Xiaodong Tang,Qiankun Guo,Miaomiao Zhou,Shengwen Zhong*

Faculty of Materials Melallurgy and Chemistry,Jiangxi University of Science and Technology,Ganzhou 341000,China

Keywords:Spent lithium-ion batteries LiNi0.5Co0.2Mn0.3O2 cathode material Direct regeneration

ABSTRACT At present,metal ions from spent lithium-ion batteries are mostly recovered by the acid leaching procedure,which unavoidably introduces potential pollutants to the environment.Therefore,it is necessary to develop more direct and effective green recycling methods.In this research,a method for the direct regeneration of anode materials is reported,which includes the particles size reduction of recovered raw materials by jet milling and ball milling,followed by calcination at high temperature after lithium supplementation.The regenerated LiNi0.5Co0.2Mn0.3O2 single-crystal cathode material possessed a relatively ideal layered structure and a complete surface morphology when the lithium content was n(Ni+Co+Mn):n(Li)=1:1.10 at a sintering temperature of 920°C,and a sintering time of 12 h.The first discharge specific capacity was 154.87 mA·h·g-1 between 2.75 V and 4.2 V,with a capacity retention rate of 90% after 100 cycles.

1.Introduction

Lithium-ion batteries have gained wide applications in computers,communication equipment,electronics,electric vehicles,etc.because of their high capacity,high energy density and excellent cycle performance [1-5].However,after several charging and discharging cycles,lithium-ion batteries eventually lose their efficiencies and ultimately fail [6],so generally lithium-ion batteries only last for 1-3 years [7].Despite the great importance of Ni,Co and other metals contained in lithium ion batteries [8-10],waste lithium ion batteries are potential pollutants to the environment as they contain large numbers of toxic metals and organic electrolytes[11-14].Therefore,the efficient recycling of waste lithium ion batteries is an important direction for the sustainable development of the battery industry [15].

At present,there are two main ways to recycle and regenerate the anode materials in failed li-ion batteries.One is the use of traditional metallurgical methods of recovery,mainly hydro-,pyro-,and biological leaching[16].For example,Yanget al.[17]proposed an effective method for recovering valuable metals.First,Selective precipitation of nickel and cobalt with sodium ethyl-xanthate.Then,Ni and Co in precipitate were separated by ammonia leaching.Finally,Li and Mn left in leaching liquor were separated by solvent separation with D2EHPA.Zhouet al.[18] proposed for the leaching spent LiCoO2material,recovery Al foil and separation cathode materials in one step by direct electrolytic leaching of spent electrode plate.Besides,she also proposed a new process for recycling waste LiCoO2cathode materials.The new process includes ultrasonic enhanced leaching and spray drying for onestep regeneration of LiCoO2materials [19].However,this method has obvious disadvantages,such as the possibility of generating secondary pollutants from the acids and bases used for the recycling process,and the loss of some metals [20,21].In view of this,Physical direct repair and regeneration method is proposed for the waste cathode materials of lithium-ion battery whose structures are not completely broken.The process is simple,and with relatively low risk of environmental pollution.For example,Zhouet al.[22] reapplied a green and simple technology to LiNi0.5Co0.2-Mn0.3O2anode material by adding LiAc solid phase to make up for the absence of lithium in the material lattice,and the cracked and broken particles were observed to disappear,the LiMn2O4/NiO was completely removed,and most of the LiF/Li2CO3were consumed.Furthermore,the specific discharge capacity was 147 mA·h·g-1at 1 C rate,and after 100 cycles,it still had a specific discharge capacity of 131 mA·h·g-1,and the capacity retention rate of 89.12%.Yanget al.[23] evenly mixed failed LiNi0.6Co0.2Mn0.2O2and Li2CO3by ball-milling,and the mixture was sintered at 800 °C to regenerate LiNi0.6Co0.2Mn0.2O2.It was observed thatn(Li)/n(Ni+Co+Mn)=1.05 was the best process condition.The electrochemical test showed that the specific capacity was 173.8 mA·h·g-1at 0.2 C multiplier,and the retention capacity rate was 99.5% after 55 cycles.The results showed that lithium source could be added to a lithium-less nickel-cobalt-manganese material,and the lithium refilled the vacancy resulting from the loss of active lithium by high temperature sintering crystal state,and the in-situ reversible repair was achieved.The material properties were found to be in close similarities to the original fresh material,thus validating the simplicity and feasibility of the method.

