Effect of Nb⁵⁺ doping on LiNi_0.5Co_0.25Mn_0.25O₂ cathode material

Abstract

Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂ (x = 0, 0.005, 0.01, 0.02, 0.03) cathode material was synthesized by co-precipitation. X-ray diffraction spectroscopy (XRD) and scanning electron microscopy (SEM) were used to analyze the crystal structure characteristics and morphology of the powder. The charge and discharge test, electrochemical impedance spectroscopy (EIS), and cyclic voltammetry (CV) were used to study the electrochemical properties of the battery in detail. XRD results show that Nb⁵⁺ substitution does not destroy the crystal structure, but it can enlarge the interplanar spacing, which is beneficial to the diffusion of lithium ions. The electrochemical properties of the material Li (Ni_0.5Co_0.25Mn_0.25)_0.99Nb_0.01O₂ are the best. The discharge specific capacity is 204.6, 186.0, 163.5, 141.6 mAh/g at 0.1C, 0.2C, 0.5C, and 1.0C, respectively. And the discharge specific capacity is as high as 174.1 mAh/g when returning to 0.1C again. After circulating 45 cycles at 0.1C, the capacity retention rate was 89.08%.

Introduction

With the development of electric vehicles, large-scale electronic equipment, and hybrid vehicles, lithium-ion batteries have received extensive attention [1]. Lithium-ion batteries have enormous global potential for achieving energy sustainability and significantly reducing carbon emissions [2]. The requirements for lithium-ion batteries are increasing, such as high energy density, excellent cycle stability, and reliable safety [3, 4]. Lithium-ion batteries of widely used LiCoO₂ materials are hexagonal α-NaFeO₂-type belonging to the R-3m space group. This layered structure allows LiCoO₂ to have good lithium-ion intercalation/deintercalation capabilities. However, the structural stability of LiCoO₂ deteriorates drastically when lithium ions are half-extracted from the structure during charging. Therefore, in order to ensure its cycle performance limits its cut-off voltage, it can only provide half of the theoretical capacity of about 140 mAh/g capacity [5, 6]. The ternary cathode material LiNi_1−x−yCo_xMn_yO₂ has a structure similar to LiCoO₂, and because it has the advantages of relatively low cost, high capacity, and better thermal stability, it can be used instead of LiCoO₂ [7, 8]. However, studies have shown that LiNi_1−x−yCo_xMn_yO₂ (0 < x < 1, 0 < y < 1 x + y = 1) has a faster capacity loss. It is because of stone and rock salt phase transition due to Ni²⁺ being more likely to occupy the Li⁺ position [9,10,11,12], oxygen in the lattice released from the surface [3, 13, 14], and other shortcomings resulting in poor crystal structure [15].

In order to improve the performance of lithium-ion batteries, people mainly study cathode materials and electrolytes [16, 17]. The main modification methods for lithium-ion ternary cathode materials are doping [18,19,20], coating [21, 22], and core-shell structure [23].

The method of element doping refers to doping with other small amounts of metal or non-metal elements inside the crystal lattice, and it is desirable to improve certain electrochemical properties of the original material. The radius of Nb⁵⁺ is 0.064 nm, and the radius of Mn⁴⁺ is 0.053 nm. Since the radius of Nb⁵⁺ is close to the radius of the transition element and Nb⁵⁺ has a larger diameter than Mn⁴⁺ [24], Nb⁵⁺ can enhance the lattice parameters. It is important that it has a higher metal-oxygen bond energy. Hu et al. studied the effect of Nb doping on the positive electrode material 0.5Li₂MnO₃·0.5LiNi_1/3Co_1/3Mn_1/3O₂, and found that Nb is most effective in improving cycle performance and reducing voltage attenuation [25]. Yi et al. performed Nb doping on LiMn₂O₄. The results show that Nb doping can reduce electrode polarization and increase lithium-ion diffusion coefficient [26].

Considering the substitution of Nb⁵⁺ may improve the electrochemical properties of the material LiNi_0.5Co_0.25Mn_0.25O₂ in some extent. In this paper, we prepared Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂ (x = 0, 0.005, 0.01, 0.02, 0.03) by co-precipitation. The effects of Nb⁵⁺ doping on the structure, morphology, and electrochemical properties of the cathode material LiNi_0.5Co_0.25Mn_0.25O₂ were investigated in detail.

