DOI:10.30919/es8d502

Received: 21 Feb 2019
Revised: 24 Mar 2019
Accepted: 26 Mar 2019
Published online: 28 Mar 2019

A Novel Core-shell Structured Nickel-rich Layered Cathode Material for High-energy Lithium-ion Batteries

Hui Tong1, Qijie Zhou1, Bao Zhang1, Xu Wang1, Yingying Yao1, Zhiying Ding2, Hezhang Chen1,3, JunChao Zheng1 and Wanjing Yu1,*

1School of Metallurgy and Environment, Central South University, Changsha 410083, China

2School of Chemistry and Chemical Engineering, Central South University, Changsha 410083, China

3School of Chemistry and Chemical Engineering, Hunan University of Science and Technology, Xiangtan 411201, China

* Corresponding Author.  E-mail address: yuwj2005@163.com (Wanjing Yu)


Abstract:

For enhancing the electrochemical properties of lithium transition-metal oxide, a novel core-shell structured cathode material LiNi0.75Co0.12Mn0.13O2@V2O5 was designed and synthesized. The precursor of the material was made of metal hydroxide in the interior and metal carbonate in the exterior, and with full concentration gradient structure. In this cathode material, the core was the high-Ni material, and the shell was the high-Mn material. The sphere material was coated and penetrated by V2O5 due to the existence of pores. Furthermore, the material was also doped by vanadium element. These were investigated and confirmed by X-ray diffraction, Focused Ion beam, EDS, TEM, etc. The 3 wt% V2O5-coated sample exhibited a remarkable cycling performance, with capacity retention of 86.5% at the current rate of 1 C after 100 cycles. Besides, the rate capability of the V2O5-coated sample was obviously enhanced at high rates (2, 5 and 10 C). The interfacial charge transfer resistance of the material after cycling was obviously decreased by V2O5 coating. The cyclic voltammetry analysis showed that the interfacial polarization of the material was inhibited due to V2O5 coating.


Table of Content

A novel core-shell structured cathode materialLiNi0.75Co0.12Mn0.13O2@V2O5with full concentration gradient was synthesized, and exhibited excellent electrochemical performance.

 


Keywords: Lithium transition-metal oxide; Cathode material; Core-shell structure; Full concentration gradient; Lithium-ion batteries


1. Introduction

Recently, people are seeking for the alternatives to fossil fuels due to the dramatically serious problem of environmental pollution in the world.1-5Lithium ion batteries (LIBs) are widely applied in electric vehicle battery series and energy storage systems due to excellent energy density, performance and long life.6-8 In LIB system, the cathode material restricts the overall performance of the batteries.7,9 The cathode materials, represented by LiCoO2, LiNi1/3Co1/3Mn1/3O2, LiMn2O4 and LiFePO4, are successfully applied to batch production of industrialization. In order to further improve the energy density of the battery, the high nickel cathode material for LIBs is developed.10,11 However, severe capacity attenuation of high nickel cathode materials hinders further application, especially in high rate and long period of cycling. Such limitation is caused by the formation of Ni4+ ions,12 which present a strong oxidizing property and react with the electrolyte.13,14

Y. K. Sun, et al. synthesized core-shell spherical material Li(Ni0.8Co0.1Mn0.1)0.8(Ni0.5Mn0.5)0.2O2 with high-nickel core and low-nickel shell by co-precipitation method.15 Shell material LiNi0.5Mn0.5O2 showed a lower capacity compared with that of the core material. But, the shell material could enhance the cycling stability of the core-shell material. Although the core-shell structure can improve the cycling stability and thermal stability of the material, the junction between the core and the shell were separated after a long period of cycling. So, the cycling stability of the core-shell structure material cannot be guaranteed. Subsequently, a new core-shell structural material with concentration gradient was reported in 2009,16-18 and a full concentration gradient (FCG) material was reported in 2012.19,20 The electrochemical properties of the two kinds of materials were both improved. However, the transition metals still could be dissolved and reacted with the electrolyte. Furthermore, as the promising energy materials, the rate performance should be further improved. At present, there are several routes to solve the problems and improve the properties of the electrode materials: nano-treatment, morphological control, surface modification and composition optimization, etc.21-26 Surface coating is an effective way to slow down the side reactions, inhibit the dissolution of the transition metal ions, and enhance the cycling performance. Some coating materials are usually used, such as Al2O327, TiO228, AlPO429, AlF330, ZrO231, and LiAlO232, etc. Vanadium pentoxide (V2O5) is not only considered as one of the high capacity cathode materials for LIBs,33 but also a coating material with good performance.34,35 In addition, V2O5 possesses a relatively higher lithium-ion diffusivity than most metal oxides, metal phosphates and metal fluorides, as well as electrical conductivity.36

