Effect of valence states of Ni and Mn on the structural and electrochemical properties of Li_1.2Ni_xMn_0.8−xO₂ cathode materials for lithium-ion batteries

Abstract

Severe capacity fading and voltage decay of Li-rich layered oxides for lithium-ion batteries remain the major bottlenecks to commercialization. Herein, we studied the effect of valence states of Ni and Mn on the structure and electrochemical performances of Li_1.2Ni_xMn_0.8−xO₂. The oxidation states of the transition-metal (TM) were evaluated by X-ray photoelectron spectra (XPS) and validated by cyclic voltammetry (CV). With increasing Mn³⁺ content, the Li₂MnO₃ component increases due to less Li loss. The phenomenon of capacity increase upon cycling has been observed owing to the gradual activation of Li₂MnO₃ caused by the existence of Mn³⁺. It is worth mentioning that Li_1.2Ni_0.1Mn_0.7O₂ reaches the highest discharge capacity value of 231.5 mA h g⁻¹ from the initial value of 76 mA h g⁻¹ after 75 cycles and still keeps 100% of the capacity at 0.1C after 100 cycles. With the increase of Ni³⁺, Li/Ni mixing decreases because of its smaller size than Ni²⁺, the content of Li₂MnO₃ decreases as a result of the decrease of Li and Mn in the TM layers, and the discharge voltage increases on account of the higher reduction potential of Ni⁴⁺/Ni³⁺ compared with Ni⁴⁺/Ni²⁺ and Mn⁴⁺/Mn³⁺. Furthermore, Li_1.2Ni_0.4Mn_0.4O₂ delivers a first capacity value of 192.5 mA h g⁻¹ and operating voltage of 3.89 V at 0.1C, and capacity retention of 92.13% and a voltage drop of 0.25 V after 100 cycles, which indicates that the stability of capacity and voltage is improved by a large margin with increasing Ni³⁺.

---

1. Introduction

Rechargeable lithium-ion secondary batteries are considered as the most valuable energy storage devices for mobile phones, laptops and electric vehicles due to high energy density, safety and long cycle life.^1,2 Nowadays, the most conventional cathode materials are layered oxides LiMO₂ (M = Ni, Co and Mn) with excellent cyclability, olivine LiFePO₄ with low cost and long cycle life, and spinel oxides LiMn₂O₄ and LiNi_0.5Mn_1.5O₄ with high voltage, while all of them suffer from low discharge capacity and fail to meet the requirements of high energy density for practical applications.^3,4

Recently, a layered cathode material with a chemical formula of xLi₂MnO₃–(1 − x)LiMO₂ (M = Ni, Co and Mn) has attracted wide attention owing to its higher discharge capacity (more than 250 mA h g⁻¹) and much higher operating voltage (about 3.5 V).^5,6 However, there are many remaining problems have to be solved, such as low initial coulombic efficiency, poor rate performance, serious capacity fading and voltage decay. The low initial coulombic efficiency is associated with the release of irreversible oxygen along with the activation of Li₂MnO₃ during the first charge process, resulting in the formation of oxide ion vacancies.⁷ The poor rate performance is related to the weak intrinsic electronic conductivity resulted from insulated Li₂MnO₃ component and the thick solid-electrolyte interphase (SEI) layer.⁸ The serious capacity fading and voltage decay should be attributed to the fast layer-spinel phase transformation, which leads to a fast reduction in energy density.^3,9

Li_1.2Ni_0.2Mn_0.6O₂ cathode material has become a hotspot research because of its high initial discharge capacity recently,^10–13 while the problems of serious capacity fading and voltage decay caused by layered-spinel structural transformation upon cycling have not been solved radically. Thus, it is necessary to study the other cathode materials with a chemical formula of Li_1.2Ni_xMn_0.8−xO₂ and understand what factors affect the discharge capacity and voltage decay and further what measures should be taken to optimize the structure and ameliorate properties. It is reported that the existence of Mn³⁺ facilitates the activation of Li₂MnO₃ phase, which gives rise to a higher discharge specific capacity after the complete activation of Li₂MnO₃. Unfortunately, the gradual activation of Li₂MnO₃ induces faster layered-spinel transformation, leading to rapid capacity fading.^14,15 In addition, Knight et al. have investigated the effect of Ni oxidation state on the structural and electrochemical characteristics, proposing Ni³⁺ can improve the cyclability and mitigate voltage decay,¹⁶ while the valence of Ni and the actual content of Li haven't been measured by any characterization methods.