Herein,this study mainly investigates the recovery and regeneration of LiNi0.5Co0.2Mn0.3O2cathode material from spent lithiumion batteries.The particles were ground to a certain size after pretreatment,then fully mixed with a lithium source and sintered at a high temperature to determine the best process conditions,and obtain a new LiNi0.5Co0.2Mn0.3O2cathode material.The phase structure and surface morphology of the LiNi0.5Co0.2Mn0.3O2cathode material were determined by XRD and SEM.Finally,the electrochemical properties of the regenerated LiNi0.5Co0.2Mn0.3O2were tested under different conditions.

2.Materials and Methods

2.1.Regeneration of cathode materials

After discharging a soft-packed lithium-ion battery,it was disassembled to obtain the positive electrode material,and the waste from the positive electrode was heated in air at 550 °C and 600°C for 5 hours.In order to confirm the complete decomposition of the conductive agent,acetylene black,in the heat-treated positive electrode material,2 g of the heat-treated waste was added to excess 2 mol·L-1H2SO4,while the fluorine content of the heattreated positive electrode material was tested to confirm the complete decomposition of the binder,polyvinylidene fluoride(PVDF).Therefore,the recovered raw material calcined at 550 °C for 5 hours is denoted as H-NCM.The average particle size of the raw material (H-NCM) was about 2 μm after jet milling and ball milling.It was then mixed evenly with an appropriate amount of Li2CO3to replenish the lost lithium ions.The resulting mixture was transferred to a porcelain boat and placed in a muffle furnace for segmental calcination (the first stage involves the even penetration of Li into the material at low temperatures;while the second stage involves the solid state reaction at higher temperatures)to obtain the regenerated cathode materials.This article explores the effects of different experimental conditions on the properties of recycled materials,such as different amounts of supplementary lithium (n(Ni+Co+Mn):n(Li)=1:1.05,1:1.10,1:1.10),different sintering temperatures (720,900,920,and 940 °C) and different sintering times (6 h,9 h,12 h,15 h).

2.2.Characterization

Inductively coupled plasma-optical emission spectrometry(ICP-OES) was used to determine the elemental content of the recycled raw materials and the resulting materials after lithium supplementation.Thermogravimetric analysis (TG) was used to determine the temperature required for the decomposition of the binders and conductive agents in the raw material.X-ray diffraction (XRD) was used to analyze the phase of LiNi0.5Co0.2Mn0.3O2before and after regeneration,using Cu Kα as the radiation source,continuous 2θ scanning,scanning speed of 10 (°)·min-1,and scanning range from 10°-80°.The surface morphology of the materials before and after regeneration was characterized by a scanning electron microscope (SEM).

2.3.Electrochemical measurement

The regenerated cathode material was assembled into a button cell for electrochemical performance test.The active materials,acetylene black and polyvinylidene fluoride (PVDF,dissolved inN-methyl-2-pyrrolidone reagent) were uniformly mixed in a mass ratio of 90:4:6,coated on aluminum foil,dried at 120 °C and cut into pole piece of required size.Lithium sheet was used as the reference electrode,imported polypropylene microporous film was used as the diaphragm,the electrolyte was 1 mol·L-1LiPF6,and the solvent was EC+DMC (volume ratio 1:1),which was assembled into a button cell in argon atmosphere.The voltage range was 2.75-4.2 V using Neware battery test system.It was formed by constant current discharge at 0.5 C rate.Electrochemical impedance spectroscopy(EIS)was performed on an IVIUM electrochemical workstation.

3.Results and Discussion

3.1.Pretreatment of recycled raw cathode materials

The TG-DTA curve of the recovered raw material as shown in Fig.1 shows the decomposition of its impurities (mainly polyvinylidene fluoride and acetylene black) with temperature changes.It can be seen from the figure that the decomposition was significantly faster from 550°C with obvious drops in quality.Therefore,the raw material was calcined at 550 °C and 600 °C for 5 h.Through chemical analysis,the mass percentage of fluorine was found to reduce massively from 1.32% to 0.06% after calcination at 550 °C for 5 h.This is an indication that the impurities(PVDF and acetylene black) were removed during the heat treatment.

