Temperature-Correlated Characterization of EoL Lithium Cobalt Oxide Batteries with Microwave-Based Pyrometallurgical Recovery

Abstract

With the increasing volumes of spent lithium-ion batteries from electric vehicles and the concurrent increase in raw materials cost for cathode production, finding effective methods for recycling battery materials has become critically important. This study investigated a pyrometallurgical approach using microwave irradiation to achieve carbothermal reduction of LiCoO₂. FactSage thermodynamic calculations were performed for process simulation and an infrared thermal camera was employed for temperature measurements, allowing the authors to optimize the process parameters to obtain metallic cobalt. Specifically, the research included microwave experiments on mixed black mass samples of anode and cathode materials in different proportions, treated at varying power levels and exposure times under air atmosphere. The effect of the process parameters and therefore of the temperature on microstructure was studied with SEM-EDS and XRD analysis. The feasibility of a wet magnetic separation method between cobalt and lithium compounds formed during the reaction was also evaluated. The results obtained from the final separation process indicated that individual compounds can be obtained at the end of the cycle; moreover, the optimization of time, temperature, and graphite additions during the tests allowed the authors to obtain promising results.

1. Introduction

In a scenario where electromobility and energy storage markets are constantly growing, the mineral demand for use in the manufacturing of lithium-ion batteries (LIBs) is forecasted to grow by at least thirty times by 2040 [1]. With the need for lithium being at the top of the list, followed by graphite, cobalt, and nickel, finding a sustainable and effective recycling strategy is of the utmost importance. The increase in stockpiling of end-of-life batteries and the environmental impact of mining critical raw materials, like cobalt and lithium, has prompted many researchers to investigate the LIB recycling field [2,3,4,5,6].

Current mainstream techniques can be divided into four categories: pyrometallurgical, hydrometallurgical, bioleaching, and mechanochemical processes. Most of them still have strong limitations and require further optimization in order to be implemented in industrial processes.

The high selectivity and low energy consumption of hydrometallurgical processes make them an appealing alternative for companies, although they have notable drawbacks; for example, the use of inorganic acids generates secondary pollution in the form of reactive gas emissions, production of acidic wastewater, and environment contamination [7,8,9].

Pyrometallurgy is a technique that uses high-temperature heating to induce a carbothermic reduction by using a reducing agent like coke or charcoal; the process is also employed for the recovery of valuable metals from spent LIBs, as it activates physico-chemical transformations that lead to the extraction and purification of metals [10,11,12,13,14]. Pyrometallurgical recovery is considered a highly reliable method owing to its high reaction rate, its large treatment capacity, and simplicity of operation. Integrated off-gas treatment equipment can be installed in order to limit gas evolution and production of gas [12].

In recent years, many studies have been conducted on pyrometallurgical recovery of waste LIBs with microwaves, with outstanding results regarding its efficiency and lower energy requirements [15,16,17,18,19]. Its working principle is based on the effect of electromagnetic waves on the dipoles within the molecules of the sample. The alternating electric field causes the rotation/realignment of such dipoles, which can occur trillions of times per second, generating heat due to friction.

In the case of LIBs, microwave heating with the use of a reducing agent (graphite) allows to obtaint a carbothermic reduction that converts the starting transition metal oxides into oxides with a lower valence. Lithium instead reacts with graphite and transforms into lithium carbonates, which can be easily separated from the other products by dissolving them in water. For example, Scaglia et al. found that by treating black mass at 600 W for 5 min, a lithium recovery of 85% was reached [20]. Pyrometallurgical treatment of spent LIBs can also be a helpful pre-treatment before leaching with green solvents (hybrid microwave treatment). A study performed by Pindar et al. [21] investigated the in situ effect of citric acid leaching after MW treatment of mixed cathode oxides. They observed that after 8 min of treatment at 900 W, the extraction efficiency of Li, Co, and Ni was 94, 90, and 98%, respectively. Other works instead demonstrated the high leaching efficiency of L-malic acid on mixed NMC-type batteries after just 4 min of heating at 1000 W [15,22]. In this case organic acids, although weaker than inorganic ones (e.g., HCl, H₂SO₄, HNO₃), act as strong chelating agents, allowing the almost complete recovery of valuable metals from the leaching solution. Other studies proposed post-MW separation of the products by exploiting the magnetic properties of Co and Mn with a wet magnetic separation technique [23,24,25].

Although MW heating technology is currently at an advanced stage of technology, still some gaps in the literature regarding the exact mechanisms with which the spent LIBs undergo an MW-induced carbothermic reaction exist. In particular, a detailed morphology and composition characterization of the reaction products has never been correlated to the temperature reached during the process, although it represents a key step in understanding the oxide breakdown behaviour.

