Enhanced Electrochemical Properties of Zr⁴⁺-doped Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ Cathode Material for Lithium-ion Battery at Elevated Temperature

Abstract

The typical co-precipitation method was adopted to synthesized the Li-excess Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) series cathode materials. The influences of Zr⁴⁺ doping modification on the microstructure and micromorphology of Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ cathode materials were studied intensively by the combinations of XRD, SEM, LPS and XPS. Besides, after the doping modification with zirconium ions, Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ cathode demonstrated the lower cation mixing, superior cycling performance and higher rate capacities. Among the four cathode materials, the Li_1.20[Mn_0.50Zr_0.02Ni_0.20Co_0.08]O₂ exhibited the prime electrochemical properties with a capacity retention of 88.7% (201.0 mAh g⁻¹) after 100 cycles at 45 °C and a discharge capacity of 114.7 mAh g⁻¹ at 2 C rate. The EIS results showed that the Zr⁴⁺ doping modification can relieve the thickening of SEI films on the surface of cathode and accelerate the Li⁺ diffusion rate during the charge and discharge process.

Introduction

Recently, the Li-excess Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ (0.6Li₂MnO₃·0.4LiNi_0.50Co_0.20Mn_0.30O₂) materials have attracted much study as cathodes for LIBs owing to the high specific discharge capacity (up to 250 mAh g⁻¹) and less cost^1,2,3. With further research, people have discovered that the Li₂MnO₃ phase (one of the components in the Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂) will be activated and participate in the electrochemical reactions only when the cell voltage is charged to exceed 4.5 V^4,5. However, the high working voltage will cause some drawbacks, such as severe capacity degradation and poor thermal stability, which restrict the practical applications, especially in EV and HEV^6,7. Moreover, when the batteries have been used in the high temperature circumstance, the side reaction between the cathode and electrolyte will be more severe in comparison with the traditional cathode materials, such as LiCoO₂ or LiNi_0.50Co_0.20Mn_0.30O₂ owing to the high working voltage for the Li-excess Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ materials⁸.

To maintain the stability of the cathode at high temperature, considerable effort has been made to resolve the intrinsic defects. For examples, the surface coating modification can effectively protect the cathode from reacting with the electrolyte and retard the thickening of SEI film during cycling. In addition, the suppression of the layered-to-spinel transformation for the Li-excess cathode materials can be obtained by the compact coating layer, leading to the improved electrochemical properties^9,10,11,12. The ion doping modification can stabilize the cathode crystal structure and suppress the layer structural damage^13,14. However, the surface coating modification technology has been complicated and the coating effect demonstrates to be difficult to control, while the ion doping modification shows the easy accessibility and obvious synthetic efficiency¹⁵. Therefore, the ion doping modification has been regarded as the competitive method to enhance the electrochemical properties of the Li-excess Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ materials.

Numerous studies have shown that the Zr⁴⁺ doping modification can effectively enhance the cyclical stability and rate capacity of cathodes. For example, when Zr⁴⁺ was doped into the LiCoO₂ by using the ultrasonic spray pyrolysis method, the LiCo_0.99Zr_0.01O₂ delivered the discharge capacity of 108 mAh g⁻¹ at 1 C in the voltage range of 3.0–4.2 V after 50 cycles, while the un-doped sample rapidly dropped down to 23 mAh g⁻¹ at the same condition¹⁶. When the LiNi_0.5Co_0.2Mn_0.3O₂ was doped modification with Zr⁴⁺ by solid-state method reaction, the Li(Ni_0.5Co_0.2Mn_0.3)_0.09Zr_0.01O₂ demonstrated the much more enhanced rate capability than that of the LiNi_0.5Co_0.2Mn_0.3O₂ by the suppression of electrode polarization¹⁷. While the radius of Zr⁴⁺ (0.072 nm) is larger than those of Mn⁴⁺ (0.053 nm), Ni²⁺ (0.069 nm), Co³⁺ (0.0685 nm) in the transition-metal layer, the Zr⁴⁺ adulteration will expand the diffusion path of Li⁺ insertion/extraction, leading to the improved electrochemical properties. On the other hand, the bond energy of Zr-O has found to be stronger than those of Ni-O Co-O and Mn-O, which will contribute to stabilizing the structure of cathode¹⁸. Based on the above evidence, Zr⁴⁺ will be an attractive doping element to dope into the Li-excess Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ and enhance the electrochemical properties.