Herein, we report a layered oxides with a chemical formula of Li_1.2Ni_xMn_0.8−xO₂ (x = 0, 0.1, 0.2, 0.3 and 0.4) prepared by a simple combustion method and a detailed study of the oxidation states of TM on the structure and electrochemical performances for the Li-rich cathode materials by changing the relative proportion of Ni and Mn. It should be noted that, the phenomenon of capacity increase upon cycling is observed by reason of the existence of Mn³⁺, which leads to a much higher capacity and more excellent cyclability. In addition, the stability of capacity and voltage is improved by a large margin with increasing the Ni³⁺. The findings herein may shed light not only on the effects understanding of the oxidation states of TM in Li-rich oxides but also on the design and improvement measures of high energy density lithium-ion batteries by controlling the content of Mn³⁺ and Ni³⁺.

2. Experimental

2.1 Synthesis of Li_1.2Ni_xMn_0.8−xO₂

The Li_1.2Ni_xMn_0.8−xO₂ (x = 0, 0.1, 0.2, 0.3 and 0.4) cathode materials were synthesized by a facile combustion method using Li, Ni, Mn acetates and citric acid as starting materials. Stoichiometric amounts of C₂H₃O₂Li·2H₂O, C₄H₆NiO₄·4H₂O and C₄H₆MnO₄·4H₂O (Sinopharm Chemical Reagent) were dissolved in ethanol under magnetic stirring to form solution A and C₆H₈O₇·H₂O (Sinopharm Chemical Reagent) was dissolved in ethanol to form solution B. A 5% excess of Li was used to offset the loss of Li during sintering under high temperature and the molar ratio of metal and citric acid was 1 : 1. Then solution A and solution B were instilled separately into 100 ml ethanol base solution under continuous mechanical stirring. The base solution turned into white emulsion gradually as ethanol was evaporated at 80 °C and further dried at 110 °C for 12 h to obtain precursor. Finally the mixture was calcined at 450 °C for 5 h in air to remove the organics and then heated at 900 °C for 5 h in air with heating rates of 3 °C min⁻¹.

2.2 Material characterization

X-ray photoelectron spectra (XPS, Axis Ultrabld, Kratos) was used to detect the surface elemental chemical state of Li_1.2Ni_xMn_0.8−xO₂ (x = 0, 0.1, 0.2, 0.3 and 0.4) cathode materials. The actual elemental chemical composition of the samples was determined by a Varian 715-ES inductively coupled plasma atomic emission spectroscopy (ICP-AES). The crystal structures of prepared materials were identified by X-ray diffraction (XRD, UltimaIV-185, Rigaku) using Cu Kα radiation at 40 mA and 40 kV in the range of 10–80°.

2.3 Electrochemical characterization

The electrochemical performance of Li_1.2Ni_xMn_0.8−xO₂ (x = 0, 0.1, 0.2, 0.3 and 0.4) cathode materials was measured on a battery test system (Neware, China) using a CR2025-type coin-cell. The cathode electrode was fabricated by coating an aluminum foil with a slurry which consisted of 80 wt% active material, 10 wt% acetylene black and 10 wt% polyvinylidene fluoride (PVDF) in N-methyl pyrrolidinone (NMP). Then the aluminum foil was dried at 100 °C for 12 h in a vacuum oven and was punched out a round cathode electrode disk of diameter 12 mm. Coin cells were assembled in an argon-filled glove box (Super 1220/750, Mikrouna) using fabricated round electrode disk as cathode, lithium foil as anode, celgard 2300 porous polypropylene film as separator and 1 M LiPF₆ in EC/DMC/DEC (1 : 1 : 1, vol%) as electrolyte. The galvanostatic charge/discharge was measured with a current density of 20 mA g⁻¹ (0.1C) between 2.0 V and 4.8 V (vs. Li/Li⁺) at room temperature. Cyclic voltammetry (CV) was performed on a CHI750D electrochemical workstation in the potential range of 2.0–4.8 V at a scanning rate of 0.1 mV s⁻¹.

