Application of Hydrophobic Deep Eutectic Solvents in Extraction of Metals from Real Solutions Obtained by Leaching Cathodes from End-of-Life Li-Ion Batteries

Abstract

This paper presents the results of applying hydrophobic deep eutectic solvents (HDESs) for the extraction of metal ions from a real hydrochloric acid solution after leaching the cathodes of three different types of Li-ion batteries. Aliquat 336-, D2EHPA- and menthol-based HDESs developed by us were used in this study. The optimal HCl leaching conditions chosen are 80 °C, 2 M HCl, 6 h, solid:liquid ratio = 1:25. The results of stepwise separation of the major elements using extraction with HDESs are presented. The HDESs used in the cross-current extraction made it possible to extract all elements with extraction ratios above 98%. It was shown that the suggested method could potentially be used in the process of recycling end-of-life Li-ion batteries.

1. Introduction

The progress in technology, intensive marketing strategies and increasing consumer demand have led to a rapid turnover of electronic devices and equipment (e.g., cell phones, laptops, electric cars) that use lithium-ion batteries (LIBs). LIBs consist of three main components: a cathode, an anode, and a porous separator impregnated with an electrolyte. Typically, the anode consists of porous carbon on copper foil and intercalated metallic lithium. As the battery discharges, lithium ions move from the anode to the cathode. In turn, the cathode materials differ in composition and may contain large amounts of valuable elements such as Co, Ni, Mn, and Li. For this reason, the cathodes of end-of-life LIBs correspond to secondary resources that can potentially replace ores [1,2,3]. In addition, the flammable organic liquid electrolytes used in LIBs pose serious safety problems if they are overheated or overcharged. For all these reasons, they require complex recycling.

The best world practices show that, of the traditional methods that include leaching, extraction, precipitation, membrane methods, ion exchange, etc., the hydrometallurgical method is the most attractive approach for the extraction of valuable metals from end-of-life LIBs [4,5,6,7]. In practice, these processes are usually combined to achieve high recovery efficiency. Extraction is a widely known and frequently used method for the recovery of metal ions from aqueous solutions. It has important advantages such as high selectivity, low energy consumption in a continuous process, cost-efficiency, and easy industrialization. The processes for the separation of metal ions contained in LIBs most commonly employ neutral and cation-exchange extractants such as organophosphorus and carboxylic acids (D2EHPA, Cyanex 272, Cyanex 301, PC-88A, Versatic 10), tributyl phosphate, trioctyl phosphine oxide, etc. [8,9]. Owing to these advantages, they are widely used in many industrial processes [10]. However, traditional extraction systems employ many toxic and inflammable organic solvents and sometimes have poor selectivity for the target components. New-generation extraction systems based on deep eutectic solvents can solve the problems outlined above.

Deep eutectic solvents (DESs) are a new class of solvents that are often mixtures of two compounds linked by hydrogen bonds. Due to the wide range of starting compounds, DESs can be adapted to specific applications in catalysis, hydrometallurgy, organic synthesis, electrochemistry, etc. [11,12,13,14,15,16]. Hydrophobic deep eutectic solvents (HDESs) can serve to replace expensive hydrophobic ionic liquids or at least allow the use of the latter to be diminished in processes of separation and extraction of metal ions from aqueous media. HDESs have physicochemical properties that can be easily controlled in a wide range by changing the type of the hydrogen bond donor (HBD) and acceptor (HBA), which makes it possible to solve many specific practical problems. Most frequently, HDESs are composed of carboxylic acids, quaternary ammonium base salts (QASs), organophosphorus compounds, terpenes, etc. [17,18].

Among other things, the interest in HDESs in the extraction of metal ions from aqueous solutions is due to the possibility of selective extraction [19,20,21]. Schaeffer et al. [22] studied an HDES based on terpenes and fatty acids to separate Cu(II) from other transition metals in weakly acid solutions. An interesting result was that this HDES showed high selectivity for Cu(II) compared to Ni(II) and Co(II), with separation ratios of over 500 and 154 for the Cu/Co and Cu/Ni pairs, respectively. The selective extraction of Li(I) ions using HDES was first demonstrated by Hanada and Goto [23]. An HDES based on tenoyltrifluoroacetone (HTTA) and trioctylphosphine oxide (TOPO) made it possible to extract Li for 95.7% with negligible extraction of Na and K. The Li/Na separation factor was 2000, which is an extremely high value for the selective extraction of Li.