In the work, the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) series samples have been synthesized via using carbonate co-precipitation method. And then the combination of microstructural, particle morphology and electrochemical properties has been surveyed to evaluate the influence of different Zr⁴⁺ doping contents into Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ cathode.

Experimental

The Li-excess Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) series cathode materials were synthesized via using the carbonate co-precipitation method to synthesize the carbonate precursors, followed by sintering with LiOH·H₂O powder at high temperature to obtain the cathode materials. The typical synthesis route has been shown as follows: (1) The stoichiometric amounts of MnSO₄·H₂O, NiSO₄·6H₂O, CoSO₄·7H₂O and Zr(NO₃)₄·5H₂O were dissolved in distilled water to obtain a transparent solution; (2) Then the appropriate amount of NH₃·H₂O, as chelating agent and Na₂CO₃, as precipitant, were dropped into the above solution to make the metal ions deposit uniformly; (3) The acquired [Mn_0.52−xZr _x Ni_0.20Co_0.08](CO₃)_0.80 precursors were segregated, washed with deionized water until the impurities eliminate completely; (4) Then the stoichiometric amount of [Mn_0.52−xZr _x Ni_0.20Co_0.08](CO₃)_0.80 precursors and an excess 3 wt.% amount of LiOH·H₂O powder were mixed uniformly, followed by pre-heated at 500 °C for 6 h and finally calcined at 950 °C for 12 h in tube furnace to acquire the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples.

To investigate the influence of the Zr⁴⁺ doping on the crystal structure of Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂, the XRD measurement were carried out by using Rigaku RINT2400 X-ray diffractometer with Cu Kα radiation in the 10° ≤ 2θ ≤ 80°, accompanied by a step size of 0.02° and a count time of 10.0s. Rietveld refinement of the cathode powder diffraction patterns were performed by using the GSAS/EXPGUI program. The morphologies of Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) were observed by using scanning electron microscopy (SEM, Ultra 55, Zeiss) and high-resolution transmission electron microscopy (TEM, FEI Titan G2 60–300) equipped with energy-dispersive X-ray spectroscopy (EDX, Oxford) to test the elemental distributions of cathode material (x = 0.02). The particle size was measured by using laser particle size Analysis (LPS, TOOLSO, 2005A). The chemical states of the doping element were determined by using X-ray photoelectron spectroscopy (XPS, Perkin Elmer PHI 1600). And the XPS spectra were fitted by using XPSPEAK software. The elemental composition, i.e. Ni, Co, Mn and Yb, was detected by ICP-OES (Inductively Coupled Plasma Optical Emission Spectrometer, iCAP 6000). Phase transformation studies of original and cycled Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.02) were carried out using a micro-Raman spectrometer (LabRAMHREvolution, HORIBA).

The electrochemical properties of Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples were measured by using galvanostatic charge and discharge with the coin cell of type CR2025. The coin cells were assembled as follows: (1) The 85 wt.% Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples, 10 wt.% carbon black and 5 wt.% polyvinylidene fluoride were evenly mixed to form the cathode slurry; (2) Then the slurry was casted onto Al foil by using a smudge stick and dried at 110 °C for 12 h in vacuum drying oven, followed by squeezed and punched into a circular disc with d = 12 mm; (3) The as-prepared cathode plate, the lithium metal plate as anode, the Celgard 2400 as the separator and 1 M LiPF₆ dissolved in EC/DMC at mass ratio of 1:1 as the electrolyte were assembled in an argon-filled glove box to form the coin cells. The Galvanostatic charge-discharge tests were carried out by on a Land CT2001A (Wuhan, China) tester.

