Structural Restoration of Degraded LiFePO 4 Cathode with Enhanced Kinetics Using Residual Lithium in Spent Graphite Anodes

Open AccessCCS ChemistryRESEARCH ARTICLES16 Jun 2022Structural Restoration of Degraded LiFePO4 Cathode with Enhanced Kinetics Using Residual Lithium in Spent Graphite Anodes Min Fan, Xin Chang, Xin-Hai Meng, Chao-Fan Gu, Chao-Hui Zhang, Qinghai Meng, Li-Jun Wan and Yu-Guo Guo Min Fan CAS Key Laboratory of Molecular Nanostructure and Nanotechnology, CAS Research/Education Center for Excellence in Molecular Sciences, Beijing National Laboratory for Molecular Sciences (BNLMS), Institute of Chemistry, Chinese Academy of Sciences (CAS), Beijing 100190 University of Chinese Academy of Sciences, Beijing 100190 , Xin Chang CAS Key Laboratory of Molecular Nanostructure and Nanotechnology, CAS Research/Education Center for Excellence in Molecular Sciences, Beijing National Laboratory for Molecular Sciences (BNLMS), Institute of Chemistry, Chinese Academy of Sciences (CAS), Beijing 100190 University of Chinese Academy of Sciences, Beijing 100190 , Xin-Hai Meng CAS Key Laboratory of Molecular Nanostructure and Nanotechnology, CAS Research/Education Center for Excellence in Molecular Sciences, Beijing National Laboratory for Molecular Sciences (BNLMS), Institute of Chemistry, Chinese Academy of Sciences (CAS), Beijing 100190 University of Chinese Academy of Sciences, Beijing 100190 , Chao-Fan Gu CAS Key Laboratory of Molecular Nanostructure and Nanotechnology, CAS Research/Education Center for Excellence in Molecular Sciences, Beijing National Laboratory for Molecular Sciences (BNLMS), Institute of Chemistry, Chinese Academy of Sciences (CAS), Beijing 100190 University of Chinese Academy of Sciences, Beijing 100190 , Chao-Hui Zhang CAS Key Laboratory of Molecular Nanostructure and Nanotechnology, CAS Research/Education Center for Excellence in Molecular Sciences, Beijing National Laboratory for Molecular Sciences (BNLMS), Institute of Chemistry, Chinese Academy of Sciences (CAS), Beijing 100190 University of Chinese Academy of Sciences, Beijing 100190 , Qinghai Meng *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] CAS Key Laboratory of Molecular Nanostructure and Nanotechnology, CAS Research/Education Center for Excellence in Molecular Sciences, Beijing National Laboratory for Molecular Sciences (BNLMS), Institute of Chemistry, Chinese Academy of Sciences (CAS), Beijing 100190 University of Chinese Academy of Sciences, Beijing 100190 , Li-Jun Wan *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] CAS Key Laboratory of Molecular Nanostructure and Nanotechnology, CAS Research/Education Center for Excellence in Molecular Sciences, Beijing National Laboratory for Molecular Sciences (BNLMS), Institute of Chemistry, Chinese Academy of Sciences (CAS), Beijing 100190 University of Chinese Academy of Sciences, Beijing 100190 and Yu-Guo Guo *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] CAS Key Laboratory of Molecular Nanostructure and Nanotechnology, CAS Research/Education Center for Excellence in Molecular Sciences, Beijing National Laboratory for Molecular Sciences (BNLMS), Institute of Chemistry, Chinese Academy of Sciences (CAS), Beijing 100190 University of Chinese Academy of Sciences, Beijing 100190 https://doi.org/10.31635/ccschem.022.202201996 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Enormous LiFePO4 (LFP)/graphite batteries retired from the market need urgent rational disposal and reutilization based on the degradation analysis of the evolutional mechanism for electrodes. Typically, Li inventory loss is one of the main reasons for the degradation of LFP-based batteries. The reduced portion of lithium in a cathode is inevitably consumed to form solid electrolyte interphase or trapped in the anode. Herein, we propose a comprehensive strategy for battery recycling and conduct the work by simply regenerating the degraded LFP materials directly with the extracted lithium compounds from spent anodes. Moreover, inter-particle three-dimensional (3D) conductive networks are built via an in situ carbonization to reinforce the electronic conductivity of regenerated cathodes. An improved electrochemical performance was achieved in the regenerated LFP materials even compared with the pristine LFP. This integrated recycling strategy not only brings more added value to the recycled materials by leveraging the recycling process but also aims to apply the concept of "treating waste with waste" and spur innovations in battery recycling technologies in the future. Download figure Download PowerPoint Introduction In response to the call to develop a low-carbon economy, multiple technologies encompassing energy transition strategies are continuously evolving.1–3 Among them, lithium-ion batteries (LIBs) are commonly used for portable electronics and electric vehicles and are growing in popularity for grid-scale energy storage systems and aerospace applications.4,5 Despite the