3. Results and discussion

3.1 Structure

From chemical point of view, the Mn and Ni ions in Li_1.2Ni_0.2Mn_0.6O₂ (x = 0.2) are expected to show the +4 and +2 oxidation states, respectively.¹¹ In order to maintain the chemical formula, the valence states of TM ions in the synthesized Li_1.2Ni_xMn_0.8−xO₂ are Li_1.2Mn_0.8^3+/4+O₂ (x = 0), Li_1.2Ni_0.1²⁺Mn_0.7^3+/4+O₂ (x = 0.1), Li_1.2Ni_0.3^2+/3+Mn_0.5⁴⁺O₂ (x = 0.3) and Li_1.2Ni_0.4³⁺Mn_0.4⁴⁺O₂ (x = 0.4), respectively.

XPS was used to determine the exact oxidation states of Mn and Ni in the layered composite material, the corresponding spectra are presented in Fig. 1. Mn_2p peaks of Li_1.2Mn_0.8O₂ (Fig. 1a), Li_1.2Ni_0.1Mn_0.7O₂ (Fig. 1b), Li_1.2Ni_0.2Mn_0.6O₂ (Fig. 1d), Li_1.2Ni_0.3Mn_0.5O₂ (Fig. 1f) and Li_1.2Ni_0.4Mn_0.4O₂ (Fig. 1h) can be fitted by multiple peaks with binding energies (BEs) (reported BEs of Mn³⁺ and Mn⁴⁺ are: 640.6 eV and 641.4–644.8 eV, respectively),¹⁴ the fitting results and relative percentage of Mn³⁺ and Mn⁴⁺ shown in Fig. 2. This means that the oxidation states of Mn for Li_1.2Ni_0.2Mn_0.6O₂, Li_1.2Ni_0.3Mn_0.5O₂ and Li_1.2Ni_0.4Mn_0.4O₂ present 4+. However, Mn ions in Li_1.2Ni_0.1Mn_0.7O₂ and Li_1.2Mn_0.8O₂ consist of Mn³⁺ (40.3%) and Mn⁴⁺ (59.7%), Mn³⁺ (49.6%) and Mn⁴⁺ (50.4%), respectively. The existence of Mn³⁺ can also be proved by the oxidation peak of Mn³⁺/Mn⁴⁺ during initial charge process in Fig. 4(b and c). The Ni spectra are shown in Fig. 1c (Li_1.2Ni_0.1Mn_0.7O₂), Fig. 1e (Li_1.2Ni_0.2Mn_0.6O₂), Fig. 1g (Li_1.2Ni_0.3Mn_0.5O₂) and Fig. 1i (Li_1.2Ni_0.4Mn_0.4O₂). Ni_2p3/2 peak fitting procedure reveals two types of BEs: 853.7–855 eV (ref. 17 and 18) and 855.3–855.9 eV.¹⁹ The lower BEs is in agreement with that of Li[Li_1/3−2x/3Ni_xMn_2/3−x/3]O₂ compounds, which confirms the presence of Ni²⁺, and the higher BEs is consistent with that of LiNiO₂, which proves the presence of Ni³⁺. The fitting results in Fig. 2 reveal that the valence states of Ni in Li_1.2Ni_0.1Mn_0.7O₂ and Li_1.2Ni_0.2Mn_0.6O₂ are all +2. However, Ni ions in Li_1.2Ni_0.3Mn_0.5O₂ and Li_1.2Ni_0.4Mn_0.4O₂ are composed of Ni²⁺ (87.9%) and Ni³⁺ (12.1%), Ni²⁺ (19.9%) and Ni³⁺ (80.1%), respectively. Therefore, the experimentally determined oxidation states of TM ions are basically consistent with the theoretical ones.

Fig. 1 Fitting results of XPS spectra of (a) Mn_2p for Li_1.2Mn_0.8O₂. (b and c) Mn_2p and Ni_2p for Li_1.2Ni_0.1Mn_0.7O₂. (d and e) Mn_2p and Ni_2p for Li_1.2Ni_0.2Mn_0.6O₂. (f and g) Mn_2p and Ni_2p for Li_1.2Ni_0.3Mn_0.5O₂. (h and i) Mn_2p and Ni_2p for Li_1.2Ni_0.4Mn_0.4O₂.