The cells were charged and discharged in the voltage range of 2.0 to 4.8 V at the different current densities (1C = 250 mA g⁻¹). In addition, the CHI660D workstation was used to perform the electrochemical impedance spectroscopy (a frequency range from 0.01 Hz to 100 kHz and perturbation amplitude of 5 mV) and the cyclic voltammogram (a voltage range from 2.0 V to 4.8 V with a scanning rate of 0.1 mV s⁻¹).

Results and Discussion

Figure 1 shows the X-ray diffraction patterns of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03)samples. The as-prepared samples have mainly demonstrated the typical XRD patterns of the hexagonal α-NaFeO₂ structure with the space group R-3m (the LiMO₂ features), except for the weak super lattice peaks between 20° and 25°, which are related to the Li₂MnO₃ phase, corresponding to the monocline unit cell C2/m^19,20. In addition, the distinct splitting of (006)/(102) and (018)/(110) peaks have indicated that the as-prepared cathode materials have formed a well-developed hexagonal layered structure²¹. Besides, to further investigate the cation mixing between the Ni²⁺ and Li⁺ in the LiMO₂ main phase, the Rietveld refinement of the diffraction patterns was performed based on the R-3m (used for LiNi_0.50Co_0.20Mn_0.30O₂ phase) and C2/m (used for Li₂MnO₃ phase) structure, as is shown in Fig. 1. And the structural parameters obtained from the refinement for the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples are listed in Table 1. It can be seen that with the Zr⁴⁺ doping content increasing, the lattice parameters a and c of LiNi_0.50Co_0.20Mn_0.30O₂ phase have gradually risen owing to the larger radius of Zr⁴⁺. The larger lattice parameters a and c will contribute to enhancing the Li⁺ diffusion rate during the charge and discharge process²². Besides, the c/a ratio is related to the cation mixing and a high ratio represents the well cation ordering has been formed²³. It can be observed the Zr⁴⁺-doped samples deliver the higher c/a ratio than that of the un-doped cathode, indicating the cation mixing of the as-prepared samples has been improved after the Zr⁴⁺ doping. Besides, according to the reports of J.R. Dahn^24,25, the nominal formula of Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ can be assumed as [Li_1−δNi _δ ][Li _δ Mn_0.52−xZr _x Ni_0.20−δCo_0.08]O₂. The GSAS/EXPGUI program has been adopted to calculate the refined lattice structural data of as-prepared samples, as is shown in Table 1. It is clear that the amount of Ni in Li site for the Zr⁴⁺-doped samples is lower than that of the pristine Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ sample. And when the Zr⁴⁺ doping content aggrandizes, the amount of Ni in Li site first decreases from 0.059 to 0.041 and 0.032, then increases to 0.039, the Li_1.20[Mn_0.50Zr_0.02Ni_0.20Co_0.08]O₂ has demonstrated the optimal cation ordering. The lower cation mixing will not only suppress formation of spinel-like phase, but also improve the layered structure stability, finally contribute to enhancing the cyclic performance. Besides, the occupancy of Zr cations in 3b-site are respectively 0, 0.012, 0.019 and 0.031 with the Zr doping contents increasing, indicating the molar ratio for Zr doping can be designed experimentally. The Zr⁴⁺ doping can enlarge the lattice parameters, which facilitates Li-ion diffusion and subsequently enhances the high-rate capability.

Figure 1

Profile fits for Rietveld refinement of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples.

Table 1 Structural parameters obtained from the refinement by the Rietveld method of the X-ray diffraction data recorded for the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples.

