that show unconventional interaction toward various metal compounds.[41−43] Initially, DESs are considered to be quasi-ionic liquids due to their characteristic properties including low vapor pressure and high thermal and chemical stability,[44] as well as high viscosity,[45] while further research discovered that DESs are distinguished from ionic liquids due to the existence of a considerable molecular portion.[46,47] This enables DESs to be more compatible in practical application due to the easier preparation. Noteworthingly, various metal oxides are reported to show high solubility in DESs,[48−50] which is attributed to the breaking of the metal-oxide bonds of the solid in the DESs, which have rich hydrogen bond donor (HBD) species.[50] In 2019, the extraction of cobalt element from lithium cobalt oxides (LCOs) was reported in a choline chloride/ethylene glycol (ChCl/EG) type DES without the addition of additional leaching reagent,[51] opening a new frontier for the search of green solvent in lithium recycling. Considering the rich variety of DESs, many follow-up steps have been taken to optimize the leaching efficiency.[52−54] Fundamentally, the efficiency of a chemistry process involves different rate-determining steps (RDSs) like reaction kinetics and mass transfer rate, thus requiring the optimization of the DESs composition (For example, the use of ternary DES) to realize the full potential. However, to our knowledge, current research works mainly focus on binary DESs, which are composed of a hydrogen bond acceptor (HBA) and an HBD. The intrinsic property of the selected HBA and HBD in a binary DES often limits the simultaneous realization of high reaction kinetics and the mass transfer rate. Despite the fact that ternary DESs have been proved powerful in many applications including the dissolution of lignin,[55] deconstruction of wheat straw, and CO absorption.[56,57] This design principle has not yet been studied to improve the efficiency for recycling LIBs.

Herein, we report the use of a ternary ChCl/SA (succinic acid)/EG type DES for extracting cobalt element from LCO materials. The design consideration is based on the high activity of the hydrogen bond donor of SA, which provides high kinetics for chemical reaction, as well as the high mobility of EG, which provides better diffusivity. At an optimized molar ratio of 1 ChCl/1 SA/1 EG, a high leaching efficiency of 99.62% can be realized at 140 °C for 16 h. Meanwhile, we found that the precipitation process can be achieved by tuning the complexation environment with the addition of sodium oxalic Na2C2O4 (Na2C2O4; 99%; Shanghai Saen Chemical Technology Co. Ltd.) The cobalt ions can precipitate in the form of CoC2O4 without the neutralization of excessive protons, which is more superior than conventional alkali precipitation strategy in terms of the consumption of chemicals. The precipitated cobalt compounds can be used to refabricate LCO active materials with high purity. The recovered LCOs show a performance of electrochemical that is comparable to commercial LCO raw materials. Such a strategy of regulating the rate-determining step based on a ternary DES is expected to be a general approach for developing a green solvent with high efficiency in LIBs recycling.

Experimental Section

Preparation of the DES

Choline chloride (C5H14ClNO; 98%; Shanghai Saen Chemical Technology Co. Ltd.), succinic acid (C6H10O4; 99%; Shanghai Technology Co. Ltd.), and ethylene glycol (C2H6O2; 99%; Shanghai Saen Chemical Technology Co. Ltd.) were used to prepare the ChCl/SA/EG ternary DES. ChCl was mixed with SA to prepare binary ChCl/SA DES at 90 °C, forming a transparent solution. A given amount of EG was then added to obtain the ternary DES with various ratios to investigate the DES with optimized extraction kinetics.

Leaching and Extraction of Cobalt

The leaching efficiency was studied by adding 3 g of ternary DES in a glass vial. Then, 10 mg of lithium cobalt oxide (LiCoO2; 99%; Guangdong Canrd New Energy Technology Co. Ltd., LCO) was added into the solution. The glass vial was heated in an oil bath at various temperatures and heating times. The transparent solution turned blue, indicating the dissolution of the LCO.