The main studies in the field of recycling end-of-life LIBs aimed at the use of hydrophilic DESs to extract metals from active materials, mainly from the cathodes. In this work, we suggest a new promising method that combines leaching with hydrochloric acid and extraction with a modern class of solvents, HDES, for the recovery of transition metals (Ni, Co, Cu, Fe), Al and Li from the cathode material of end-of-life LIBs. The novelty of our work is that it employs two-component menthol-based HDESs with a variable second component that is responsible for the overall extraction capacity of an HDES. The menthol in the HDESs acts as a replacement for organic solvents in the case of Aliquat 336 and D2EHPA. It is important to note that all the experiments were performed using real hydrochloric acid-based leaching solutions, which resulted in the development of schemes for the treatment of cathodes from LIBs of various types.

2. Materials and Methods

2.1. Materials and Reagents

In this work, we used the cathode materials from LIBs of various types (two cylindrical batteries and one from a charging device): Momax PL 985797 (3.7 V) (LIB 1), Panasonic NCR18650B (3.6 V) (LIB 2), and Robiton LiFe18650 (3.2 V) (LIB 3). The cathode material (cathode powder and aluminum substrate) was obtained from end-of-life LIBs after the disassembly of the battery case. The main parts of LIBs are shown in Figure 1: the battery case (1), the cathode and anode material (2), and the separator (3).

In the first stage, the composition of the materials being treated should be monitored by X-ray spectrometry, one of the main chemical identification methods. The composition of the powdered cathode materials without the aluminum substrate was determined by X-ray fluorescence analysis (XRF, wave-dispersive X-ray fluorescence spectrometer SPECTROSCAN-MAX GVM, Spectron, Russia) (

Table 1. Composition of the surface of cathode materials determined by XRF.

Element	wt.%
	LIB 1	LIB 2	LIB 3
Si	0.09	0.09	0.07
P	0.49	0.44	36.68
S	0.28	0.07	0.41
Cl	0.14	0.12	0.04
K	0.12	0.13	0.19
Ti	0.19	0.03	0.02
Cr	0.05	0.02	0.03
Mn	0.07	0.06	0.07
Fe	0.08	0.06	62.12
Co	97.03	4.34	0.21
Ni	0.24	94.54	0.01
Cu	1.22	0.09	0.15

).

These semi-quantitative data are used for identification purposes since a versatile calculation procedure by the fundamental parameters method is employed. However, this information is the basis for the further analytical study of the identified target components and for the use of inductively coupled plasma atomic emission spectroscopy, which is a more accurate quantitative method.

The composition of the cathode materials from the end-of-life LIBs was also determined using an inductively coupled plasma optical emission spectrometer (ICP-OES, Thermo Scientific ICAP PRO XP, Waltham, MA, USA). The method involves nebulizing a solution into a high-frequency inductively coupled plasma in order to excite the spectrum of the sample, followed by automatic measurement of the analytical line intensities of the elements and determination of the amount of these elements from the calibration characteristics.

To achieve the maximum sensitivity in the measurement of micro-impurities, the analysis was performed in the axial plasma overview mode. Elements with concentrations above 50 mg/L were determined in the radial plasma overview mode. Analysis by the ICP-AES method was performed at the following working spectrometer parameters: the generator output power (forward power) was 1150 W, the nebulizer gas flow rate was 0.60 L/min, the auxiliary gas flow rate was 0.35 L/min, the coolant gas flow rate was 10 L/min, and the peristaltic pump rate (sample flow rate) was 60 rpm.

The solutions for building the calibration curve were prepared from MES-1, MES-2 and MES-4 multi-element standard solutions (Skat, RF) with a concentration of 50 mg/L and from single-element standard solutions of Co, Ni, Fe (Inorganic Ventures, Christiansburg, VA, USA) with a concentration of 1 g/L by diluting the solutions to concentrations of 100, 50, 10, 1 and 0.5 mg/L with 1% nitric acid solution. The 1% nitric acid solution was prepared from nitric acid of ‘special purity’ grade according to GOST 11125 and deionized water with a resistivity of 18.2 MOhm·cm.

The main criterion for selecting the analytical lines was that it had to be possible to take the background in the vicinity of a line into account. The wavelengths of the analytical lines are shown in

Table 2. Analytical lines of the elements being analyzed.

Element	Wavelength, nm
Fe	259.940
	239.562
Al	167.079
	396.152
Li	610.362
	812.645
Ti	334.941
	323.452
Nb	309.418
	316.340
Co	238.892
	237.862
Ni	300.249
	221.647
Cu	324.754
	327.396
Mn	257.610
	259.373
K	766.490
Si	251.611
	212.412

.