Fig. 2 Average valence states of transition metal in Li_1.2Ni_xMn_0.8−xO₂.

ICP-AES analysis results of Li_1.2Ni_xMn_0.8−xO₂ (x = 0, 0.1, 0.2, 0.3 and 0.4) are shown in Table 1. The accurate elemental ratios of Ni/Mn are in excellent agreement with nominal values and the relative lithium content of Li_1.2Ni_0.2Mn_0.6O₂ are well close to target stoichiometries. However, the relative lithium contents of all samples decreased with increasing x value. Owing to the appearance of Ni³⁺ and the difficulty in maintaining Ni³⁺ content, more than 5% extra Li was used to compensate the Li loss during high temperature heating for Li_1.2Ni_0.3Mn_0.5O₂ and Li_1.2Ni_0.4Mn_0.4O₂.¹⁶ Furthermore, Li_1.2Mn_0.8O₂ required little excess of Li and Li_1.2Ni_0.1Mn_0.7O₂ required less than 5% of Li to counteract the Li loss for producing the desired composition thanks to the fact that more Mn⁴⁺ was replaced by Mn³⁺ and there's no need to maintain the Mn⁴⁺ to increase in the degree of lithium volatilization.

Table 1 ICP-AES analysis results and lattice parameters and the ratios of I₍₀₀₃₎/I₍₁₀₄₎ calculated from the XRD data

x	Sample	ICP results	a/Å	c/Å	c/a	I(003)/I(104)
0	Li1.2Mn0.8O2	Li1.249Mn0.8O2	2.8647	14.2159	4.962	1.781
0.1	Li1.2Ni0.1Mn0.7O2	Li1.226Ni0.1Mn0.702O2	2.8737	14.2984	4.976	0.671
0.2	Li1.2Ni0.2Mn0.6O2	Li1.204Ni0.2Mn0.599O2	2.8676	14.2765	4.979	0.948
0.3	Li1.2Ni0.3Mn0.5O2	Li1.114Ni0.3Mn0.501O2	2.8779	14.2874	4.965	1.287
0.4	Li1.2Ni0.4Mn0.4O2	Li0.991Ni0.4Mn0.399O2	2.8799	14.2990	4.965	1.433

---

Fig. 3a shows the XRD patterns of Li_1.2Ni_xMn_0.8−xO₂ (x = 0, 0.1, 0.2, 0.3 and 0.4). The main diffraction peaks can be indexed as a hexagonal α-NaFeO₂ layered structure with the R3̄m space group.

Fig. 3 (a) XRD patterns of Li_1.2Ni_xMn_0.8−xO₂ (x = 0, 0.1, 0.2, 0.3 and 0.4), respectively. (b) Expanded regions of XRD patterns of all samples within a 2θ range of 20.0–22.5°, respectively.

The clear splitting of (006)/(102) and (018)/(110) doublets indicate the formation of a highly ordered layered structure for all samples.^20,21 The lattice parameters calculated from the XRD data are listed in Table 1. The c/a ratios of all samples are more than 4.9, which further provides the evidence for an explicit crystalline layer structure.^22,23 Moreover, the ratio of I₍₀₀₃₎/I₍₁₀₄₎ can be used to evaluate the degree of the cation mixing between Li⁺ and Ni²⁺ in the Li-layers, and the value of I₍₀₀₃₎/I₍₁₀₄₎ > 1.2 is an indication of low degree of cation dislocation.^24,25 It is indicated that there is no cation mixing in Li_1.2Mn_0.8O₂ because the content of Ni is 0. However, the values of I₍₀₀₃₎/I₍₁₀₄₎ for Li_1.2Ni_0.1Mn_0.7O₂ and Li_1.2Ni_0.2Mn_0.6O₂ are respectively 0.671 and 0.948, which are both lower than 1.2, indicating more cation dislocation between Li⁺ and Ni²⁺. For Li_1.2Ni_0.3Mn_0.5O₂ and Li_1.2Ni_0.4Mn_0.4O₂, the values of I₍₀₀₃₎/I₍₁₀₄₎ are higher than 1.2, indicating lower level of cation disordering. As can be seen from above, the ratio of (003)/(104) peak increases as the Ni³⁺ increases because Ni³⁺ ions (0.56 Å) are much smaller than Ni²⁺ ions (0.69 Å) and Li⁺ ions (0.76 Å), which leads to lower Li/Ni site mixing.¹⁶