However, temperature monitoring during microwave tests can be a challenging task, as conventional high-temperature thermocouples are usually constituted of a metallic probe, which interacts with electromagnetic waves by creating localized heating and field distortion [26]. Moreover, the time lag when measuring the temperature after the end of microwave exposure might lead to inaccurate results.

This study proposes the investigation of a MW treatment of LCO-type batteries by varying the contents of graphite, followed by wet magnetic separation of the resulting products to obtain metallic cobalt and lithium carbonates. Specifically, MW heating was monitored at different time intervals by means of a thermal camera, with the aim of determining at which temperature the carbothermic reaction was most efficient and which structural and composition changes the samples underwent. In order to gain preliminary insight into the development and expected temperatures of the carbothermic reductions, thermodynamic modelling was performed with thermochemical software FactSage^TM with different graphite contents.

2. Materials and Methods

2.1. Materials

The spent LiCoO₂ batteries employed in this study were derived from end-of-life portable electronic devices. Before dismantling, the batteries were first discharged for 24 h in a 5% NaCl solution. This pre-treatment was required in order to ensure safe handling of the cells, avoiding fire hazards. The outside metallic shell was then removed and the cathode and anode foils were dried for 48 h at 80 °C in order to remove the liquid electrolyte.

Cathode and anode sheets were treated separately by grinding them with a Retsch GmbH Mixer Mill MM 301 (Retsch GmbH, Haan, Germany) at a frequency of 30 Hz for 10–20 s. In order to remove the residual fragments of the current collector foils (Al and Cu) and homogenize the particle size, the obtained powders underwent a series of sieving steps (75, 60, and 45 µm).

2.2. Methods

2.2.1. Thermodynamic Analysis

Thermodynamic calculations were carried out by using FactSage^TM 8.3 software, with the Equilib module and FactPS, FT oxid databases. The thermodynamic data for modelling LiCoO₂ at equilibrium condition were taken from the database proposed in the work of Nuraeni et al. [27], simulating the carbothermic reduction of the transition metal oxide with different amounts of graphite additions in air-like atmosphere. For the calculations, only solid states were considered.

2.2.2. Experimental Design

Microwave experiments were carried out with a commercial Panasonic NN-GD351W microwave (Panasonic, Milan, Italy) oven operating in air atmosphere; the powder samples were placed inside alumina crucibles, together with SiC susceptors to aid the transmission of heat.

After performing a preliminary testing campaign, a power level of 950 W was selected and kept constant for the MW tests reported in this article, while the content of graphite and the exposure time were varied.

The temperature at the end of each experiment was measured by means of a FLIR T420 thermal imaging infrared camera, FLIR Systems, Wilsonville, OR, USA.

A TGA-DSC analysis in air atmosphere was performed by heating a specimen characterized by LiCoO₂ and graphite at 5 °C/min from 20 to 1000 °C.

X-ray diffraction spectra were obtained for the virgin material and after the MW experiments using a Bruker AXS D8 Advance (Bruker, Karlsruhe, Germany) diffractometer with Cu K_α radiation operated in Bragg–Brentano configuration at 40 kV and 40 mA. The selected 2θ angle ranged from 15 to 65°, with a scanning step of 0.02° and dwell time of 1 s.

Wet magnetic separation was performed in a 50 mL distilled water solution at a temperature of 80 °C, with a stirring rate of 300 rpm for 30 min.

Morphology and composition analysis was performed with a scanning electron microscope (SEM) (Thermo Fisher Scientific, Waltham, MA, USA) coupled with energy dispersive spectrometry (EDS) (Thermo Fisher Scientific, Waltham, MA, USA).

3. Results

3.1. Thermodynamic Simulations

The carbothermic reduction reaction of LiCoO₂ was simulated by varying the amount of carbon introduced into the system under air-like conditions. In the presence of carbon, thermal decomposition of LiCoO₂ may be possible through one of the three following reactions:

$$\mathrm{L}\mathrm{i}\mathrm{C}\mathrm{o}{\mathrm{O}}_2+0.5\mathrm{C}=\mathrm{C}\mathrm{o}+0.5{\mathrm{L}\mathrm{i}}_2\mathrm{C}{\mathrm{O}}_3+0.25{\mathrm{O}}_2$$

$$\mathrm{L}\mathrm{i}\mathrm{C}\mathrm{o}{\mathrm{O}}_2+0.75\mathrm{C}=\mathrm{C}\mathrm{o}+0.5{\mathrm{L}\mathrm{i}}_2\mathrm{C}{\mathrm{O}}_3+0.25\mathrm{C}{\mathrm{O}}_3$$

$$\mathrm{L}\mathrm{i}\mathrm{C}\mathrm{o}{\mathrm{O}}_2+\mathrm{C}=\mathrm{C}\mathrm{o}+0.5{\mathrm{L}\mathrm{i}}_2\mathrm{C}{\mathrm{O}}_3+0.5\mathrm{C}\mathrm{O}$$

The products of these reactions are metallic cobalt, lithium carbonate, and a gaseous phase, which can either be CO₂ or CO depending on the temperatures and atmosphere [27]. The thermodynamic calculations show that at 50% graphite content, transformation of Co₃O₄ into CoO is predicted at temperatures higher than 900 °C, together with the formation of Li₂CO₃. By increasing the amount of graphite, the temperature of formation of CoO is lowered to below 900 °C, probably due to the higher absorption power of graphite.