Figure 2 shows the SEM images of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples. The as-prepared samples are composed of numerous crystallites with a diameter of 200~700 nm. And all particles present the similar morphology of rock-shaped grains without obvious aggregation. In addition, with the Zr⁴⁺ doping content increasing, the crystal particles surface become more smooth and the size of the particles become larger, which implies the crystallinity of the particles can be enhanced after the Zr⁴⁺ doping. To further analyze the influence of the Zr⁴⁺ doping on the cathode particles size, the size distribution of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples have been measured, as is shown in Fig. 3. It is obvious that the size of D₅₀ gradually aggrandizes when the Zr⁴⁺ doping content increases, as is shown by the arrows, which is in good consistent with the observation of SEM images. A small amount of doped Zr ions may form continuous grain boundary phases in the Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ particles. These continuous grain boundary phases could enhance the mass diffusion transport at grain boundaries, finally promote the grain growth of Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂^26,27. And the well crystallization will help to ameliorate the electrochemical properties of cathode. Besides, the STEM images of Li_1.20[Mn_0.50Zr_0.02Ni_0.20Co_0.08]O₂ and corresponding elemental maps of Ni, Mn, Co and Zr is shown in Fig. 4. The Fig. 4 demonstrates that not only the Ni, Co and Mn atoms have been distributed homogeneously, but also the doping element Zr atom have been evenly distributed in the cathode particles rather than segregated on the oxide surface, indicating the Zr⁴⁺ doping technology has obtained the obvious synthetic efficiency. Based on the above analysis, it has proved that the Zr⁴⁺ has been successfully doped into the Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ cathode material with uniform dispersion. The uniform dispersion of Zr dopant will make the function of Zr⁴⁺ doping modification more stability, which may be ready to provide a better cycling performance to some extent.

Figure 2

SEM images of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples.

Figure 3

Size distribution of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples.

Figure 4

STEM images of Li_1.20[Mn_0.50Zr_0.02Ni_0.20Co_0.08]O₂ and corresponding elemental maps of Ni, Mn, Co and Zr.

Figure 5 shows X-ray photoelectron spectroscopy (XPS) results of Zr, Mn, Ni and Co for the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.02) samples. In Fig. 5(a), the obvious peaks at the binding energies of 184.9 eV and 182.6 eV are assigned to Zr 3d_5/2 and Zr 3d_3/2, respectively, which corresponds to the Zr-O bonds at the state of Zr^4+ 28. In Fig. 5(c), the obvious peaks at the binding energies of 854.2 eV is assigned to Ni_2p3/2, which corresponds to the oxidation state of Ni²⁺ and Ni³⁺ after fitting, respectively^29,30. Besides, it can be calculated that the relative content of Ni²⁺ decreased after zirconium doping owing to the reduction of cation mixing degree. Compared with the pristine Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂, the binding energies of Mn_2p and Co_2p peaks for Li_1.20[Mn_0.50Zr_0.02Ni_0.20Co_0.08]O₂ have no obvious changes, indicating the chemical properties of the Mn and Co elements have not been changed after the Zr⁴⁺ doping modification.

Figure 5

x-ray photoelectron spectroscopy (XPS) results of Zr, Mn, Ni and Co for the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.02) samples.

To acquire the elements composition of Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples, the ICP test was adopted and the results is demonstrated in Table 2. The measurement values of Ni, Co, Mn and Zr elements content are approximately equal to the theoretical analysis values, indicating that the molar ratios for Ni, Co, Mn and Zr elements have been synthesized in accordance with the experimental requirements.

Table 2 Relative contents of Ni, Co, Mn and Zr in the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples.

