Heavy liquids for rapid separation of cathode and anode active materials from recycled lithium-ion batteries

Highlights

• Separation of active materials is added to the LIBs physical recycling method.

• Separation of active materials is done using heavy liquids based on Stokes’ law.

• Separation using heavy liquids is rapid and the recovery efficiency is high.

• Separation does not change the active materials morphology and composition.

Abstract

Lithium-ion batteries (LIBs) dominate the industry of rechargeable batteries in recent years due to their advantages, including high energy and power density and relatively long lifespan. Despite these advantages, the disposal of spent LIBs into the ground is harmful to the environment, which needs to be addressed by recycling spent LIBs. The available recycling methods for spent LIBs such as pyrometallurgy and hydrometallurgy focus only on collecting valuable elements from the spent LIBs. The direct physical recycling method may be more economical than the other two methods if the mixed cathode and anode active materials are separated, directly regenerated, and then used to make new LIBs. The first obstacle in this method is the separation of different types of spent active materials that came in the form of micro-sized powder (filter cake). This study aims to separate the mixture of cathode and anode active materials by adopting Stokes' law. The focus is on the physical separation rather than the thermal or chemical separation methods to avoid damaging the morphology and composition of electrode active materials. The proposed mathematical model shows how fast and effectively different electrode materials can be separated by adjusting the heavy liquid density. For validation, several experiments are conducted to separate the cathode active materials (LiCoO₂, LiFePO_4, LiNi_0.8Co_0.15Al_0.05O_2, LiNi_1/3Co_1/3Mn_1/3O₂, and LiMn₂O₄) and the anode active material (Graphite) from each other. Overall, this study shows how rapidly and effectively (high purity) electrode active materials can be separated without damaging the morphology and the composition of electrode active materials.

Introduction

In 1991, SONY Company launched lithium-ion batteries (LIBs) because of their advantages: low self-discharge rate, high capacity to the net weight, and excellent cycle life. Since then, LIB technology has significantly advanced, and different chemistries have been introduced to the market. They are currently broadly used for many applications, such as portable electronic devices, transportation, and renewable storage (Xie et al., 2020). The rapid growth in LIB production will result in a rapid increase of retired LIBs (Wang et al., 2014). The spent materials will effectuate an influx of hazardous waste in landfills if no appropriate action is taken (Shin et al., 2005). The increase in the disposal rate of non-recycled LIBs will harm the environment because of the toxic and harmful ingredients of LIBs (Zhang et al., 2020). Additionally, the waste of valuable materials in spent LIBs necessitates exploration and extraction of new materials, further polluting. A study shows that Li, Ni, Co, and Mn in the cathode of a typical electric vehicle with the cathode chemistry of LiNi_1/3Co_1/3Mn_1/3O₂ is 3.5, 10.9, 10.9, and 9.8 kg, respectively (Diekmann et al., 2017). These high quantities and the potential impacts on the environment emphasize the necessity for recycling retired LIBs. If the spent materials collected from various LIBs are separated and appropriately regenerated, they can reuse to build new LIBs. Recycling and keeping the LIB materials in the closed-loop of build-recycle-rebuild will improve the security of nations that depend on importing these materials to make LIBs.

The current three recycling methods for spent LIBs are pyrometallurgy, hydrometallurgy, and direct physical (Li et al., 2018). Despite the tremendous benefits of these recycling methods, each has some limitations and disadvantages (Zhang et al., 2013). The pyrometallurgy method is the most adopted in recycling companies because pretreatment is unnecessary before recycling. The pyrometallurgy method usually is performed by heating the materials of the retired LIBs above their melting point. The pyrometallurgy target is to recover some valuable metals such as cobalt and nickel from the LIB. Regardless of the noticeable economic benefit, the increasing cost of energy required and risk to the environment because of the residual gases that may exhaust into the atmosphere has limited its use in the wide range of the recycling industries (Chen et al., 2015; B. Wang et al., 2019).