The weak peaks between 20 and 22.5° are attributed to short-range superlattice ordering of Li and Mn in the TM layers of the layered Li₂MnO₃ region, which can be indexed to the monoclinic unit cell with the C2/m space group.^26–28 In the magnified XRD patterns (20.0–22.5°) shown in Fig. 3b, the intensity of these weak peaks appear much more intense with increasing the content of Mn, resulting from a few Li loss during high temperature heating process and more Li content in the TM layers. However, a cubic spinel structure with a space group of Fd3m can be detected in the XRD pattern (Fig. 1a) for Li_1.2Mn_0.8O₂.

3.2 Electrochemical performance

3.2.1 Initial charge and discharge behavior. To evaluate the effect of valence states of Ni and Mn on the initial electrochemical performance of Li_1.2Ni_xMn_0.8−xO₂ (x = 0, 0.1, 0.2, 0.3 and 0.4), the initial charge–discharge curves and corresponding cyclic voltammetry profiles of all the five samples measured at a constant current density of 20 mA g⁻¹ (0.1C) between 2.0 V and 4.8 V are shown in Fig. 4. As normal Li_1.2Ni_0.2Mn_0.6O₂ layered oxides, a sloping and a plateau are observed during the initial charge process in Fig. 4a. The sloping curve below 4.5 V is ascribed to the lithium extraction from LiMO₂ (M = Ni, Mn) component accompanied by oxidation of Ni²⁺/Ni³⁺/Ni⁴⁺. And the long plateau above 4.5 V is attributed to the lithium extraction from Li₂MO₃ component with the reaction of O²⁻/O₂ redox couple,^29,30 which represents the activation of Li₂MO₃. It can be seen obviously that Li_1.2Ni_0.2Mn_0.6O₂ delivers the highest initial charge and discharge capacity in Fig. 4a. The corresponding CV curve shows the oxidation peak of Ni²⁺/Ni³⁺/Ni⁴⁺ at about 3.85 V as well as the irreversible activation process of Li₂MO₃ region at about 4.7 V during initial charge process, and a broad Ni reduction peak at about 3.7 V as well as a very weak reduction peak corresponding to Mn⁴⁺/Mn³⁺ couple at approximately 3.2 V during initial discharge process in Fig. 4d.

Fig. 4 (a) Initial charge and discharge curves of all five samples at 0.1C between 2.0 and 4.8 V. (b–f) Corresponding cyclic voltammetry profiles.

However, the capacity of sloping below 4.5 V and plateau above 4.5 V decrease as the content of Mn³⁺ increases, which leads to a much lower initial charge and discharge capacity for Li_1.2Mn_0.8O₂ and Li_1.2Ni_0.1Mn_0.7O₂. For Li_1.2Ni_0.1Mn_0.7O₂, the much lower sloping capacity is most likely attributable to the serious Li/Ni site mixing. Also, the appearance of Mn³⁺ suppresses the complete activation of Li₂MO₃,¹⁴ resulting in shorter plateau above 4.5 V. The most of peaks for the CV curve of Li_1.2Ni_0.1Mn_0.7O₂ in Fig. 4c are similar to Li_1.2Ni_0.2Mn_0.6O₂, while an oxidation peak at approximately 3.25 V of Mn³⁺/Mn⁴⁺ is observed, which provides an evidence of Mn³⁺ ion formation in consistent with the result of XPS. For Li_1.2Mn_0.8O₂, the much lower capacity can be attributed to two factors. On the one hand, the presence of Li_1+xMn_2−xO₄ with the spinel structure leads to the diminished sloping capacity.³¹ On the other hand, Mn³⁺ and spinel structure suppress irreversible extraction of Li⁺ and O²⁻ from Li₂MnO₃ component during the first charge process.^14,31 Furthermore, the CV of Li_1.2Mn_0.8O₂ is particularly informative. Two intense oxidation peaks of Mn³⁺/Mn⁴⁺ at 3.1 V and 4.25 V are observed clearly, characteristic of a Li–Mn oxide spinel phase in the system Li_1+xMn_2−xO₄,^30,32 except for the normal oxidation peak of Mn³⁺/Mn⁴⁺ in layer LiMnO₂ at about 3.8 V and a much weaker irreversible activation of Li₂MO₃ at about 4.7 V during initial charge process in Fig. 4b, which is consistent with the result of XPS and XRD.