Figure 6 shows the initial charge-discharge curves of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples in the voltage range of 2.0~4.8 V at 0.1 C rate. All samples have demonstrated the similar charge curve for the two typical charge steps. The first step of charging process exists in the potential region from 2.0 V to 4.5 V, corresponding to the Li⁺-extraction from layer LiNi_0.50Co_0.20Mn_0.30O₂ component and the oxidation of Ni²⁺ to Ni⁴⁺ and Co³⁺ to Co^4+ 31,32. For the second step, all samples exhibit a long voltage plateau at about 4.5 V, where the irreversible Li⁺ extract and oxygen release from the Li₂MnO₃ phase^33,34. Table 3 shows the initial cycle electrochemical data of Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) cathodes at 0.1 C rate in the voltage range of 2.0~4.8 V. With the Zr⁴⁺ doping content increasing, the initial charge capacities of as-prepared samples gradually decline owing to the electrochemical inactive of doped Zr⁴⁺, While the discharge capacities first enhance and then decrease and the Li_1.20[Mn_0.50Zr_0.02Ni_0.20Co_0.08]O₂ sample delivers the highest discharge capacity of 272.4 mAh g⁻¹. In addition, the lowest irreversible capacity loss for the Li_1.20[Mn_0.50Zr_0.02Ni_0.20Co_0.08]O₂ sample has promoted the highest initial coulombic efficiency, which indicates that the Zr⁴⁺ doping can restrain the release of oxygen from the Li₂MnO₃ and decrease the irreversible capacity loss. Compared to the bonds break energy values for theΔHƒ₂₉₈(Ni-O) = 391.6 kJ∙mol⁻¹, ΔHƒ₂₉₈(Co-O) = 368 kJ∙mol⁻¹ and ΔHƒ₂₉₈(Mn-O) = 402 kJ∙mol⁻¹, the Zr-O delivers the higher bonds break energy value ofΔHƒ298(Zr-O) = 760 kJ mol⁻¹, therefore with the Zr⁴⁺ doping, the oxygen release of the Zr⁴⁺-doped samples will face more resistance than the un-doped sample, subsequently the irreversible capacity loss has been suppressed¹⁷.

Figure 6

Initial charge-discharge curves of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples in the voltage range of 2.0~4.8 V at 0.1 C rate.

Table 3 Initial cycle electrochemical data of Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) cathodes at 0.1 C rate in the voltage range of 2.0~4.8 V.

Figure 7 shows the rate capabilities of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples with various current densities in the voltage range of 2.0~4.8 V. Obviously, the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0.01, 0.02, 0.03) samples have all demonstrated the higher discharge capacities than those of the pristine Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ at the rate of 0.1 C, 0.2 C, 0.5 C, 1 C, 2 C and 5 C, thereinto the Li_1.20[Mn_0.50Zr_0.02Ni_0.20Co_0.08]O₂ sample delivers the optimum rate capacity. In addition, with current density increasing, the superiority has become particularly evident, indicating the advantage of Zr⁴⁺ doping on the rate capacity of Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ is much more significant at high rate. As is seen in Table 4, the discharge capacity of the Li_1.20[Mn_0.50Zr_0.02Ni_0.20Co_0.08]O₂ is only 8.8 mAh g⁻¹ higher than that of the bare Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂. However when the current density enhances to 5 C rate, the bare sample shows a discharge capacity of 86.6 mAh g⁻¹ and this value is increased to 105.3, 114.7 and 108.6 mAh g⁻¹ for the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0.01, 0.02, 0.03) samples, respectively. The superior rate capacity of the Zr⁴⁺-doped Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ samples have mainly been attributed to the fast Li⁺ migration speed during the charge and discharge process. One reason is that with the Zr⁴⁺ doping, the larger lattice parameters of the Zr⁴⁺-doped samples have contributed to enhancing the Li⁺ diffusion speed. Besides, the better crystallization property of the Zr⁴⁺-doped samples will also help to strengthen the conductivity ability of ions and electrons during the charge-discharge process.

Figure 7

Rate capabilities of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples with various current densities in the voltage range of 2.0~4.8 V.

Table 4 Discharge capacity of Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) at various current densities in the voltage range of 2.0~4.8 V.