Experiment

Synthesis

The precursor Ni_0.5Co_0.25Mn_0.25(OH)₂ was first prepared by hydroxide coprecipitation, and then Nb⁵⁺ was introduced in the ball milling stage. The mixture which was composed of NiSO₄·6H₂O, CoSO₄·7H₂O, and MnSO₄·H₂O in a molar ratio of 5:2.5:2.5 was dissolved in a small amount of deionized water and then configured as a solution having a transition metal ion concentration of 1 mol/L. The precipitant NaOH solution was prepared to be 2 mol/L, and the complexing agent NH₃·H₂O was set to 1 mol/L. The three solutions were simultaneously pumped into the reactor while maintaining a reaction temperature of 55 °C and a pH of 11. After the reaction was completed, the mixture was stirred at a constant temperature for 2 h, and finally it was aged overnight. The suspension was thoroughly washed and filtered to remove impurity ions (Na⁺, SO₄²⁺, NH₃⁺, etc.). After the end of the suction filtration, the filter cake was placed in a blast drying oven at 102 °C to remove moisture.

According to the stoichiometric molar ratio of the chemical formula Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂ (x = 0, 0.005, 0.01, 0.02, 0.03), the prepared powder was ball milled with Li₂CO₃ and Nb₂O₅, and the amount of Li₂CO₃ added was 5% higher than the calculated value, and Nb₂O₅ was used as the niobium source. The milling time is 5 h to make them mix uniformly.

The resulting mixture was placed in a muffle furnace and heated to 450 °C at a heating rate of 5 °C/min in an oxygen atmosphere, and then preheated for 5 h. After cooling to room temperature in a furnace, it was taken out and carefully ground. Finally, the temperature was raised to 850 °C at a heating rate of 5 °C/min, and the ground powder was calcined at 850 °C for 12 h to obtain a final product.

Electrode film and battery assembly preparation

First, the active material, acetylene black, and PVDF (polyvinylidene fluoride) were in an amount of 8:1:1, and the NMP (N-methyl-pyrrolidone) was taken in an appropriate amount. Acetylene black was added as a conductive agent, and PVDF was used as a binder. The active material and acetylene black were thoroughly ground in an agate mortar, and then PVDF was dissolved in NMP. The ground mixture was added to the solution. Stirring was carried out to obtain a uniformly dispersed slurry, which was then uniformly coated on an aluminum foil. The coated aluminum foil was dried at 110 °C for 10 h. After drying, it was compacted by a tableting machine and then punched into circular electrode sheets having a diameter of 10 mm.

Second, the preparation of the battery assembly was carried out in a dry inert gas glove box. The operation sequence was first placed with a negative electrode (lithium plate), a separator (Celgard 2400 porous polypropylene film), an appropriate amount of electrolyte 1 mol/L LiPF₆ (EC+EMC+DMC volume ratio 1:1:1), and a positive electrode sheet. It was sealed and subjected to electrochemical test after standing for 24 h.

Physical characterization and electrochemical performance of materials

The crystal structure was analyzed using an X-ray diffraction analyzer ((XRD), Bruke D8-Fouse, Germany) which uses Cu-Kα as a radiation source with a scanning range of 2θ = 10–80° and a scanning speed of 12°/min. The morphology of the synthesized powder was analyzed by scanning electron microscopy ((SEM), Nova Nano SEM450 FEI), and energy dispersive spectrometer ((EDS), manufactured by AMETEK, model OCTANE PLUS) was used to test the element content in the material. The surface chemical compositions of the samples were measured by X-ray photoelectron spectroscopy ((XPS), ESCALAB 250Xi).

Charge and discharge were tested by the battery test system Land CT2001A with a test voltage range of 2.7–4.3 V. Both cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) were tested on PARSTAT 4000 electrochemical analyzer. The test parameters for CV were a scan rate of 0.1 mV/s and a voltage range of 2.7–4.5 V. The EIS test amplitude is 5 mV in the frequency range of 100 kHz to 0.01 Hz. The EIS data was analyzed using the ZsimpWin 3.10 software. All the characterizations and measurements were taken at room temperature.