Obviously, the capacity of sloping below 4.5 V becomes higher with the increase of Ni³⁺ on account of much lower level of Li/Ni cation mixing. On the contrary, the capacity of plateau above 4.5 V decreases with increasing the content Ni³⁺, resulting in a much lower plateau capacity for Li_1.2Ni_0.3Mn_0.5O₂ and disappearance of plateau for Li_1.2Ni_0.4Mn_0.4O₂. This is most likely due to the reduction in the Li⁺ and Mn⁴⁺, resulting in decreasing the Li₂MnO₃ component.¹⁶ The CV profiles show the oxidation peaks of Ni²⁺/Ni³⁺/Ni⁴⁺ at about 4.2 V and 4.3 V for Li_1.2Ni_0.3Mn_0.5O₂ (Fig. 4e) and Li_1.2Ni_0.4Mn_0.4O₂ (Fig. 4f), respectively. However, the corresponding reduction peak splits into two separate peaks because the voltage difference between Ni³⁺/Ni²⁺ and Ni⁴⁺/Ni³⁺ redox couples.³³ Moreover, the peak of Li₂MnO₃ activation can't be observed for Li_1.2Ni_0.4Mn_0.4O₂ during the initial charge process, in accordance with the disappearance of plateau and the results of XRD. In addition, it is clear that the discharge median voltage in the initial cycle increases with increasing the content of Ni³⁺, because Ni³⁺ shows the much higher redox potential compared with Ni²⁺ and Mn³⁺.

3.2.2 Cyclic performance. The cyclic performances of Li_1.2Ni_xMn_0.8−xO₂ (x = 0, 0.1, 0.2, 0.3 and 0.4) at 20 mA g⁻¹ (0.1C) in the voltage range of 2.0–4.8 V for the first 100 cycles are shown in Fig. 5a. Li_1.2Ni_0.2Mn_0.6O₂ exhibits the highest initial discharge capacity of 244 mA h g⁻¹, while it shows only capacity retention of 62.97% after 100 cycles, which indicates a fast structure transformation from layer to spinel.³⁴ However, the discharge capacity of Li_1.2Mn_0.8O₂ and Li_1.2Ni_0.1Mn_0.7O₂ shows a trend of gradual increase by reason of the gradual activation of Li₂MnO₃ upon cycling. It should be noted that, Li_1.2Mn_0.8O₂ presents a much slower growth rate that the discharge capacity increases from 22.08 mA h g⁻¹ to 39.13 mA h g⁻¹ after 100 cycles, which is attributed to the slow activation rate of Li₂MnO₃ suppressed by existing spinel phase Li_1+xMn_2−xO₄.³¹ And for Li_1.2Ni_0.1Mn_0.7O₂, the discharge capacity reaches the highest value of 231.5 mA h g⁻¹ from the initial value of 76 mA h g⁻¹ after 75 cycles and still retains 100% of the capacity after 100 cycles. This phenomenon of capacity increase in the initial dozens of cycles is also ascribed to the gradual activation of Li₂MnO₃ caused by the existence of Mn³⁺.

Fig. 5 (a) Cycling performances of all five samples at 0.1C between 2.0 V and 4.8 V. (b) Energy density curves of all the samples upon cycling at 0.1C.

It is noted that Li_1.2Ni_0.3Mn_0.5O₂ and Li_1.2Ni_0.4Mn_0.4O₂ show excellent capacity retention with 75.04% and 92.13% after 100 cycles, respectively. The reason for this superior cyclic stability, which generally enhances as Ni³⁺ content increases, involves two aspects. One reason can be the reduction in Mn ion dissolution upon cycling resulting from decreasing the content of Mn, which leads to less capacity loss.^16,35 The other reason is the reduction in Li ion irreversible dissolution and elimination of more Li vacancies upon cycling, for the smaller in size of Ni³⁺ than Ni²⁺ can't occupy the sites of Li layers after charge process, which can protect the active materials from structure transformation.¹⁶