In this work, a novel FCG material LiNi0.75Co0.12Mn0.13O2 with core-shell structure was designed and prepared. The precursor was composed of metal hydroxide in the core and metal carbonate in the shell. The core-shell structured FCG LiNi0.75Co0.12Mn0.13O2 sphere material was prepared through using this special precursor. The shell material contained higher concentration of Mn and lower concentration of Ni compared with those of the FCG material prepared by metal hydroxide precursor, which is beneficial for inhibiting the side reactions and improving the electrochemical performance. The LiNi0.75Co0.12Mn0.13O2@V2O5 cathode material was also successfully synthesized by a wet-chemical method. In this material, the interior and exterior of the sphere particle were coated and penetrated with V2O5. Consequently, the V2O5-coated sample exhibited outstanding electrochemical performance.


2. Experimental section

The precursor of the material was synthesized by co-precipitation method. NiSO4·6H2O and CoSO4·7H2O with molar ratio of 0.9:0.1 were mixed in deionized water, marked as high-nickel solution; NiSO4·6H2O, CoSO4·7H2O and MnSO4·H2O with molar ratio of 0.7:0.1:0.2 were dissolved in deionized water, marked as low-nickel solution. The low-nickel solution was gradually added to the high-nickel solution, and the high-nickel solution was pumped into the reactor at the same time. In the reaction process, Ar was used as a shielding gas and the reaction temperature was about 60 ℃. The NaOH solution and the NH3·H2O solution were fed into the reactor separately. With constantly stirring, Ni0.9Co0.1(OH)2 core was generated first and the precipitation continuously deposited on the core. In the last stage, NaOH solution was changed as Na2CO3 solution, so that metal carbonate shell grew on the surface of the precursor. So, the FCG precursor with core-shell structure was obtained. The precursor and Li2CO3 were mixed with a molar ratio of 1:1.05. The mixture was calcined under oxygen atmosphere at 450 ℃ for 5 h and then at 750 ℃ for 12 h. In the end, core-shell structured FCG LiNi0.75Co0.12Mn0.13O2 cathode material was obtained as pristine material. The V2O5-coated LiNi0.75Co0.12Mn0.13O2 sample was prepared by a wet-chemical method. Vanadium(IV)-oxy acetylacetonate (C10H14O5V) was dissolved in alcohol and mixed with pristine material. It was dispersed by ultrasonic treatment, and then heated until dried by stirring. After that, the powder was calcined in air at 500 ℃ for 6 h. Finally, the 3 wt% V2O5-coated LiNi0.75Co0.12Mn0.13O2 material was obtained.

The crystalline phases of the pristine and V2O5-coated samples were identified with an X-ray powder diffraction (XRD, X’Pert Powder, PANalytical, USA) equipped with Cu-Kα radiation. The morphologies of these samples were observed by scanning electron microscopy (SEM, JSM-6360LV, JEOL, Japan). The microstructures of the samples were examined by using transmission electron microscopy (TEM, Tecnai G2 F20 S-TWIN, FEI, USA). The morphologies of the cross-section of the samples were observed by Focused Ion beam (FIB, HELIOS NanoLab 600i, FEI, USA). The element distributions of samples were characterized with an energy-dispersive X-ray spectrometer (EDS). The chemical valence states were analyzed by X-ray photoelectron spectroscopy (XPS, Escalab 250Xi, Thermo Fisher Scientific, China).