Results and discussion

Physical characterization

Figure 1 shows the XRD pattern of Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂, all of which are consistent with the layered features of α-NaFeO₂. Compared with the undoped material, it is found that the position of the Nb⁵⁺ doped material peak does not change. Each peak has a characteristic peak of a typical α-NaFeO₂ structure, and the shape of the peak is clear, and the intensity is large. This indicates that Nb⁵⁺ doping does not change the crystal structure of the original material. By carefully comparing the samples x = 0, x = 0.005, x = 0.01, x = 0.02, x = 0.03, it is found that the (006)/(102) peak is equally sharp except for sample x = 0.03. The (108)/(110) peaks all have good splitting, indicating a layered structure. Table 1 summarizes the unit cell parameters after refining of all samples. As the Nb⁵⁺ doping content increases, the lattice parameters a and c show a slight increase, indicating that the modification has induced the structure [27]. It can be seen from the volume parameter that the lattice volume after doping is larger than the original one. This may be due to the ionic radius of Nb⁵⁺ (0.64 Å) being larger than that of Mn⁴⁺ (0.53 Å) [24]. After doping, the intensity ratio of I(003)/I(104) are all larger than the original intensity ratio, which indicates that Li⁺/Ni²⁺ mixing is reduced [25].

Fig. 1

XRD pattern of Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂

Table 1 Lattice parameters of Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂

In order to observe the surface topography of the material, Fig. 2 was obtained by scanning electron microscope. It can be seen that the Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂ particles consist of densely packed particles. The micro-spherical particles can effectively shorten the Li⁺ diffusion path, reduce volume expansion/compression during charging/discharging, and promote diffusion of electrolyte into the electrode material [28]. The particle surface of Li (Ni_0.5Co_0.25Mn_0.25)_0.99Nb_0.01O₂ material is very smooth and clean. Its particle size is the most uniform, so its electrochemical performance should be the best.

Fig. 2

SEM image of Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂

In order to determine the elemental composition and distribution of the sample, and further prove the uniformity of Nb⁵⁺ doping, EDS spectrum testing was performed on this sample in Fig. 3. According to the test results, each region is found to contain nickel, cobalt, manganese, and niobium. The relative content of the elements in these two regions is approximately the same, and the atomic percentage ratio is very close to the target chemical formula. It is also proved that Nb⁵⁺ was successfully doped.

Fig. 3

EDS image of Li (Ni_0.5Co_0.25Mn_0.25)_0.99Nb_0.01O₂

XPS was used to detect the chemical state of various elements on the surface. All spectra were calibrated by assigning the C 1s peak at 284.6 eV. Background type was Shirley. Figure 4a is the binding energy spectrum of Li (Ni_0.5Co_0.25Mn_0.25) _0.99Nb_0.01O₂, and Ni 2p, Co 2p, Mn 2p, O 1s, C 1s, Nb 3d, and Li 1s peaks can be observed from it. The C 1s are mainly due to adventitious carbon formed during the atmospheric exposure [29]. The XPS spectrum of Ni 2p has two main peaks and two satellite peaks. The two main peaks are at the peaks of 854.9 eV and 872.4 eV, which are attributed to 2p_3/2 and 2p_1/2. According to the fitted XPS data, Ni 2p_1/2 and Ni 2p_3/2 both contain two splitting peaks corresponding to two different oxidation states of + 2 and + 3 [30, 31], respectively. In Fig. 4d, the two main peaks of the Co 2p spectrum are Co 2p_3/2 at 780.0 eV and Co 2p_1/2 at 794.9 eV, indicating that Co exists mainly in the oxidation state of Co³⁺ [32, 33]. There are two main peaks in the Mn XPS spectrum as shown in Fig. 4e. They are Mn 2p_3/2 at 642.4 eV and Mn 2p_1/2 at 653.9 eV, which are related to Mn⁴⁺ for the sample. In Fig. 4f, the satellite peaks of Nb appear at 206.4 eV and 209.2 eV corresponding to 3d_5/2 and 3d_3/2, respectively. Consistent with previous literature reports [34, 35], it was proved that Nb⁵⁺ was successfully doped.

Fig. 4

XPS spectra for Li (Ni_0.5Co_0.25Mn_0.25)_0.99Nb_0.01O₂. a Full-spectrum. b Li 1s. c Ni 2p. d Co 2p. e Mn 2p. f Nb 3d