The analysis was performed after the complete dissolution of the samples in an HNO₃:HCl mixture (1:3) under the following conditions: solid:liquid ratio = 1:25, 2 h, 80 °C (

Table 3. Mean content of the main components in the waste cathode material powder (p = 0.95, n = 4).

Element	Li	Al	Co	Fe	Ni	Cu	Battery
wt.%	6.89	8.67	54.06	-	-	0.2	LIB 1
	4.25	8.72	2.8	0.18	64.42	-	LIB 2
	1.31	14.83	-	25.32	-	-	LIB 3

).

The relative determination error did not exceed 1%. The validity of the results obtained was verified by analyzing standard samples of the (HPS, North Charleston, SC, USA) formulation and model mixtures based on them.

Information on the chemicals used in the experiments is presented in

Table 4. The chemicals.

Compound	Supplier	CAS	Purity, wt.% *
Aliquat 336	Acros	63393-96-4	98
D2EHPA	Acros	298-07-7	95
L-Menthol	Acros	2216-51-5	99
HCl	Aldosa	7647-01-0	99
NaOH	Chimmed	1310-73-2	98
Distilled water	–	–	–

* Declared by the suppliers.

. All the chemicals were used as received from the supplier without further purification.

Preparation of DESs

All the HDES mixtures were prepared by adding an HBA to the corresponding HBD in various molar ratios. The mixture was stirred in an Enviro-Genie SI-1202 temperature-controlled shaker (Scientific Industries, Inc., Bohemia, New York, USA, accuracy ± 0.2 °C) at 60 °C for 30 min until a homogeneous transparent liquid formed. After the reaction, the resulting liquid was slowly cooled to room temperature and stored in the air until use.

The following mixtures were used:

HDES 1: Aliquat 336—menthol, 1:1 [24];

HDES 2: D2EHPA—menthol, 1:1 [25].

2.2. Leaching Experiments

Leaching experiments were performed in a 25 mL glass reactor equipped with a temperature control system (accuracy ± 0.2 °C) and a magnetic stirrer (750 rpm). A hydrochloric acid solution prepared in advance was added to the reactor, heated to 80 °C, and then the cathode material was added to a solid:liquid mass ratio of 1:25. The leaching process was performed for up to 6 h. If required, small samples of the solution were taken at certain time intervals and analyzed by ICP-OES.

The degree of metal leaching (L, %) was calculated as the ratio of the amount of the metal in the leaching solution being studied to the amount of the metal dissolved in aqua regia.

2.3. Extraction Experiments with HDESs

All the experiments on extraction with HDESs were performed in graduated centrifuge tubes in a temperature-controlled shaker at a spin rate of 35 rpm, with various phase ratios at room temperature (22 ± 2 °C). The phases contacted for 30 min in the extraction experiments. The re-extraction process was performed using a technique similar to the extraction.

The metal extraction efficiency (E%) was calculated using the equation:

$$\mathrm{E}\left(\%\right)=\frac{{\mathrm{n}}_{\mathrm{in}}-{\mathrm{n}}_{\mathrm{aq}} }{{\mathrm{n}}_{\mathrm{in}}}\times 100$$

(1)

where n_in and n_aq are the amounts of metal ions in the initial solution and in the aqueous solution after the extraction, respectively. The data was measured in triplicate for each extraction, and mean values were calculated with a relative standard deviation of less than 5%.

3. Results and Discussion

3.1. HCl Leaching of the Cathodes of Three Different LIB Types

The metal leaching efficiency (L) was studied as a function of three factors: leaching temperature, process time, and hydrochloric acid concentration.

It was shown in the previous studies [26,27] that the leaching efficiency of metal ions with an HCl solution increases with the process temperature. To find out how the leaching degree depends on temperature, the values of 25 and 80 °C were chosen.

The studies were performed at the same HCl concentration (2 M) in the time range from 15 to 360 min using LIB 1 as an example. Figure 2 shows that at 25 °C (Figure 2a), L reaches a plateau after 60 min, while at 80 °C (Figure 2b), the leaching degree continues to increase and reaches higher values: quantitative recovery of Al and Cu is achieved, against 94.42 and 89.21% for Li and Co, respectively. It can be concluded that for the types of LIBs investigated, there is also a tendency for the degree of metal leaching to increase with increasing process temperature. Based on the results obtained, 80 °C was chosen as the temperature for the further leaching experiments.

The time dependences of the metal leaching degree for LIB 2 and LIB 3 are shown in Figure 2c,d, respectively. In both cases, the metal leaching degree increased with increasing process time. In the case of LIB 2 (Figure 2c), this effect was most pronounced, and the leaching rates reached almost quantitative values at 360 min. In the case of LIB 3 (Figure 2d), similar data were obtained. Based on the results obtained, it can be concluded that 360 min is the optimal leaching process time since nearly quantitative extraction of the majority of the metals is observed.