3.2.Characterization of recycled cathode materials

3.2.1.Elemental analysis

Fig.1. TG-DTA curve of the recovered cathode material.

Table 1 Percentage mass content of Li,Ni,Co and Mn in cathode materials

Table 1 lists the mass content of each metal after the pretreatment of the recycled raw material.Compared to the standard,the content of Li was significantly depleted.This may be due to the consumption of some active lithium ions during the formation of the solid electrolyte interphase(SEI),and the failure of the lithium ions to return to the layered structure due to the destruction of some parts of the cathode material structure.The Ni,Co,and Mn contents slightly increased,which were accompanied by the decrease in Li.Therefore,it can be calculated,and the depleted Li in the material can be supplemented by the addition of Li2CO3for sintering.

3.2.2.XRD analysis

Fig.2 is the XRD pattern of recycled materials and commercial materials.The peak position and number of recycled materials are almost consistent with that of commercial materials.But the peak intensity of each peak is lower.Therefore,the structure and properties of the material can be improved by supplementing and adjusting its lithium content under specific sintering system.

Fig.2. The XRD patterns of commercial materials and recycled material.

3.2.3.SEM analysis

Comparative SEM micrographs of commercial LiNi0.5Co0.2Mn0.3-O2and recycled LiNi0.5Co0.2Mn0.3O2are shown in Fig.3.It can be seen from the images that the morphology of the commercial LiNi0.5Co0.2Mn0.3O2was spherical in shape,and the secondary particles were formed by the accumulation and cluster of the primary particles.However,the spherical structures of the recycled cathode material were mostly damaged to some extent.Some primary particles appeared to be scattered and detached from the secondary particles,thus rendering incomplete morphologies.This is possibly due to the expansion and contraction of the LiNi0.5Co0.2Mn0.3O2particles caused by the insertion and extraction of lithium,as the change in volume caused mechanical stress within the particles.The uncompensated anisotropic stress between adjacent primary particles in a secondary particle eventually leads to particle breakage [24].Hence,it becomes difficult to restore the original morphology.Therefore,jet milling and ball milling were employed in this study to break all the secondary particles and ensure even dispersion of the primary particles.By these methods,the structures of the particles can be adjusted to achieve the intended recycling.

3.3.Characterization of LiNi0.5Co0.2Mn0.3O2 cathode materials

3.3.1.Elemental analysis

Table 2 lists the pretreated recycled raw materials and different mixing ratios of Li2CO3(n(Ni+Co+Mn):n(Li)=1:1.05,1:1.10,1:1.15)sintered at 920°C for 12 hours.The Li content of the calcined materials are close to that of the standard LiNi0.5Co0.2Mn0.3O2material,which is presumably related to the addition of Li2CO3.To measure the free Li+content of the sintered material,appropriate amounts of the materials were dissolved and theconcentration of the solutions were measured.The obtained values were 0.107%,0.146%,and 0.162% for the different mixing ratios (1:1.05,1:1.10,1:1.15) respectively.These indicate that only a small amount of Li2CO3was attached to the surface of the materials as most of the Li were embedded in the bulk materials.

Fig.3. SEM images of LiNi0.5Co0.2Mn0.3O2.(a) Commercial LiNi0.5Co0.2Mn0.3O2,(b) recycled LiNi0.5Co0.2Mn0.3O2.(c) LiNi0.5Co0.2Mn0.3O2 after jet milling and ball milling.

Table 2 Elemental contents of each sintered material with different Li2CO3 mixing ratios(n(Ni+Co+Mn):n(Li)=1:1.05,1:1.10,1:1.15)

Fig.4. The XRD patterns of cathode materials prepared under different The XRD patterns of cathode materials prepared under different(a)ratios of precursors to Li2CO3,(b)calcination temperatures,(c) calcination times.

Table 3 Cellular parameters of LiNi0.5Mn0.2Co0.3O2 prepared under different conditions

Fig.5. XPS spectra of (a) Ni 2p,(b) Mn 2p,(c) Li 1s for regenerated material.

Fig.6. TEM images of the regenerated cathode material.