Electrochemical performance

In order to evaluate the electrochemical properties of the material, Fig. 5 shows the initial charge-discharge curve of Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂ (x = 0, 0.005, 0.01, 0.02, 0.03) for each material in the range of 2.7–4.3 V at a rate of 0.1C. All the curves in the figure have a charging or discharging platform, and the charging and discharging curves reflect the reversibility of Li insertion/extraction in Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂ crystals during electrochemical process [36]. The generation of the charging or discharging platform in the curve is due to the phase change. If the platform lasts longer, we will find that the voltage changes more slowly. This results in a higher charge/discharge specific capacity. Due to the precipitation/dissolution of lithium on the electrode surface during charge and discharge, the surface state of the electrode changes and the polarization changes at the same time. This result may make the battery’s first charge-discharge curve not smooth [37]. It can be seen from Fig. 5 that when x = 0, 0.005, 0.01, 0.02, 0.03, the first discharge specific capacities are 199.0, 135.1, 204.6, 164.7, 188.7 mAh/g, respectively. When x = 0.01, the material has the largest initial discharge specific capacity. This indicates that an appropriate amount of Nb⁵⁺ doping can increase the discharge specific capacity of the original material. This may be due to the fact that an appropriate amount of Nb⁵⁺ doping can increase the layer spacing and increase the diffusion capacity of Li⁺. However, as the amount of Nb⁵⁺ doping continues to increase, this may result in a decrease in the first discharge specific capacity due to a decrease in the content of the active material Ni²⁺/Ni³⁺.

Fig. 5

First charge and discharge curve of Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂

Figure 6 shows the rate performance of materials, and Fig. 7 shows the average specific discharge capacity of the materials at different discharge rates. Each battery is charged to 4.3 V at 0.1C each time, and then discharged to 2.7 V at 0.1C, 0.2C, 0.5C, 1.0C. It can be seen from Figs. 6 and 7 that the discharge specific capacities of different samples at the same discharge rate and the same sample at different discharge rates are significantly different. The former is due to the difference in the electrochemical properties of the material due to the different content of doped Nb⁵⁺. In the latter case, the polarization increases as the discharge rate increases. For detailed analysis, the specific discharge capacities of the original materials at 0.1C, 0.2C, 0.5C, 1.0C, and 0.1C are 199.0, 183.3, 154.0, 113.7, and 155.4 mAh/g, respectively. And its average discharge capacities are 196.72, 177.84, 142.16, 105.48, and 155.4 mAh/g, respectively. However, when x = 0.01, the discharge specific capacities of the material are 204.6, 186.0, 163.5, 141.6, and 174.1 mAh/g, respectively. And the corresponding average discharge specific capacities are 201.62, 181.34, 161.43, 138.36, and 169.2 mAh/g, respectively. Therefore, the sample Li (Ni_0.5Co_0.25Mn_0.25)_0.99Nb_0.01O₂ has the best rate performance. This result may be due to the proper amount of Nb⁵⁺ doped material, the better insertion/extraction reversibility of Li⁺ [25], thereby increasing the reversibility of lithium-ion transfer between the two electrodes [38].

Fig. 6

Rate performance of Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂

Fig. 7

Average discharge specific capacity curve of Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂

Figure 8 shows the cycle performance of materials for charging and discharging at 2.7–4.3 V at a rate of 0.1C. The cycle performance analysis of the material is shown in Table 2. The capacity retention rate of the original material is 72.95%. When x = 0.005, x = 0.01, x = 0.02, x = 0.03 capacity retention rates were 75.93%, 89.08%, 87.17%, and 66.16%, respectively. It is apparent that the capacity retention of the original material is low due to surface structure changes of the active material and decomposition of the electrolyte, which is induced by highly reactive Ni⁴⁺ and/or some Li residues [5, 39]. The capacity retention of the material is improved when the proper amount of Nb⁵⁺ is doped, and the cycle stability of the material is best when x = 0.01. This may be because the bond energy of Nb-O is stronger than that of MO (M = Ni, Co, Mn), and an appropriate amount of Nb doping can stabilize the bulk structure of the cathode material during Li⁺-ion intercalation/deintercalation [40].

Fig. 8

Cycle performance of Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂

Table 2 Cycle performance analysis of Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂

Figure 9 is a cyclic Nyquist diagram of the electrode, and Fig. 10 is an equivalent circuit diagram. It is clear that the Nyquist plots have similar shapes. The pattern consists of a small intercept, a semicircle in the high frequency region, and a straight line in the low frequency region. They are attributed to the ohmic resistance (R_s) determined by the resistance of the electrolyte and the electrode, the resistance (R_f) of the solid electrolyte interface ((SEI) film) layer, the charge transfer resistance (R_ct) at the electrode/electrolyte interface, and Warburg impedance of Li⁺ diffusion, respectively [41,42,43].