The hydrometallurgy method has a higher recovery efficiency than the pyrometallurgy due to the larger number of recycled elements. Yet not all the components are recovered (Yao et al., 2018). In this method, the first step is a complete discharge of the spent LIBs to avoid any explosion during the process. Second, all types of spent LIBs are crushed in a hydraulic chamber until converted to a powder called filter cake. Then, the filter cake materials dissolved in an aqueous acid solution containing reagents. After that, the resulting metal-rich solution has further treated by solvent extraction to recover the valuable metals. Although this method has higher efficiency than pyrometallurgy, the procedure may generate toxic substances that can harm the environment. This procedure may not be economically efficient because of the massive consumption of chemical reagents (Chagnes and Pospiech, 2013). Both methods mentioned above focus only on recovering the valuable metals from the spent materials such as Mn, Ni, and Co. The remaining elements, in some cases, are disposed of as waste.

The direct physical recycling method of the spent LIBs materials has attracted more attention recently because it does not change the morphology of the electrode active materials and provides the opportunity for their direct regeneration and using to make new LIB. This makes the LIB recycling process more economical. In an experiment (Shi et al., 2018), researchers have regenerated LiNi_1/3Co_1/3Mn_1/3O₂ (NMC-111) from the spent cathode scrap. The regeneration process to fix the material defects was done by hydrothermal treatment followed by short thermal annealing to restore the missing lithium into the spent materials. The results indicated that the discharge capacity of the regenerated NMC is 158.4 mAh/g, which is very close to the capacity of the fresh NMC. Another experiment has conducted to regenerate LiNi_0.8Co_0.15Al_0.05O₂ (NCA) cathode material from retired lithium-ion batteries. They started by separating the cathode materials from the current collector using the acid-leached technique. Then, the cathode material has synthesized by adding Lithium carbonate in a ratio of 1:1.1 and a calcination temperature of 800°C for 15 h. The results have shown that the specific capacity was 248.7 mAh/g for the charge phase, and the discharge specific capacity was 162 mAh/g, which is very close to the performance of the new NCA (Y. Wang et al., 2019).

Because of advantages of the direct physical recycling method ̶ such as the low emission, low consumption of energy, and simplicity ̶compared to the pyrometallurgy and hydrometallurgy recycling methods, several studies have been conducted by researchers to regenerate and make new LIBs from the spent electrode materials (Chen et al., 2016; Ganter et al., 2014; Song et al., 2017). However, most of these studies have focused on one type of LIB's chemistry with known active material. In practice, all LIBs chemistries are recycled together in recycling facilities and the output of these facilities is a filter cake that contains a mixture of different anode and cathode active materials. In fact, it is not practical and economical for recycling facilities to sort the spent LIBs based on their chemistries and manufacturers to achieve a filter cake that contains only one anode and one cathode active material. We can count at least two reasons for this practical limitation. One reason is that the battery manufacturers do not share the details of battery materials data with recycling facilities to enable them to effectively sorting LIBs. The other reason is that the recycling facilities cannot store a LIB with specific chemistry until enough similar LIBs are collected and then recycle together. Because of these two reasons, all spent LIBs are recycled together without consideration of the LIB chemistry and manufacturer once they arrive at the recycling facility. To address this practical limitation, a separation process to separate different filter cake materials should be added to the direct physical method. To date, no study has been done to separate different electrode active materials from the filter cake obtained from recycling facilities. Therefore, this study is focused on introducing a separation process for the filter cake obtained from spent LIBs of varying chemistries. It is shown that the separation of different electrode active materials of spent LIBs can be accomplished rapidly and effectively (high purity) by a physical method with minimum damage to the material morphology and composition. The target of the separation process is to recover each type of the spent LIBs materials (e.g., LFP, LMO, NCA, NMC-111, LCO, and Graphite) and prepare them for the regeneration process. Regardless of the enormous differences in each material's size distributions, as each density value is fixed, it is possible to separate the materials by adopting Stokes' law. The proposed process adopts only a physical separation model based on density differences with no need for any thermal or chemical treatments to avoid any emission of byproduct gasses that are responsible for significant risks to the environment. This proposed process separates the filter cake materials from each other, focusing on cathode active materials that contain the most valuable compounds and elements in spent LIBs.

The separation process is validated by running an experiment where mixture types of cathode materials (LMO, NMC, NCA, LCO, and LFP) and the anode materials (Graphite) were mixed homogeneously before being separated. Finally, the SEM/EDX and XRD analyses were conducted to validate the success of the separation procedure. The results have proven the complete separation of the materials from each other in a short time.