Figure 8 shows the cycling performance of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples at 0.5 C rate in the voltage range of 2.0~4.8 V at room temperature (25 °C). It can be observed that the Zr⁴⁺-doped samples have delivered the higher discharge capacity than that of the bare Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂. And with the cycles going on, the cycling performance of the bare Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ is similar to those of the Zr⁴⁺-doped samples, the discharge capacities have all gradually attenuated followed the same trend. Table 5 shows the discharge capacity of Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) at 0.5 C rate in the voltage range of 2.0~4.8 V at 25 °C. With the Zr⁴⁺ doping content increasing, the initial discharge capacities are198.0, 202.0, 208.3 and 203.0 mAh g⁻¹, respectively. And after 100 cycles, the corresponding capacity retentions still maintain 86.9%, 88.5%, 90.4% and 88.7%, respectively. It has proved that the Zr⁴⁺ doping modification can enhance the specific capacity and cycling performance of the Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ cathode, owing to the lower cation mixing and faster Li⁺ migration speed for the Zr⁴⁺-doped samples. Besides, the discharge voltage plateau will gradually decrease during the cyclic process, owing to the enlargement of polarization and the formation of spinel-like phase for cathode materials³⁵. It can be observed that the discharge voltage drops to lower plateau for the all cathodes after different cycles, as the arrows pointed in Fig. 9. Table 5 shows the declining value of voltage plateau between 1st and 100th (ΔV) for the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples and with the Zr⁴⁺ doping contents increasing, the ΔV values are 0.298, 0.259, 0.211 and 0.236 V, respectively. The smaller ΔV values of the Zr⁴⁺-doped cathodes have indicated that the Zr⁴⁺ doping modification can improve the layered structural stability by restraining the cation mixing between the Ni²⁺ and Li⁺ and the formation of spinel-like phase. While the smaller ΔV values of Zr⁴⁺-doped cathodes will contribute to maintaining the high power output of cells.

Figure 8

Cycling performance of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples at 0.5 C rate in the voltage range of 2.0~4.8 V at 25 °C.

Table 5 Discharge capacity and the difference value of discharge mid-point voltage (∆V) of Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) at 0.5 C rate in the voltage range of 2.0–4.8 V at 25 °C.

Figure 9

Discharge profiles of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples from 2.0 V to 4.8 V at 0.5 C rate in the 1st, 30th, 60th and 100th cycles.

The poor cycling performance at high temperature for the Li-excess Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ has become one of the main drawbacks for the commercial application owing to the enhancement of the side reaction between cathode and electrolyte. Figure 10 shows the cycling performance of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) at 0.5 C rate in the voltage range of 2.0~4.8 V at 45 °C. In comparison with the cycling performance at room temperature, the bare Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ demonstrates the more severe capacity fading. During the early cycle period, the fast capacity attenuation can be observed owing to the bare cathode particles surface. After several cycles, the side reaction between the cathode and electrolyte can generate some by-product, which will deposit at the electrode/electrolyte interface to form the Solid Electrolyte Interface (SEI) film. And the SEI film will protect the cathode materials from erosion by the electrolyte, making the capacity attenuation slightly slow^36,37. The initial discharge capacities are 221.8, 226.6 and 218.0 mAh g⁻¹ for Zr⁴⁺-doped Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ electrodes with the doping contents of 0.01, 0.02 and 0.03, respectively, larger than that (208.3 mAh g⁻¹) of the un-doped Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂, as is seen in Table 6. After 100 cycles, with the Zr⁴⁺ doping contents increasing, the Zr⁴⁺-doped Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ samples exhibit the discharge capacity of 180.2, 190.7 and 176.9 mAh g⁻¹ respectively, corresponding that the capacity retentions first enhance from 86.3% to 88.7% and then decline to 86.5%. As for the bare Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂, the discharge capacity decreases acutely to 172.5 mAh g⁻¹ with the capacity retention of only 82.8%. During the charge-discharge process at high temperature, the cathodes have suffered from the attack of HF, dissolution of the Mn ions, structural change and decomposition of electrolyte on the cathode surface³⁸. While the stronger total metal–oxygen bonding for the Zr⁴⁺-doped samples can contribute to stabilizing the structure of cathode during cycling, leading to the improved cycling performance. However, when the Zr⁴⁺ doping content reaches to 0.03, the cycling performance of Li_1.20[Mn_0.49Zr_0.03Ni_0.20Co_0.08]O₂ is not as good as that of the Li_1.20[Mn_0.50Zr_0.02Ni_0.20Co_0.08]O₂ for that the inhomogeneity phase of the ZrO₂ existed in the compound can hinder the Li⁺ intercalation/deintercalation from the cathode¹⁷.