expanding market of LIBs, the limitation of their lifespan subject a rising number of retired LIBs to tricky problems such as the dramatic increase of anxiety in accumulation and disposal. From an economic and environmental point of view, there is an urgent call to spent LIB recycling, which is the last yet critical piece of the jigsaw puzzle to ensure the long-term economic sustainability of LIBs.6–8 Traditional pyro-/hydrometallurgical methods have been proved effective in extracting the valuable metals in the cathodes, which are generally considered high-quality minerals of urban mines.9–11 However, most of them are far from meeting the criterion of sustainability, which has become a burning hot topic in the spent LIBs recycling recently.12–16 Many efforts have been made to open a new path for recycling spent LIBs in recent years. Direct regeneration, an alternative to the traditional strategies with a prominent edge in energy consumption and pollution prevention over the former steps, can directly achieve closed-loop reuse of the active materials without extraction of mineral elements and resynthesis, which is especially suitable for the low-cost cathodes. Though the methods for direct regeneration have evolved from solid-state synthesis17,18 to hydrothermal relithiation19–21 and electrochemical relithiation,22 the regenerated cathode obtained in direct regeneration usually suffers from poor performance, consistency, and the market acceptance. Only if the performance of regenerated materials is equal or superior to that of pristine materials can they win market acceptance.23 Beyond the cost reduction and pollution control, sustainability also lies in the resource reutilization of the entire process. Given that the cathode occupies the largest share of cost among all the components in LIBs, it is a major concern to researchers or recyclers at the moment.24 It is estimated that utilizing recovered cathode materials could cut down the total cost of LIBs by over 20%.25 By contrast, other components with low cost and an abundance of raw materials are usually discarded.26 From this point, more potential economic profits can be realized if other components beyond cathode materials in spent LIBs are reused properly.27,28 As shown in Supporting Information Figure S1, spent LIBs which have undergone a long-term cycling exhibit remarkable changes in both cathode and anode sides compared with their pristine state. Because of irreversible lithium-ion consumption, increasing internal resistance, and damaged structure, the number of lithium ions that can eventually return to the lattice structure of the cathode is decreasing, resulting in a gradual decrease of capacity and, ultimately, failure of the battery.29 In this sense, the lithium content in cathode material reduces significantly with performance degradation, while that in anode material increases due to the growth of solid electrolyte interphase (SEI),30,31 dead lithium,32,33 and accumulation of unstripped Li.34 Herein, based on challenges in current recycling processes, we designed a comprehensive strategy by reutilizing residual lithium in spent graphite anodes for the direct regeneration of degraded LiFePO4 (D-LFP). As illustrated in Supporting Information Figure S2, cathode and anode strips are separated from the spent batteries after complete discharge and disassembly. Lithium in the graphite is extracted by pure water to form a lithium-rich solution, which is then assisted by introducing a reducing agent to directly relithiate the spent LFP electrode. It is worth noting that the alkalinity of the solution also helps separate the Al foil from the electrode,18 while the binder and conductive additives are retained with LFP particles. Subsequently, short annealing in Ar is adopted to further repair the structure of the relithiated material, and concurrently in situ carbonization of the binder together with the original conductive additives help build a three-dimensional (3D) conductive network around the regenerated LFP particles, enabling faster charge transfer kinetics; thus, exceeding the performance of virgin materials. Our recycling strategy provides new ideas for efficient resource utilization and high-performance direct regeneration, which has excellent guiding significance in the following recycling innovations of the spent LIBs. Experimental Methods LiFePO4/graphite pouch cell cycling and calculation of the theoretical Li content in the spent anode The pouch cell with LFP as cathode and graphite as the anode (3.08 Ah, n/p = 1.17) was purchased from Beijing IAmetal New Energy Technology Co., Ltd. and experienced 2010 cycles at 1 C (1 C = 170 mAh g−1) in a voltage range of 2.5–3.65 V using Neware battery tester (CT-4008-5V12A-DB-F, Neware, Shenzhen, China). After fully discharged to 2 V at 0.1 C, the degraded cathode and anode electrodes were gained by manual disassembly and solvent rinse with dimethyl carbonate (DMC), respectively. We assumed