Fig. 9

Electrochemical impedance spectroscopy (EIS) of Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂

Fig. 10

Electrochemical impedance spectroscopy fitting circuit diagram

The values of the impedance parameters of Table 3 were obtained by fitting the ZSimpWin software, in which the diffusion coefficient (D_Li+) of lithium ions in the cathode can be calculated by the following formula [44]:

$$ {D}_{\mathrm{Li}+}={R}^2{T}^2/{2\mathrm{A}}^2{\mathrm{n}}^4{F}^4{C}^2{\upsigma}^2. $$

(1)

$$ Z\hbox{'}=k+{\upsigma \upomega}^{-1/2}. $$

(2)

Table 3 EIS fitting data of Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂

In Eq. 2, k represents a constant, and σ represents the Warburg factor opposite to Z'. Therefore, to obtain the slope σ, only a function graph of Z' and ω^−1/2 is drawn (Fig. 11). In Eq. 1, n is the number of electrons required per unit reaction, R is the gas constant, T is the absolute temperature, n is the number of electrons required to participate in the unit reaction, F is the Faraday constant, A is the area of the cathode/electrolyte interface, and C is the concentration of lithium ions [45, 46].

Fig. 11

Linear fitting relationship between Z′and ω^−1/2 of Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂

By comparing the diffusion coefficients (D_Li+) of lithium ions of various samples by Table 3, it was found that the original material is lower than Li (Ni_0.5Co_0.25Mn_0.25)_0.99Nb_0.01O₂, which were 2.3301 × 10⁻¹⁴ and 2.9515 × 10⁻¹⁴, respectively. The results show that doping Nb⁵⁺ can enlarge the interplanar spacing and promote the diffusion of lithium ions, which improves the rate performance and cycle stability of the electrode materials.

Figure 12 is a cyclic voltammogram of the material with a scan voltage of 2.7–4.5 V and a scan rate of 0.1 mV/s. As shown, all materials have only a pair of redox peaks, and no phase transition from the hexagonal phase to the spinel phase occurs during Li-ion intercalation/deintercalation [47]. The oxidation peak on the corresponding curve during charging is related to the oxidation which is related to the oxidation of Ni²⁺/Ni³⁺ to Ni⁴⁺ and the oxidation of Co³⁺ to Co⁴⁺, indicating that Li⁺ is detached from the compound Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂. Upon discharge, a similar reduction peak which corresponds to a decrease in Ni-ions (Ni⁴⁺→Ni²⁺/Ni³⁺) and Co-ions (Co⁴⁺→Co³⁺) indicated in the reverse lithium intercalation layered structure was found. Mn⁴⁺ is inactive in the structure of Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂ because the valence is 4 [34]. When x = 0, 0.005, 0.01, 0.02, 0.03, the oxidation potential peaks were 3.9427 V, 3.9903 V, 3.8281 V, 3.8613 V, 3.9771 V, respectively; the reduction potential peaks were 3.6223 V, 3.6256 V, 3.6410 V, 3.6072 V, 3.6062 V, respectively; the peak difference between the two peaks are 0.3204 V, 0.3647 V, 0.1871 V, 0.2541 V, 0.3709 V, respectively. The peak difference is smaller, which indicates that the polarization of the material is smaller. Therefore, the reversibility of the material during charging and discharging is higher, and the electrochemical performance is better [48]. When the Nb⁵⁺ doping amount is x = 0.01, the minimum potential difference is 0.1871 V, which is smaller than the potential difference of the original material. Thus, an appropriate amount of Nb⁵⁺ doping reduces the electrochemical polarization of the original material [49]. Therefore, the material Li (Ni_0.5Co_0.25Mn_0.25)_0.99Nb_0.01O₂ has better electrochemical performance, which is consistent with the cycle performance and EIS analysis results.

Fig. 12

Cyclic voltammogram of Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂

Conclusion

In this paper, the material Li (Ni_0.5Co_0.25Mn_0.25)_1−xNb_xO₂ (x = 0, 0.005, 0.01, 0.02, 0.03) was synthesized by co-precipitation method, and the materials were structurally analyzed and electrochemically tested. Electrochemical data show that the material Li (Ni_0.5Co_0.25Mn_0.25)_0.99Nb_0.01O₂ has the best rate performance, and the specific discharge capacities at 0.1C, 0.2C, 0.5C, and 1C are 204.6, 186.0, 163.5, 141.6 mAh/g, respectively. From 1C to 0.1C, a discharge specific capacity of 174.1 mAh/g is still obtained. And after circulating 45 times at 0.1C, the capacity retention rate is the highest, which is 89.08%. The results show that an appropriate amount of Nb⁵⁺ doping can reduce the mixing of Li⁺/Ni²⁺ and increase the interlayer spacing. Furthermore, the cycle performance and rate performance of the material LiNi_0.5Co_0.25Mn_0.25O₂ are improved.