Figure 10

Cycling performance of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) at 0.5 C rate in the voltage range of 2.0~4.8 V at 45 °C.

Table 6 Capacity retention and discharge capacity of Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) at 0.5 C rate in the voltage range of 2.0–4.8 V at 45 °C.

To further understand the influence of Zr⁴⁺ doping on the electrochemical properties of Zr⁴⁺-doped Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂, the electrochemical impedance spectroscopy (EIS) for the four samples have been carried out after charging to 4.5 V in the 1st, 30th cycles. Figure 11 shows the Nyquist curves of the four cathodes and all the Nyquist curves demonstrate the similar characteristics, containing a small semicircle in the high frequency, a large semicircle in the high to medium frequency and a quasi-straight line in the low frequency, which respectively correspond to the impedance of Li⁺ migration across the SEI film (R _sf and CPE _sf ), the impedance of charge transfer (R _ct and CPE _dl ) and the impedance of Li-ion migration in the cathode (Z _W )^39,40. The corresponding equivalent circuit in Fig. 10(e) is used to simulate the Nyquist curves and the corresponding R_s, R_sf and R_ct values can be acquired, as is shown in Table 7. In the 1st cycle, the R _sf values of Zr⁴⁺-doped Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ are lower than that of the bare one, therefore the superior initial discharge capacity can be obtained for the Zr⁴⁺-doped Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂, which is in consistency with the results of Table 3. With the cycles going on, the SEI film will thicken, causing the increase of the R _sf value. After 30 cycles, with the Zr⁴⁺ doping contents increasing, the Zr⁴⁺-doped Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ samples deliver the R _sf values of 445.8, 363.1 and 428.8 Ω respectively, corresponding that the ΔR _sf values first enhance from 287.1 to 210.8 and then drop to 284.9 Ω. As for the bare Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂, the R _sf value rise promptly to 544.5 Ω, with the ΔR _sf value of 371.4 Ω. It indicates the samples after the Zr⁴⁺ doping can relieve the thickening of SEI films on the surface of cathode, which contributes to decreasing the Li⁺ migration resistance across the SEI films and enhancing the electrochemical properties. Besides, the Li⁺ diffusion rate (D _Li ⁺) in the cathode can be calculated using the following equations⁴¹:

$${D}_{L{i}^{+}}=\frac{{R}^{2}{T}^{2}}{2{F}^{4}{n}^{4}{A}^{2}{C}^{2}{\tau }_{W}}$$

(1)

$${Z}_{re}={R}_{S}+{R}_{ct}+{\tau }_{W}{\omega }^{-1/2}$$

(2)

where, F, n, A, C R is gas constant, T is the absolute temperature, F represents the Faraday constant, n is the number of electrons per molecule during oxidation, A corresponds to the area of the electrode-electrolyte interface, i.e. 1.13 cm² and C is the concentration of lithium ion, respectively. Besides, τ _W is the Warburg coefficient of the bulk cathode, which is can be calculated by the Eqs (2). Thereinto, the Z_re is the real part of impedance, ω is the angular frequency⁴² and Fig. 12 shows the plots comparison of Z_re vs. ω^−1/2 for the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples after 30 cycles. Thus τ _W can be obtained from the linear fitting of Z_re vs. ω^−1/2.