The slurry was prepared by mixing prepared sample, carbon black (super P) and polyvinylidene fluoride in the weight ratio of 8:1:1. After that, the slurry was cast onto aluminum foil for drying in a vacuum oven. The CR2025 coin cell was fabricated in glove box. Metal lithium was used as the negative electrode. PE and PP composite films were used as separator, and 1 mol·L-1 lithium hexafluorophosphate dissolved in DMC: EC: EMC (volume ratio 1:1:1) was used as electrolyte. The electrochemical performances of the material were studied by an automatic galvanostatic charge-discharge unit, a LAND battery cycler (CT2001A, China), 2.5-4.3 V versus Li/Li+ electrode. The cyclic voltammetry (CV) curve was obtained by using an electrochemical workstation (CHI660D, China). The test voltage range was 2.5-4.3 V and the scanning speed was 0.1 mV s-1. Electrochemical impedance spectroscopy (EIS) was detected by the same workstation and the frequency range was from 0.01 Hz to 100 kHz.


3. Results and discussion

The schematic illustration of the preparation process is shown in Scheme 1. Firstly, the FCG precursor made of metal hydroxide was prepared by co-precipitation method. Secondly, metal carbonate was deposited on the surface of metal hydroxide to form a thin shell, in which the concentration of Mn was high. Finally, after wet-chemical and calcination treatments, the surface of the sphere was coated by V2O5 layer, and the interior was penetrated by V2O5 particles.

Scheme 1 Schematic illustration of the preparation process of LiNi Co Mn O @V O cathode material.

 

Fig. 1 (a) Cross-section SEM image of the precursor particle; (b) linear EDS spectrum of the particle section.

The cross-section SEM image and the corresponding EDS spectrum of the precursor material are shown in Fig. 1. It can be seen from the cross-section SEM image (Fig. 1a) that there are many pores inside the particle. Such a structure is designed to provide physical channels for the lithium source and the coating material to penetrate into the interior of the precursor material. Figure 1b shows the molar concentration variations along the red arrow. It can be observed that, from inner core to outer shell, the molar concentration of Ni element decreases slowly in the inner core and decreases fast in the outer shell; the molar concentration of Co element remains almost unchanged; and the molar concentration of Mn element increases slowly in the core, and increases fast in the shell. The reason for the molar concentrations of Ni element decreases fast and Mn element increases fast in the shell, is that MnCO3 deposits easier compared with NiCO3 in the co-precipitation process.

Fig. 2 XRD patterns of pristine and V O -coated samples.

The crystalline structures of the pristine and V2O5-coated samples were investigated by XRD, as shown in Fig. 2. The two samples both belong to the hexagonal α-NaFeO2 structure, and no characteristic peaks of any impurities can be observed in the patterns. The characteristic peak of V2O5 was not observed in the XRD pattern of V2O5-coated sample. The possible reason is that the V2O5 was amorphous or the coating amount of V2O5 is too low to be detected. In the XRD patterns, (006)/(102) and (018)/(110) peaks were significantly split, which indicates that the layered structure of the material is good37. In addition, the lattice parameters of pristine and V2O5-coated samples were investigated and summarized in Table 1. It can be seen from the data, a value has no obvious change before and after V2O5 coating; however, the c value and the unit cell volume increase after V2O5 coating. It is suggested that vanadium element was doped into the crystal structure of the material. The increases of lattice parameter and unit cell volume are advantageous to the lithium ion transport, which could improve the electrochemical properties of the material. The ratios of (003)/(104) of the samples were 1.32 (pristine sample) and 1.53 (V2O5-coated sample), respectively. Generally, the both ratios are greater than 1.2, indicating that the cathode materials possess good layered structures38.

Table 1 Lattice parameters of pristine and 3 wt% V2O5-coated samples.

samples

a (Å)

c (Å)

Volume (Å3)

I003/I104

pristine

2.8918

13.7226

99.38

1.32

V2O5-coated

2.8856

14.2092

102.46

1.53

 

Fig. 3 (a, b) SEM images and (c) EDS spectrum of pristine sample; (d, e) SEM images and (f) EDS spectrum of V O -coated sample.

Figure 3 shows the typical SEM images and EDS spectra of the pristine and V2O5-coated samples. In Fig. 3a, it is seen that the particles of pristine sample were spherical or quasi-spherical, with an average particle size of about 5 μm. It can be seen from Fig. 3b that the primary material particles were 100-300 nm. After the V2O5 coating treatment, the morphology of the sample was not changed, as shown in Fig. 3d and 3e. There was no characteristic peak of V element in the pristine sample (Fig. 3c); however, the characteristic peak of V element can be observed in V2O5-coated sample (Fig. 3f), indicating the existence of V2O5 coating on the particle surface.