To extract the cobalt ions, 0.536 g of sodium oxalate (Na2C2O4; 99%; Guangdong Canrd New Energy Technology Co. Ltd., LCO) solid was dissolved in 100 mL of deionized water to prepare 0.04 mol/L sodium oxalate solution, and 5 mL was taken in ternary DES dissolved with 0.01 g of LCO. One can see that solids are being produced in the cloudy solution. After the solid is completely precipitated, filtrate and dry, pink cobalt oxalate solid can be obtained. Black Co3O4 can be obtained by burning cobalt oxalate in a muffle furnace at 400 °C for 5 h. The obtained Co3O4 and Li2CO3 were evenly mixed in a muffle furnace at 600 °C for 5 h, and then, the temperature was adjusted to 900 °C for another 10 h to obtain the recycled LCO. The structures of the recycled materials were characterized by X-ray diffraction (XRD, RigakuD/Max-3B) and scanning electron microscopy (SEM, Hitachi S-4800)

Electrochemical Performance of Recycled LCO

The electrochemical performance of the recycled LCO was tested with coin cells. The active materials were mixed with PVDF binder and carbon black with a weight ratio of 8:1:1 using NMP solvent. The slurry was then coated on the aluminum foil and then placed in an oven at 80 °C for 1 h and then further dried at 110 °C under a vacuum for 12 h. The electrode was then cut into circular sheets with a diameter of 12 mm and assembled using Li foil as the counter electrode and 1 M LiPF6 in ethylene carbonate (EC)–dimethyl carbonate (DME) (1:1 in volume) as the electrolyte in a coin cell. Finally, various performance tests of the battery were carried out. Cyclic voltammetry (CV) was performed using a CHI660e electrochemical workstation (Chenhua, Shanghai). Galvanostatic charge–discharge tests were conducted on a LAND battery program-control test system.

Results and Discussion

Improving the Leaching Efficiency by Regulating the RDS in Ternary DES

The recycling of LIBs through a hydrometallurgical approach based on DESs has attracted extensive research for its potential to extract metal elements with high purity and environmental friendliness. Currently, considering the existence of the large variety of DESs, extensive researches have been devoted to search for the type of DES with optimized leaching efficiency. Still, the leaching efficiency needs improvement to meet the demand for the LIBs recycling at large scale. The complete leaching of cathode powders using binary DES composed of a HBD and a HBA species generally require tens of hours or even days.[52,53,58] Fundamentally, the leaching process of the solid materials in solvents involves the chemical reaction process and the mass transfer process. The chemical reaction rate is limited by the reaction kinetics while the mass transfer rate is limited by the mobility of the substance. The overall leaching rate is determined by the slowest step involved (RDS). Therefore, when a certain type of DES is used for leaching cathode materials, the apparent leaching rate is limited by either the chemical reaction kinetics or the diffusion capability of the solvent (Figure a). For example, choline chloride (ChCl) has been widely used as the HBA species to prepare DES due to its easy availability. When another HBD species is mixed with ChCl, the HBD species (e.g., succinic acid) with high proton activity would promote the debonding between the metal and the oxygen atoms, thus benefiting the kinetics of the chemical reaction. But, such ChCl/SA type DES provides poor diffusion capability for mass transfer. On the contrary, when the HBD species (e.g., ethylene glycol) with high molecular mobility is selected, their chemical reaction kinetics toward dissolving the cathodes are usually not satisfactory. As a result, the overall leaching rate of cathode materials in binary DES is limited even though the system might have one superior property in terms of chemical reaction kinetics or mass transfer capability.

Figure 1

Schematic illustration showing the design concept based on optimizing the rate-determining-step (RDS). (a) Overall reaction kinetics of a ternary type DES can theoretically surpass those of a binary DES due to the synergistically regulated rate-determine-step (RDS). (b) Leaching experiment. The dissolution of LCO in ChCl/SA/EG type ternary DES is significantly faster than those of binary type DESs composed of either ChCl/EG or ChCl/SA.