that the Li loss is caused by the Li trapped in the graphite anode that cannot move back to the cathode. Thus, the theoretical Li mass ( m Li ) in this graphite can be calculated from the following equation: m Li = C f − C i 26.8 × M Li (1)where C f , C i , and M Li represent the final discharge capacity, initial charging capacity, and atomic mass of Li, respectively. The total mass of the anode ( m a ; including graphite, binder, and conductive additives) can be obtained from the following equation: m a = C 0 370 × w g (2)where C 0 is the designed anode capacity and w g is the mass fraction of the graphite in the anode. Thus, the theoretical Li content in a degraded anode is the mass fraction of Li in the total mass of Li and anodes. Testing process of the specific Li content in the degraded anodes An appropriate amount of spent anode powder scraped from the electrode was added to a 2% HNO3 solution (2 mL BV-3 nitric acid in 100 mL deionized water) with lithium content in the range of the required concentrations. After completing the reaction, the mixture was filtrated, and the filtrate was analyzed by inductively coupled plasma mass spectrometry (ICP-MS) for quantitative estimation of Li. Preparation of the solution for relithiation Based on the Li content in the degraded graphite (D-graphite) acquired from ICP-MS and the process parameters of electrode preparation, quantitative spent anode electrodes were added to the pure water to prepare 0.2 M Li+ solution. After reacting for 2 h at room temperature, the mixture was filtrated to collect the solution. Quantitative ascorbic acid (0.08 M) was added to the solution to create the reductant environment. Solution relithiation and regeneration of the D-LFP To accelerate mass transfer, the spent LFP sheets were cut into 2 × 2 cm square pieces. Then the pieces were added to the aforementioned solution with a specific solid–liquid ratio of 10 g L−1. Here, the mass of the solid was considered as the mass of LFP and calculated from the process parameters of the pouch cell. Then the D-LFP particles were adequately relithiated during the quiescence at high temperature at different times, followed by filtration and removal of the Al foils. The relithiated materials were dried at 80 °C for 8 h and then annealed at 600 °C in pure Ar for 2 h. Preparation of Super P/poly(vinylidene fluoride) composite A conductive additive, Super P (SP) and poly(vinylidene fluoride) (PVDF) with the same mass ratio in the LFP electrode of the pouch cell were mixed thoroughly in N-methyl-2-pyrrolidone (NMP), followed by solvent evaporation. Then the as-obtained mixture was calcinated in Ar at 600 °C for 2 h. Materials characterization The composition and structure of the materials were characterized by an X-ray diffractometer (D8 Bruker, Bruker, Beijing, China) using a Cu Kα radiation source (λ = 1.5418 Å). The specific ratio of metal elements in different samples was tested by ICP-MS. The morphologies measurements of the samples were examined by scanning electron microscopy (SEM) (Hitachi S4800, Hitachi limited, Japan) operated at 10 kV. The materials were characterized by transmission electron microscopy (TEM) and high-resolution TEM (HRTEM), performed using JEM 2100F, JEOL Ltd., Tokyo, Japan), obtaining the images at 200 kV, and energy-dispersive X-ray spectroscopy (EDS; X-MAX-80T, Oxford Instruments, UK). X-ray photoelectron spectra (XPS) of the samples were tested on ESCALab 250Xi (Thermo Scientific, Shanghai, China) spectrometer based on hydrocarbon C 1s (284.8 eV). Nuclear magnetic resonance (NMR) spectra were collected on a Bruker (AVANCE III 400, Beijing, China) with a 12 kHz rotation speed. The chemical shifts for 7Li in different samples were referenced to LiCl solution. Thermogravimetric (TG) analysis was performed on the Netzsch STA 449 (Netzsch, Germany). Raman spectra (LabRAM HR Evolution, HORIBA, France) were recorded via spectrometer excitation wavelength at 532 nm provided by an argon-ion laser. Electrode preparation and electrochemical measurements To keep the same carbon content, pristine LFP (P-LFP) was first mixed with 4.74 wt % SP, and the mixture was counted as the active materials. The cathode electrodes were prepared by mixing the active material, conductive additives, and PVDF binder with a mass ratio of 80:10:10 in NMP. The as-prepared slurry was then cast on a fresh carbon-coated Al foil dried at 70 °C for 3 h, followed by an 80 °C vacuum-dried process for 8 h. Standard CR2032-type coin cells were assembled in an argon-filled glovebox (H2O < 0.1 ppm, O2 < 0.1 ppm) with the cathode, electrolyte (1M LiPF6 in ethylene carbonate (EC) : diethyl carbonate (DEC) : dimethyl carbonate (DMC) = 1∶1∶1), Celgard 2500 as separator and Li metal as anode to evaluate the electrochemical performance. Both cycling and rate properties were examined in a voltage range of 2.5–4.2 V on the BT2000, LAND Electronic Co. Ltd., Wuhan, China at 25 °C. The electrochemical impedance