TEM images of V2O5-coated sample were also investigated, as shown in Fig. 4. It can be clearly seen that a relatively uniform V2O5 coating layer on the surface of the particle, and the thickness is about 4 nm. Figure 5a shows the cross-section SEM image of V2O5-coated sample. It is found that there are fewer pores compared with those in the precursor. EDS spectrum of the corresponding cross-section part of the sample is shown in Fig. 5b. It is seen that V element is detected in the spectrum. The red arrow in Fig. 5a is the linear scanning direction of EDS spectrum. The distributions of Ni, Co, Mn and V elements in the EDS spectrum are shown in Fig. 5c. It is found that the molar concentration of Ni element decreases slowly in the core and decreases fast in the shell, and the molar concentration of Mn element increases slowly in the core and increases fast in the shell, which is in line with the original material design of high-nickel core and high-manganese shell. Figure 5d-g show the EDS mapping of the four elements, corresponding to the particle section in Fig. 5a. It can be seen that all the elements are evenly distributed in the particle. V element was also found on the section surface of the particle, which further indicates that V2O5 particles penetrate into the interior of the material.

Fig. 4 Typical TEM images of V O -coated sample.

 

Fig. 5 (a) Cross-section SEM image of V O -coated sample particle; (b) EDS spectrum and (c) linear EDS spectrum of the particle section; EDS mappings of (d) Ni, (e) Co, (f) Mn and (g) V elements of the particle section.

The chemical valence states of the elements in pristine and V2O5-coated samples were studied by XPS, as shown in Fig. 6. The XPS spectra of the samples are shown in Fig. 6(a). The V2p peak at 524.81 eV can be observed in Fig. 6(b), corresponding to the characteristic peak of V5+ ions in V2O539. This indicates that the coating on the surface of the pristine material is V2O5. Figure 6(c) shows the oxygen peak of pristine sample before and after V2O5 coating. The oxygen peaks at 531.24 eV (V2O5-coated sample) and 531.29 eV (pristine sample) correspond to the oxygen peak of layered ternary material; 529.41 eV (V2O5-coated sample) corresponds to the oxygen peak of V2O5.40 In Fig. 6(d-f), the peaks of Ni2p3/2, Co2p3/2 and Mn2p3/2 were observed at 854.42, 780.14 and 642.03 eV, respectively, corresponding to Ni2+, Co3+ and Mn4+ in the material,41 proving that the binding energies of the V2O5-coated sample are not changed by V2O5 coating.

The electrochemical properties of the samples were further investigated. Figure 7 shows the first cycle charge-discharge profiles of pristine sample and V2O5-coated sample at the current density of 0.1 C. The initial discharge capacities of the pristine and V2O5-coated sample were 199.1 and 192.7 mAh g-1, respectively. It can be observed that the discharge capacity of V2O5-coated sample was a little lower than that of pristine sample, as V2O5 is inactive in the electrochemical reaction and has no contribution to the capacity. Figure 8(a) shows the rate performances of the pristine and V2O5-coated samples. The discharge capacities of pristine and V2O5-coated samples achieved 138.6 and 144.2 mAh g-1 at 10 C, respectively. It is easily found that V2O5-coated sample exhibited higher discharge capacities especially at high rates (2, 5 and 10 C). The cycling performances of the pristine and V2O5-coated samples were shown in Fig. 8(b). The batteries were charged and discharged for 100 cycles at the current density of 1 C. The discharge capacity of pristine sample was from 188.2 mAh g-1 at first cycle to 141.6 mAh g-1 at 100th cycle, with capacity retention of only 75.2%. However, the discharge capacity of V2O5-coated sample was from 186.1 mAh g-1 at first cycle to 160.9 mAh g-1 at 100th cycle, with the capacity retention of 86.5%, which is much higher than that of pristine sample. The higher capacity retention of V2O5-coated sample is attributed to the coating and penetration of V2O5, which inhibits the damage of the structure and side reactions.