Herein, we attempt to break through the bottleneck of conventional binary DES by designing ternary DES system with regulated RDS. The ternary DES composed of ChCl, SA, and EG was selected as a proof of concept to demonstrate the synergistic effect for improving the overall leaching efficiency. Figure b briefly illustrates the synthesis and comparison of the DESs. By directly mixing the starting raw materials with a known ratio, binary or ternary DES with a given composition can be prepared. Afterward, LCO powders were added into the DES. The whole system was then kept at various reactions conditions. The leaching process is accompanied by the generation of the [CoCl4]2– complex, which endows the solvent with blue color, thus enabling the direct estimation of the leaching degree by comparing the colors. As shown in the inset of Figure a, at the same condition (90 °C, 2 h), the leaching rate changes with the composition and was optimized when the molar ratio of the ternary ChCl/SA/EG was set as 1:1:1. This result provides solid evidence for the feasibility of using ternary DES to push beyond the conventional binary whose leaching rate is limited by their RDS.

Leaching Kinetics of the Ternary DES

To demonstrate the superior overall reaction kinetics of the ternary ChCl/SA/EG system, we systematically studied the leaching process at various conditions (90–150 °C, 0.5–16 h) using an oil bath. As shown in Figure b, when 10 mg of LCO powder was dissolved in 3 g of DES, the color of the leaching solution deepened with the increase of the reaction temperature and time. When the reaction proceeded at 140 °C for 16 h, no solid residues could be observed, indicating the complete dissolution. UV–vis was used to quantify the concentration of the dissolved ions in the supernatant leaching solution. Three characteristic bands (630, 667, and 696 nm) were detected, which correspond to the absorption of [CoCl4]2– complex species.[59] The peak centered at 696 nm was used to quantify the concentration of the dissolved cobalt ion. First, solutions with a series of known concentration were prepared to obtain the standard concentration curve (Figure a). Afterward, the UV–vis spectrum of the leaching solutions could be measured (Figure and Figure S2) and used to calculate the concentration of the dissolved cobalt ions at given temperature and time. To describe the progress of the leaching process, the leaching efficiency (η, ratio between the dissolved and input cobalt elements) was calculated and is given in Figure b. The value of η can be defined aswhere C is the measured concentration of the cobalt element (mg/L), V is the volume of the solution (L), and MCo is the mass of the cobalt element in the added LCO. The leaching efficiency increases with the reaction temperature or time. When the reaction time was set as 16 h, the leaching efficiencies at both 140 and 150 °C were close to 1.0, which corresponds to the complete dissolution as visualized in the digital pictures in Figure a.

Figure 2

Dissolution efficiency of LCO ternary ChCl/SA/EG with various reaction conditions. (a) Digital photos of the reaction solution in glass vials. (b and c) UV–vis spectra of the supernatant leaching solutions at temperatures of 90 (b) and 150 °C (c).

Figure 3

Leaching kinetics of the LCO in the ternary DESs. (a) Standard concentration curve used to convert the tested UV–vis to the concentration of cobalt ions. (b) Leaching efficiency of solution under different conditions. (c and d) Kinetics fitting and the linear fitting of activation energy of cobalt leaching in the leaching process.

On the basis of the above leaching efficiency at various reaction conditions, the leaching kinetics can be calculated for fundamental understandings. The leaching process is accompanied by the continuous conversion of solid cobalt in the lattice to the ionic cobalt complex in the solution, during which the size of the solid particle decreases until complete dissolution in a solid–liquid leaching process. Therefore, an unreacted nucleus contraction model can be used to describe the reaction and calculate the overall leaching kinetics.[60] If the ratio between various leached ions stays constant, the apparent leaching rate equation can be established aswhere K is the reaction rate constant while t stands for the leaching time (min). At a given temperature, the reaction rate constant can be fitted (Figure c). Afterward, the activation energy (Ea) can be derived by fitting the reaction rate constant at various temperature usingwhere A is the pre-exponential factor, R is the molar gas constant, and T is the thermodynamic temperature (K). By fitting the −ln K to 1000/T, the apparent activation energy Ea can be calculated to be 11.77 kJ/mol. The low activation energy of the ternary DES is superior than binary DES and is essential for realizing efficient leaching. The above results provide solid evidence for the feasibility of our proposed strategy to improve leaching kinetics by regulating the RDS in a ternary DES system.