3.2. Microwave Tests

Table 1. Process parameters selected for MW testing.

Sample Name	Graphite wt%	Graphite [g]	LCO [g]	Power [W]	Time [s]
50G11	50	0.2	0.2	950	660
60G6	60	0.24	0.16	950	360
80G8	80	0.32	0.08	950	480

reports the data related to the main experiments carried out during this study.

For sample 50G11, thermal camera images were taken after 4, 6, and 8 min of MW treatment, as shown in Figure 1. The temperature rises very rapidly after just 4 min of treatment (569 °C) and after another 2 min it reaches a temperature of 729 °C, at which point the carbothermic reduction takes place. The test was repeated two times, and it was observed that after 6 min of heating, the specimen reached a temperature of 793 ± 32 °C, indicating a good repeatability. The reaction temperature was confirmed by the TGA-DSC results of Figure 2, which exhibits a sharp decrease in the weight linked to an exothermic peak between 650 and 750 °C, caused by the oxidation of carbon that occurs during the carbothermic reduction of LiCoO₂ [28].

The weight of the MW-treated samples was measured after 6 and 11 min of heating, showing that the weight loss amounted to 21 and 40.5%, respectively, in accordance with the TGA results.

For this specimen, the SEM-EDS elemental map was acquired in different sites (Figure 3), which conveyed important information regarding the development of the reaction products after 11 min of heating. In particular, the presence of cobalt throughout the analyzed surface does not match that of oxygen, meaning that the two elements are not necessarily bound together as in the starting LCO active material. Moreover, it can be observed that the structure of the polymeric binder (PVDF) has also changed, as fluorine and carbon can be found separately within the surface, meaning that they were transformed during the treatment.

In Figure 4 the elemental map of specimen 80G8 is shown; the morphology and element distribution are notably different from those observed in sample 50G11. In this case, MW treatment induced the formation of fluorine- and carbon-enriched spherical particles on which the formation of cobalt crystals occurred, although smaller and more limited in size and distribution compared with the ones in sample 50G11. The mass change for this sample was 32%.

Sample 60G6 was heated for 6 min, reaching the following temperatures: 799 °C (2 min), 916 °C (4 min), and 874 °C (6 min). The measured weight loss after MW treatment was calculated to be 55.75%, much higher compared to that of the other specimens.

Figure 5 shows the X-ray diffraction spectra of the specimens. Comparing the spectrum of the untreated LCO material to the ones obtained for the treated specimens, it can be observed how the main diffraction peaks change significantly by increasing the content of graphite during the experiments. In particular, at a content of 50% graphite, the appearance of peaks related to lithium decomposition products like Li₂CO₃ (ICDD 00-022-1141), LiAlO₂ (ICDD 00-038-1464), and LiF (ICDD 00-004-0857) are noticed. Moreover, the peaks related to the starting rhombohedral LiCoO₂ (ICDD 00-050-0653) are substituted by peaks associated with cubic metallic Co (ICDD 00-015-0806) and CoO (ICDD 00-048-1719). For graphite percentages of 60 and 80, the peaks related to the cathodic decomposition are gradually substituted by high-intensity graphite ones; however, the presence of weak LiF and LiAlO₂ peaks can be observed.

Some authors reported that by heat-treating the black mass and reaching a temperature of 650 °C, the phase peak of LiAlO₂ can be detected, suggesting that Al foils can participate in the reaction [28]. Impurities in the electrode materials also participate in the thermal reduction process, probably leading to a significant effect on the phase transition of Li and its water-leaching behaviour.

3.3. Wet Magnetic Separation Tests

After MW treatment, the specimens underwent a wet magnetic separation process, during which the magnetic part obtained from the previous step is separated from the other reduction products, namely lithium-bearing compounds and residuals, by using a magnetic stirrer. At 25 °C, lithium carbonate (Li₂CO₃) is characterized by a high water solubility, equal to 1.29 g/100 mL [29], meaning that by increasing the solution temperature, the compound dissolved more easily. The residual deposit (graphite) was collected through filtration.

The post-wet magnetic separation products were characterized by means of X-ray diffraction, as reported in Figure 6, and SEM-EDS (Figure 7).