3.3.2.XRD analysis

There are many factors that affect the properties of materials during lithium supplementation and sintering.To determine the appropriate amount of lithium required,the effect of different lithium contents on the structure of LiNi0.5Co0.2Mn0.3O2was investigated.In addition,the effects of calcination temperature and time were studied to obtain the materials with moderate primary particle sizes and better performances.The obtained results are presented in Fig.4 and Table 3.

Compared with the standard card (PDF#87-1562),all the samples were of α-NaFeO2structure(space group:R-3m).The obvious splitting of diffraction peaks (006)/(102) and (108)/(110) in each spectrum indicates that the materials possess good layered structures.The intensity ratio of the(003)/(104)peaks is one of the factors reflecting the degree of cation disordering in cathode materials.WhenI(003)/I(104)＞1.2,it means that the degree of cation mixing is small and the material exhibits a well-ordered layer structure.From Table 3,theI(003)/I(104)value of each sample is greater than 1.2,illustrating their desirable cation mixing.Atc/a＞4.899,materials exhibit better structural stabilities,and the larger the ratio,the better the structural stability of a material[25,26].

Fig.7. SEM images of the cathode material under different conditions.(a),(b)and(c)correspond to calcination temperatures of 940°C,920°C and 900°C,respectively.(d),(e)and (f) correspond to calcination times of 15 h,12 h and 9 h,respectively.

Fig.8. First charge and discharge tests of the cathode material prepared under different conditions.(a) Different calcination temperatures.(b) Different calcination times.

Summarily,in combination with the results of ICP,the optimal addition amount of Li2CO3was determined to ben(Li):n(Ni+Co +Mn)=1:1.10.The optimum temperature and calcination time were also determined to be 920 °C or 940 °C and 12 h or 15 h,respectively.

3.3.3.XPS analysis

XPS was used to studied the valence state of elements in the regenerated material.Fig.5(a) shows the Ni 2p spectrum.The Ni spectra with two major binding energies at 872 eV and 855 eV assigned to Ni 2p3/2and Ni 2p1/2are showed which suggests the co-existence of Ni2+and Ni3+in both samples.As shown in Fig.5(b),two main peaks at 642.3 eV and 653.8 eV were observed,belonging to Mn 2P3/2and Mn 2p1/2respectively,and two main peaks at 644.9 EV,belonging to M-O-Mn respectively,indicating the oxidation state of Mn4+in the sample.Fig.5(c) shows the Li 1s spectrum.The peak around 54.3 eV is characteristic of lithium atoms in the crystalline lattice,while peak around 55.3 eV is ascribed to the Li+in lithium compound.These data are consistent with the common cathode material structure.

3.3.4.TEM analysis

The TEM image of the regenerated cathode material particles(Fig.6) shows a lattice stripe of about 0.47 nm,which fully conforms to the (003) crystal plane.The analyses of ICP,XPS and TEM above prove that the added lithium diffuses into the material structure during sintering.

3.3.5.SEM analysis

Fig.9. The rate and cycling performance of LiNi0.5Co0.2Mn0.3O2 sintered at different conditions.(a)(c)(e) Rate performance curves.(b)(d)(f) Cycling performance curves.

The SEM micrographs of LiNi0.5Co0.2Mn0.3O2cathode material under different conditions are shown in Fig.7.As observed from the images,the particles exhibit spherical-like polyhedron shapes.In addition,the secondary particles appear completely broken,while the primary particles show even dispersions,except for a few particles that have agglomerated.The figure also shows that the material is basically single crystal with its primary particle size in the range of 2-3 μm.Sintering temperature affects the particle sizes of materials.The primary particle size was obviously larger at 940 °C compared to the sizes at other temperatures.Moreover,finer particles are expected for lower calcination temperatures or shorter calcination times,which would be unfavorable to the properties of materials.Therefore,it is necessary to control the sintering temperature and time.

Fig.10. Electrochemical impedance spectra of samples sintered at 920 and 940°C.