As shown previously [28], the concentration of hydrochloric acid affects the leaching process considerably, namely, as the HCl concentration is increased from 1 M to 1.75 M, the Co, Mn and Li leaching efficiency increases by 15–30%. In our case, the leaching behavior of metals was studied as a function of the hydrochloric acid concentration in the range from 2 M to 6 M for 60 min. The results are presented in Figure 3. The plots show that the degree of leaching of the metals increases with an increase in the acid concentration. However, if the target metals (Li, Co, Ni) are considered, this effect is sometimes not observed or is insignificant, so increasing the concentration of hydrochloric acid above 2 M is inexpedient.

Thus, a solution obtained upon leaching at 80 °C in 2 M HCl for 6 h with a solid:liquid ratio = 1:25 was chosen as the initial solution for HDES extraction. The results correlate with those obtained previously by other researchers for LIBs with similar compositions [27,28]. It can be concluded that high-efficiency leaching of valuable elements from end-of-life batteries by hydrochloric acid solutions is possible even without the use of high temperatures, high HCl concentrations, or additional reagents.

3.2. Solvent Extraction

In this work, extraction is used for the stepwise separation of metals from solutions obtained by leaching three different LIB types. The discussion of the experiment, the selection of optimal conditions, and the separation schemes that were developed are presented in the sections below. Previously [24,25], we provided a detailed description of the synthesis, characterization, and application of Aliquat 336-, D2EHPA- and menthol-based HDESs as extractants for isolating metal ions contained in major amounts in LIBs from model solutions. Here we present the results of a study with real solutions after hydrochloric acid leaching.

3.2.1. LIB 1

Extraction of Cu(II). The concentration of HCl is one of the main parameters affecting the efficiency of metal ion extraction [29]. It was suggested to use HDES based on Aliquat 336 to purify leaching solutions from Cu(II) ions.

It is known that Cu²⁺ ions form stable anionic complexes such as [CuCl₂], [CuCl₃]⁻, [CuCl₄]²⁻, [CuCl₅]³⁻, etc., in hydrochloric acid solutions. They can be extracted from solutions by an anion-exchange extractant [30]. It was found previously, using a model solution of copper(II) chloride, that copper was extracted by HDES 1 at an HCl concentration of 0.1–5 M.

Figure 4 shows the results of experiments on the extraction of the major metal ions with HDES 1 at variable volume ratios of the aqueous and organic phases. At the first stage, with O/A = 1/1, Cu(II) is extracted into the HDES 1 phase by 38.7% in one extraction step. At the same time, the other metals are not extracted into the HDES phase. The extraction ratio of the remaining metal ions is <5%. To ensure complete extraction of Cu(II) from the leaching solution, cross-current extraction on a cascade of mixers/settlers was implemented. Cu(II) was extracted for 99.24% in 5 extraction steps.

Water was used for the re-extraction of Cu(II) from the HDES phase. The re-extraction ratio was 98.47% in 6 steps.

Extraction of Co(II). Once the leaching solution was purified from Cu(II) ions, the extraction of Li(I), Co(II) and Al(III) ions as a function of the HCl concentration in the aqueous phase (from 3 to 5 M) at various phase volume ratios was studied at the second stage (Figure 5). Almost in all the cases, the extraction of Al(III) was below 1%. The extraction ratio of Co(II) tends to increase with the concentration of hydrochloric acid, which is associated with an increase in the fraction of cobalt anionic complexes in the aqueous solution. Lithium extraction demonstrates a similar behavior.

The most suitable conditions for the extraction of Co(II) from the leaching solution are 4 M HCl, O/A = 1/1. The concentration of HCl in the raffinate after the first extraction step was increased to 4 M. To reach a complete extraction of cobalt(II) into the HDES phase (98.77%), nine-step cross-current extraction was implemented on a cascade of mixers/settlers.

Co(II) was re-extracted for 99% with water from the HDES phase in one step. After extraction, the resulting raffinate contained Li(I) and Al(III) ions.

Extraction of Al(III). The isolation of Li(I) and Al(III) can be performed by quantitative extraction of Al(III) in two steps by leaching from an acid solution (pH = 2.5) with an HDES based on D2EHPA and menthol (HDES 2). Al(III) was re-extracted for 99% with 1 M HCl solution from the HDES phase.