spectroscopy (EIS) was performed on a Princeton Applied Research PARSTAT® MC 1000 multi-channel electrochemical workstation at a frequency range of 10−1 to 105 Hz with an amplitude of 10 mV. Cyclic voltammetry (CV) profiles of the samples were also acquired from the Princeton electrochemical workstation at a scanning rate of 0.1–0.8 mV s−1. Results and Discussion Extraction of residual lithium in spent graphite anode The cycling performance of the LFP/graphite pouch cell is provided in Supporting Information Figure S3a. The final capacity degraded to 1948.1 mAh g−1, equal to 71.1% of the initial capacity. After fully discharging to 2.0 V, the spent cell was chosen for subsequent disposal. As we can see from the Supporting Information Figures S3b and S3c, the spent pouch cell is intact without apparent damage and swelling of its appearance. After disassembly, the separated graphite electrodes show uneven coloration, indicative of the formation of dead Li and the detachment of active materials on edge. To extract the residual lithium in the D-graphite, we first investigated the amount and the form present. ICP-MS was applied to calibrate the specific content of Li in the D-graphite. It turned out that the average content of Li in our D-graphite is 2.779 (±0.050) wt % ( Supporting Information Table S1). It is widely acknowledged that the amount of Li retained in an anode is directly proportional to the lost capacity of the cell. Thus, we calculated the theoretical content of Li based on the capacity loss of this pouch cell and confirmed the experimental result corresponds to the theoretical value of 2.75 wt %. Details of the calculation process are provided in the Experimental Section and Supporting Information Table S2. Since we have no access to the formation data of the pouch cell, we used the designed capacity to estimate the C i . SEM was carried out to further explore the morphological changes of spent graphite. Figures 1a and 1b intuitively exhibit the deteriorating state of graphite electrode microscopically. A thick SEI layer covers the particle, and the dead Li is unevenly distributed as well, conforming to the above-mentioned failure mechanisms for graphite anode. The magnified SEM image shows visible snowflake-like substances coated on the D-graphite particle, resulting in a rough surface. XPS was conducted to determine the precise composition of the SEI. From C 1s high-resolution XPS spectra presented in Figure 1c, the characteristic peaks located at 284.8, 286.7, 288.7, and 290.7 eV are attributed to C–C, C–O/C–H, O–C=O, and C–F/CO3 respectively, in line with common failure models for graphite anode.35,36 After long-time cycling, the graphite characteristic peaks were hardly observed, indicating that the thickness of SEI was beyond the depth of photoelectron penetration. Combined with the F 1s and Li 1s spectra of the D-graphite (Figures 1h and 1i), we concluded that the residual lithium in the D-graphite consisted mainly RCOLi, ROCO2Li, Li2CO3, LiF, LixPFy, LixPOyFz, and dead Li,37 which, typically, could react with water. Figure 1 | (a and b) SEM images and (c) C 1s high-resolution XPS spectra of D-graphite. (d) Optical photo of the extraction of the residual lithium in D-graphite. (e) 7Li NMR spectra of the different solutions. (f) XRD pattern of the solute drying from the extraction. (g) Survey spectra, (h) F 1s, and (i) Li 1s high-resolution XPS spectra of the D-graphite before and after water extraction. Download figure Download PowerPoint On the basis of the above-mentioned results, pure water was adopted in this study to extract the Li in the D-graphite. It is worth noting that the D-graphite in the fully discharged state contains little active Li, endowing it with enough safety to operate in the water. Actually, the reaction between the D-graphite and water in practical was observed as mild and well-controlled. After a complete reaction in water, the solid residues, mainly the graphite and conductive carbon, were removed by simple filtration. The filtrate with a transparent yellow color was obtained (Figure 1d) and implemented in the subsequent solution relithiation of D-LFP. The yellow color results from the decomposition of the water-soluble binder molecules. To investigate the chemical environment of Li+ in the filtrate, NMR was applied to compare the different Li compounds dissolved in water.38 The 7Li NMR spectra of the various solutions show no obvious differences, indicating that the Li+ in the filtrate shared a similar chemical environment with that in Li2CO3, LiOH, and LiCl solutions (Figure 1e). That is, in the filtration. Li was present as free ions (Li+) as expected.39 After evaporating the water, the residual solute was characterized by X-ray diffraction (XRD) to explore its composition. Figure 1f shows that the pattern of the solute is consistent with that of Li2CO3 without signals of other impurities, while a in the range of an in the the solute of the filtrate