4. Discussion

The unmatched presence of cobalt and oxygen in the EDS elemental maps of sample 50G11 (Figure 3) implies that the starting LCO material has reacted during MW heating and the oxide structure has broken down, partially forming metallic cobalt particles and lower valence oxides; in particular, the main characteristic peak of LiCoO₂ at 2θ = 18.96° disappears, substituted by small peaks associated with lithium-bearing reaction products formed by the consumption of graphite. In fact, as a reductant, carbon reduces the lowly soluble Co³⁺ to Co²⁺ in LCO [12], consistent with what is reported in the thermodynamic simulations.

The same can be deduced concerning fluorine and carbon; in fact, during MW treatment the PVDF binder decomposes and its elements participate in the reaction by forming other compounds. In this case, fluorine that is not bound to carbon might have formed LiF, while carbon might have reacted with lithium, forming Li₂CO₃. From a thermodynamic point of view, as long as CO is produced during the reaction, lithium oxides and metallic cobalt can be obtained; CO₂, on the other hand, can interfere with the reaction efficiency, especially in air atmosphere [24].

The cobalt particle morphology in sample 80G8 appears very different from what was observed in the previous sample, probably due to the increased content of graphite that modified the reactions occurring within the black mass. In fact, a variety of crystals nucleated on the surface of the F/C-enriched particles, which the EDS elemental map shows to be made of cobalt, without presence of oxygen. EDS spectra also reveal the presence of phosphorous contamination, probably as a residual of the LiPF₆ electrolyte. The XRD confirms the presence of metallic cobalt as well as CoO, together with weak peaks associated with LiF.

The absence of Li₂CO₃ peaks from the XRD spectra of sample 80G8 may be explained by the thermodynamic conditions prevailing at 900 °C and at higher graphite contents. At high temperatures or with carbon excess, the Boudouard equilibrium shifts toward CO formation rather than CO₂, creating a highly reducing environment. Thus, Li₂CO₃ becomes thermodynamically unstable and is decomposed into Li₂O and CO₂. In presence of Al, Li₂O reacts with it and forms LiAlO₂, which is more stable than Li₂CO₃ and insoluble in water.

The diffraction spectrum of sample 60G6 also shows that transformation of LiCoO₂ into CoO and Co took place, together with some small amounts of LiAlO₂.

The presence of such products and the absence of Li₂CO₃ peaks is probably due to the limited duration of the test, which only lasted 6 min. Moreover, the higher amount of C compared with specimen 50G11 probably led to a more rapid temperature increase, at which oxidation of carbon is more favoured.

The cobalt particles recovered after wet magnetic separation of sample 50G11 are shown in Figure 7; they exhibit a coarse porous framework in which impurities like Fe, Al, P, and Mg are present.

The SEM images of Li₂CO₃ show a crystal structure of platelets and spherulites, consistent with the morphology that is usually described in the literature [30,31,32]. In the XRD analysis of Li₂CO₃, a diffraction peak corresponding to Co is observed, likely due to wet magnetic separation residues. This minor contamination reflects the need to further optimize the conditions set for this proof-of-concept study and does not affect the identification of the main Li₂CO₃ phase. However, further investigation is required to determine the exact purity levels of the leachates. High-purity recovery of Li₂CO₃ would in fact allow for its use as a precursor in the manufacturing of new cathode materials [33].

5. Conclusions

The aim of this study was to gain a deeper understanding of the mechanisms occurring during microwave reduction of spent lithium cobalt oxide batteries.

It was observed that slight variations in the graphite content had a significant effect on the reaction products, especially on the formation of valuable lithium-bearing compounds. With a content of 50% graphite and the heating aid of a SiC susceptor, microwave treatment of the samples for 8 min led to the formation of lower-valence Co oxides and metallic Co, together with Li₂CO₃, LiF, and LiAlO₂; however, by gradually rising the graphite to 80%, the formation of Li₂CO₃ was hindered, indicating that the reaction was slowed down, probably owing to the more favourable reaction of carbon with oxygen to form CO₂. The morphology of the reacted samples also appeared inherently different by varying the graphite contents, which can also be attributed to the way the reaction develops with higher carbon contents. In general, the material seemed to have good microwave absorption properties.

The temperatures reached to obtain the formation of Co and Li₂CO₃ were relatively low, around 700–900 °C, demonstrating that microwave heating leads to efficient reactions in comparatively lower times with respect to conventional heating mechanisms.

Overall, microwave technology seems to be a promising technique for recycling spent lithium-ion batteries, although further improvements must be made to deal with off-gas production and large volumes of waste. Moreover, further investigation is required in order to determine the purity of the reaction products for possible re-integration in the manufacturing of new lithium-ion batteries.