Based on the above data, a scheme for the treatment of LIB 1 cathode material containing Li(I), Al(III), Co(II) and Cu(II) was suggested (Figure 6). The use of HDESs as extractants in the scheme of metal separation is beneficial due to their stability in acid and alkaline media, high efficiency, and the possibility to abandon toxic organic solvents, thus creating a more environmentally friendly process.

3.2.2. LIB 2

Extraction of Fe(III). It was shown previously for the model leaching solution that Fe(III) was extracted by HDES 1 in a wide range of HCl concentrations from 0.01 to 5 M. Since the leaching solution contained 0.45 M HCl, no additional injection of HCl was required for the extraction of iron ions. Figure 7 shows the results of a series of experiments at a varying volume ratio of the phases to achieve the maximum Fe(III) extraction ratios. High extraction ratios were observed at the O/A = 1/5 phase ratio, but other metal ions were also extracted with an extraction ratio of ~10%. At O/A = 1/1, the extraction of Li(I), Co(II), Ni(II) and Al(III) into the HDES phase decreased significantly, which favored the selective extraction of iron(III) from the leaching solution. In the cross-current extraction mode on a cascade of mixers/settlers, Fe(III) was extracted for 99.47% in 4 stages. Iron(III) was re-extracted with water.

Extraction of Co(II). Further, the extraction of Co(II) with HDES 1 from the leaching solution was studied as a function of hydrochloric acid concentration in the range from 3 to 5 M (Figure 8). It can be seen that the cobalt(II) extraction ratio increases with increasing acid concentration in the leaching solution to reach 72.91%, which is due to an increase in the fraction of anionic complexes that are extracted by the Aliquat 336-based HDES. However, the extraction of Al(III) from 5 M HCl solution begins with an extraction ratio of 82.63%, which correlates with the values obtained in the model solution [31]. The extraction of Co(II) from the leaching solution was carried out from a 4 M HCl solution. In 7 steps, Co(II) ions were extracted for 99.11% in the cross-current extraction mode. Re-extraction with water was carried out for 98.99% in 6 steps.

Extraction of Al(III). Al(III) extraction with HDES 2 was performed under the same conditions as in the case of LIB 1. The extraction rate in the cross-current extraction mode was 99.54% in 6 steps.

The leaching solution remaining after Al(III) extraction contains Li(I) and Ni(II), which can be separated by Ni(II) precipitation.

Based on the above data, a scheme for the treatment of LIB 2 cathode material containing Li(I), Al(III), Co(II), Ni(II) and Fe(III) was suggested (Figure 9).

3.2.3. LIB 3

Purification of Li(I) from Al(III) and Fe(II). Al(III) and Fe(II) were extracted from the LIB 3 leaching solution with HDES 2 at pH = 2.5. The extraction rate in 4 extraction stages was 99.75 and 99.98% for Fe(II) and Al(III), respectively, in the cross-current extraction mode. In this case, Li(I) was not extracted into the HDES phase.

Based on the above data, a scheme for the treatment of LIB 3 cathode material containing Li(I), Al(III) and Fe(II) was suggested (Figure 10).

Re-extraction of Fe(II) and Al(III) ions was carried out with a 4 M HCl solution. The extraction of Fe(II) and Al(III) ions into the aqueous phase was 99.5% and 46.82%, respectively, in three stages. After that, the metal ions were extracted and separated by precipitation. The residual Al(III) ions were also re-extracted from the HDES phase by cross-current extraction with 4 M HCl solution.

After re-extraction of the metals, the HDESs can be reused as extractants, while the metals can be recovered by precipitation or electrolysis [32,33]. Based on the data obtained and the schemes presented above, it may be concluded that it is possible to extract and separate the main elements contained in LIBs with various compositions, which would not only reduce the environmental impact but also return the metals to the production cycle to make new batteries.

4. Conclusions

Recycling end-of-life LIBs that are a secondary resource of valuable metals has obvious advantages from economical and environmental points of view. In this work, a combined approach to the recycling of the cathode material of end-of-life LIBs by hydrochloric acid leaching followed by metal extraction with HDESs, namely Aliquat 336/menthol (1:1) and D2EHPA/menthol (1:1), is suggested. Optimal conditions for hydrochloric acid leaching of three widely known types of LIBs are proposed.

Experimental data on the extraction of metal ions from the real leaching solution using HDES are presented for the first time. During the work, a series of experiments were carried out, the results of which showed the possibility of separation of valuable elements (Li, Co, Al, Ni) by investigated HDES. The extraction rate of valuable individual elements by multistage cross-current extraction is >98%. Principal process flow diagrams implemented on the mixer-sump cascade are proposed. Prospects for industrial applications of HDES to recycle end-of-life LIBs are demonstrated.