Regeneration of the LCO from the Leaching Solution

The synergistic effect in ternary DES system improved the overall leaching efficiency by balancing the chemical reaction kinetics and mass transfer kinetics. To evaluate its potential for practical application, the next step is to develop economical methods to extract the dissolved elements and regenerate functional materials with high performance. In the proposed ChCl/SA/EG system, we discovered that the precipitation of the cobalt ions can be achieved by adding sodium oxalate without neutralizing the solution in conventional alkaline-based precipitation approach, thus significantly simplifiying the extraction process. The precipitate can be collected and cleaned by a repeated centrifuging/washing process. The obtained cobalt compounds are CoC2O4 (Figure a) with irregular bulk morphology (Figure S3). By annealing the powders at 400 °C for 5 h, CoC2O4 can be converted to Co3O4 (Figure S6) with strip-like morphology (Figure S4), which can further be used to regenerate LCO by annealing with Li2CO3 at 600 °C for 5 h and 900 °C for 10 h. The regenerated LCO has high purity and crystallinity (Figure a) with microscale size (Figure S5). The structure of the regenerated LCO is expected to enable the materials to function well in LIBs.

Figure 4

(a) XRD of the precipitate before and after calcination after adding excessive deionized water and recycle LCO. (b) Cycle test of regenerated LCO materials under better conditions. (c) Rate capability test of regenerated LCO materials under better conditions.

To verify the suitability of the regenerated LCO as potential cathode used in LIBs, we used coin cells to compare its electrochemical performance with commercialized microsized LCO (morphology given in Figure S5). CV curves show that the recycled LCO has a pair of sharp redox peak at 4.0/4.2 V (Figure S7), which is similar to commercial LCO. The recycled LCO shows an initial discharge capacity of 164.2 mAh/g at a current density of 0.2 C and a voltage range of 3.0–4.5 V. After 50 cycles, 141.7 mAh/g can be maintained (Figure b). The charge/discharge curves at different cycles in the inset of Figure b show a plateau at 3.7 V. The rate capability in Figure c shows that the discharge capacity of the recycled LCO does not show obvious change when the current densities are in the range 0.1–2 C. The above results indicate that the regenerated LCO has comparable electrochemical performance compared with commercialized LCO in terms of specific capacity, kinetics, and stability. The high quality of the regenerated LCO proves that the ternary DES is qualified as an efficient solvent for the hydrometallurgical recycling of cathodes from spent LIBs.

Conclusions

This work demonstrates the successful improvement of the overall leaching efficiency when using DES for cobalt recycling from spent LIBs. The rational design of ternary DES provides a synergistic effect to regulate the chemical reaction kinetics and diffusion kinetics, which outperforms conventional binary DES whose overall leaching kinetic is limited by its RDS. For the ChCl/SA/EG ternary DES system shown in this work, the complete dissolution of LCO can be achieved at 140 °C for 16 h. Systematic study of the leaching experiment at various conditions show that the energy barrier during leaching is 11.77 kJ/mol. The extraction of the dissolved cobalt element can be simply achieved using oxalic ions. The precipitate can be used to regenerate LCO, which shows comparable electrochemical performance with commercial LCO in terms of discharge capacity, rate capability, and stability. The above results prove the viability for using ternary DES with regulated RDS to improve the leaching kinetics. The further extraction and regeneration of valuable materials from the leaching solution is easy and efficient, thus enabling the design strategy of ternary DES to be of both scientific and technological significance.