was a composite of Li2CO3, LiOH, and to the of to into Li2CO3 in the during solvent the was no The utilization of the residual lithium in D-graphite on the of water extraction. Thus, we characterized the graphite before and after water the a between rough D-graphite with SEI and graphite particles with ( Supporting Information Figures and was to further investigate the of the A of nm for is observed compared with nm for D-graphite ( Supporting Information Figures and which is in with a The of the with the confirmed that water the and extracted the Li compounds that in the D-graphite. Survey spectra in Figure generally the of the peaks attributed to the residual Li compounds after water extraction. Moreover, no in F 1s high-resolution XPS spectra of the the of SEI and no visible in Li 1s spectra of the for both the and the that the residual Li content is the to the removal of the thick SEI the graphite located at eV shows in the C 1s spectra of ( Supporting Information Figure A amount of and on the after Li extraction ( Supporting Information Figures and could be conducted to the into high Solution relithiation of the D-LFP cathode A of the solution relithiation process is shown in Figure Here, ascorbic acid was applied to create a reductant which and thus, the Li an solution was to separate cathode materials from the current during the solution relithiation process by the Al the practical is shown in Supporting Information Figure We first tested the relithiation on D-LFP with low Li As by the XRD in Supporting Information Figure the D-LFP electrodes in the aforementioned pouch cell an excellent relithiation with no obvious the of 80 °C for 2 h. we made a into the relithiation of the D-LFP with high Li Supporting Information Figure the SEM and TEM images of the high D-LFP and the of the and carbon By the TEM images ( Supporting Information Figures and with the XRD pattern of the D-LFP (Figure we that after of the Li, the D-LFP and a pure The of the transition and composition during the relithiation process was further by XRD and ICP-MS. As shown in Figures and by 1 h of a on and after 2 h at 80 a complete composition was The of in the relithiated LFP with a ratio of indicative of of to LFP during the solution However, as the reaction the Li content in the final and then a which could be to the between Li+ and in the solution. Figures and present the SEM images of to show the morphological of the relithiation process. with the D-LFP in Supporting Information Figures and no visible changes were from a range of particles. Since the a binder used for the cathode was in water, the solution relithiation process not the PVDF and conductive additives from the which between the LFP particles. A TEM image (Figure shows that the carbon layer was on the of indicating that the solution relithiation process caused no damage to the carbon image further confirmed the consistent LFP of (Figure which with the XRD spectra of further confirmed a reduction of the of the peaks and peaks with peaks a in the relithiated ( Supporting Information Figure Figure 2 | of the solution relithiation process. XRD and (c) of LFP composition of D-LFP and different relithiation and SEM images and and TEM images of Download figure Download PowerPoint To the of Li+ into the we tested the electrochemical performance of the The a initial charging capacity of mAh and a voltage of V compared with that of with of mAh and V, ( Supporting Information Figure The remarkable charging capacity increase was attributed to more Li+ in the the of the and cycling performance of in Supporting Information Figure that the relithiated materials could work as a new cathode. However, we a gradual increase in specific capacity in the first indicating of In this sense, solution relithiation was to the capacity of the degraded cathodes, the poor of as as PVDF residues, the performance to In situ carbonization assisted regeneration of the To the of and the of PVDF residues, a short annealing was adopted in this work to the regeneration process. After annealing in Ar at 600 °C for 2 h, regenerated LFP as was obtained for further As illustrated in Figure the solution relithiation process the to without the PVDF and between the the annealing a temperature of 600 the decomposition temperature for PVDF was for the complete carbonization of Thus, carbonization of PVDF with the residual built electronic conductive networks among the LFP particles, and a further the structure of the LFP. Figure 3 | of the in situ carbonization process. and SEM images of (d) TEM of the the of and (e) TEM and (f) images of the (g) XRD and (h) high-resolution XPS spectra of and Download figure Download PowerPoint SEM images of (Figures and show morphological changes in with D-LFP ( Supporting Information Figures and and (Figures and The were distributed without visible and the was mixed with LFP particles, indicating the complete removal of A TEM image and of the confirmed the even of and