3.4.Electrochemical tests

3.4.1.First charge and discharge tests

The first charge and discharge performances of the LiNi0.5Co0.2-Mn0.3O2cathode material prepared under different conditions are shown in Fig.8.With the addition of Li2CO3to the cathode material in the mixing ratio ofn(Ni+Co+Mn):n(Li)=1:1.10 and calcination time of 12 h,the first discharge specific capacities and the first charge and discharge efficiencies were 149.25,154.87 and 153.92 mA·h·g-1,and 86.12%,87.93% and 86.03% at 900,920 and 940 °C,respectively.Whereas,the first discharge specific capacities and the first charge and discharge efficiencies at 1:1.10 mixing ratio and 940 °C were 146.31,154.87 and 152.76 mA·h·g-1,and 84.24%,87.93% and 85.21%,after 9 h,12 h,and 15 h,respectively.Obviously,higher values were obtained at 920°C,12 h,andn(Ni+Co+Mn):n(Li)=1:1.10.The results are in good agreement with the results from XRD and SEM examinations.

3.4.2.Rate and cycling performance

The rate and cycling performances of LiNi0.5Co0.2Mn0.3O2sintered at different conditions are shown in Fig.9.

The cells were charged and discharged at different current densities,and at low rates,there were very little differences between the discharge specific capacities of the materials at 920 °C and 940 °C.However,the materials sintered at 920 °C showed better performances at high rates.In addition,the cycle performance was tested at 0.5 C,and the discharge capacity of the material sintered at 920°C decayed from 152.21 mA·h·g-1to 136.57 mA·h·g-1,while the capacity retention rate was 89.72%.The discharge capacity decayed from 151.01 mA·h·g-1to 132.06 mA·h·g-1,with a capacity retention rate of 87.45% for the material sintered at 940 °C.These mean that the calcination at 920 °C yielded better results.The temperature increment may have encouraged the agglomeration of primary particles towards the formation of larger particles,which ultimately affected the performance of the material.Besides,it can be seen from Fig.9(c) and Fig.9(d) that both rate and cycling performance are the best under the condition of sintering for 12 h.Fig.9(e) and Fig.9(f) show the rate and cycling performance of cathode materials obtained at different ratios of precursor to Li2CO3,obviously,we can see the best performance when the ratios of precursor to Li2CO3is 1:1.10.

3.4.3.Electrochemical impedance spectroscopy tests

Fig.10 shows the Nyquist plots for electrochemical impedance spectroscopy tests of the samples sintered at 920 °C and 940 °C.The impedance spectrum consists of high frequency semicircles representing the interface or membrane impedance and low frequency straight lines representing the lithium ion diffusion impedance [27,28].From the figure,the interface impedance values of the materials obtained at 920 °C and 940 °C are 77 Ωand 94 Ω,respectively.In addition,the results show that the slope of the low frequency line at 920 °C is slightly larger than that at 940 °C,which indicates that the ion diffusion impedance of the material sintered at 920 °C is comparatively smaller,thus suggesting that calcination at 920 °C was more suitable.

3.4.4.Comparison of electrochemical properties with commercial materials

Fig.11 compares the rate and cycling performance of recycled material and commercial materials.The comparison shows that the rate performance of commercial materials is better than that of recycled materials,especially in the case of high rate.The capacity difference is about 10 mA·h·g-1.However,the recycling performance of recycled materials has some advantages,after 100 cycles,the capacity retention rate is higher.Because the recycled materials come from waste lithium-ion batteries,it is also of certain significance from the aspects of environmental protection,energy and economic benefits.

Fig.11. Rate and cycling performance of recycled material and commercial materials.

4.Conclusions

The recovery and regeneration of LiNi0.5Co0.2Mn0.3O2cathode material from spent lithium-ion batteries was investigated,and the following conclusions were drawn:the raw materials were first subjected to jet milling and ball milling,and then the LiNi0.5Co0.2-Mn0.3O2material was improved by high temperature solid state reactions to supplement the depleted lithium.The resulting material possessed an ideal layered structure and single crystal morphology with better primary particle dispersion.The optimal experimental conditions weren(Li)/n(Ni+Co+Mn)=1.1,920 °C and 12 h,and the best electrochemical properties of the materials were obtained under these conditions.The first discharge specific capacity was 154.87 mA·h·g-1with a capacity retention rate of 90%after 100 cycles,while also exhibiting good rate performances.Compared with commercial materials,its rate performance is relatively poor,but its cycle performance is more advantageous.Therefore,the study provides a feasible method for the direct regeneration of recycled cathode materials.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.