Figure 11

Nyquist plots of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples at a charge state of 4.5 V in the 1st, 30th cycles and (e) the equivalent circuit used to fit the measured impedance spectra.

Table 7 The simulated data of the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) cathodes at 4.5 V from EIS spectra using the equivalent circuit shown in Fig. 11(e).

Figure 12

Plots comparison of Z_re vs. ω^−1/2 for the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.01, 0.02, 0.03) samples after 30 cycles.

According to Eqs (1) and (2), after 30 cycles, with the Zr⁴⁺ doping contents increasing, the Zr⁴⁺-doped Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ samples exhibit the D_Li⁺ values of 3.61 × 10⁻¹⁴ cm² s⁻¹, 8.32 × 10⁻¹⁴ cm² s⁻¹ and 5.46 × 10⁻¹⁴ cm² s⁻¹ respectively, higher than that (7.63 × 10⁻¹⁵ cm² s⁻¹) of the pristine electrode. Therefore, the Zr⁴⁺-doped Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ samples have demonstrated the superior rate capacity.

Figure 13 shows the Raman spectra of original and cycled Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.02) (100 cycles at 45 °C). The Raman band at 427 cm⁻¹ corresponds to the monoclinic Li₂MnO₃ phase, which can be observed in the spectrum of original Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.02) in Fig. 13(a) and disappear in the cycled electrode in Fig. 13(b) owing to the disappearance of the monoclinic Li₂MnO₃ component after cycling⁴³. Besides, the other two significant Raman bands at 483 and 595 cm⁻¹ for the Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ (481 and 593 cm⁻¹ for the Li_1.20[Mn_0.50Zr_0.02Ni_0.20Co_0.08]O₂ owing to the Zr⁴⁺ doping) belong to the bending E_g and stretching A_1g modes, respectively⁴⁴ in Fig. 13(a). After 100 cycles, the Raman bands at 595 cm⁻¹ for the Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.02) have both shifted to higher values, indicating the cathode structure transformation from the layered to defect spinel structure⁴⁵. The Raman band for the Li_1.20[Mn_0.50Zr_0.02Ni_0.20Co_0.08]O₂ have shifted from 593 to 602 cm⁻¹ after 100 cycles, much lower than that of the pristine Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ (from 595 to 615 cm⁻¹). Therefore, the Zr⁴⁺ doping have restrained the layered-to-spinel phase change of Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ during cycling, forming the superior electrochemical properties by Zr doping.

Figure 13

(a) Raman spectra of original Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.02); (b) Raman spectra of cycled Li_1.20[Mn_0.52−xZr _x Ni_0.20Co_0.08]O₂ (x = 0, 0.02) (100 cycles at 45 °C).

Conclusions

In order to enhance the electrochemical properties of Li-excess Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂, the different contents of Zr⁴⁺ have been doped into the pristine Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂. After the Zr⁴⁺ doping, the cation mixing between Li⁺ and Ni²⁺ has been lowered and the cathode particles have been aggrandized. In comparison with the pristine cathode, the Zr⁴⁺-doped Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ samples have demonstrated the more stable cycling performance and higher rete capacities. Especially at high temperature (45 °C), the Zr⁴⁺ doping modification has delivered the more obvious superiority. After 100 cycles, with the Zr⁴⁺ doping contents increasing, the Zr⁴⁺-doped Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂ samples exhibit the capacity retentions of 86.3%, 88.7% and 86.5% respectively, larger than that (82.8%) of the bare Li_1.20[Mn_0.52Ni_0.20Co_0.08]O₂. The stronger total metal–oxygen bonding for the Zr⁴⁺-doped samples has mainly contributed to stabilizing the structure of cathode and improving the cycling performance. The Zr⁴⁺ doping modification has provided a potential approach to enhance the electrochemical properties of the Li-excess cathodes for Li-ion battery.