Voltage decay caused by the structure conversion from layer to spinel phase and the deterioration of the electrode/electrolyte interface is one of the main challenges for Li-rich cathode materials, which leads to the gradual decrease in energy density during cycling.³⁶ The charge–discharge curves of all five samples at various cycle numbers at 0.1C are employed to evaluate the voltage decay, as shown in Fig. 6, respectively. Severe voltage decay upon cycling is observed in Li_1.2Mn_0.8O₂ (Fig. 6a), Li_1.2Ni_0.1Mn_0.7O₂ (Fig. 6c) and Li_1.2Ni_0.2Mn_0.6O₂ (Fig. 6e), resulting from the unexpected layered-spinel structure transformation.³⁷ In particular, Li_1.2Mn_0.8O₂ exhibits an interesting gradual increase phenomenon of the discharge median voltage in the initial tens of cycles from 3.36 V to 3.60 V, for the proportion of LiMnO₂ with higher redox potential of Mn³⁺/Mn⁴⁺ than Li_1+xMn_2−xO₄ increases gradually along with the activation of Li₂MnO₃.³² However, the voltage fading rate efficiently slowed down with the increase of Ni³⁺ content, especially for Li_1.2Ni_0.4Mn_0.4O₂ with the voltage drop of 0.25 V. The alleviated voltage decay can be attributed to the enhancement of structural stability, the elimination of Li vacancies resulting from lower level of TM migration after charge process, the reduced Mn dissolution upon cycling and the weak interface reaction between electrode and electrolyte contributed by the reduction in oxygen release as the increase of Ni³⁺.^16,37

Fig. 6 Charge and discharge curves of (a) Li_1.2Mn_0.8O₂, (c) Li_1.2Ni_0.1Mn_0.7O₂, (e) Li_1.2Ni_0.2Mn_0.6O₂, (g) Li_1.2Ni_0.3Mn_0.5O₂ and (i) Li_1.2Ni_0.4Mn_0.4O₂ at 0.1C between 2.0 V and 4.8 V after the 2^nd, 5^th, 10^th, 20^th, 50^th and 100^th cycle. (b, d, f, h and j) Corresponding cyclic voltammetry profiles.

Fig. 6(b, d, f, h and j) presents the CV curves of all materials after various cycles. It is obviously observed that the weak oxidation peaks of Li₂O removal from Li₂MnO₃ at 4.8 V always exist during 100 cycles for Li_1.2Mn_0.8O₂ (Fig. 6b), which proves the phenomenon of capacity increase upon cycling. Moreover, the oxidation peaks of Mn³⁺/Mn⁴⁺ in spinel phase at about 3.2 V continuously shift to a higher voltage upon cycling, and the reduction peaks of Mn⁴⁺/Mn³⁺ in layer phase become weaker as well as a new reduction peak of 2.95 V occurs. This behavior can be interpreted that an unexpected layered-spinel intergrowth structure formation generates voltage decay and low capacity.¹¹ As shown in Fig. 6d, a weak peak of oxygen release still appears in the 50th during charge process and disappears after 100 cycles, which verifies the interesting phenomenon of capacity change for Li_1.2Ni_0.1Mn_0.7O₂. In particular, the redox peak of Mn⁴⁺/Mn³⁺ and the peak of Li₂MnO₃ activation cannot be observed all the time for Li_1.2Ni_0.4Mn_0.4O₂, indicating that traces of inactivated Li₂MnO₃ in active materials maybe enhance the structure and improve the voltage decay.

Fig. 5b shows the energy density profiles upon cycling, which are determined by capacity loss and voltage decay. Li_1.2Ni_0.2Mn_0.6O₂ delivers the highest initial energy density of 873 W h kg⁻¹ and the lowest retention value of 446 W h kg⁻¹ after 100 cycles. However, the gradual increase of capacity for Li_1.2Ni_0.1Mn_0.7O₂ and the excellent stability of capacity and voltage for Li_1.2Ni_0.4Mn_0.4O₂ exhibit much higher energy density of 660 W h kg⁻¹ and 645 W h kg⁻¹ after 100 cycles, respectively.

4. Conclusions

In this paper, a monoclinic-hexagonal layered material with a chemical formula of Li_1.2Ni_xMn_0.8−xO₂ (x = 0, 0.1, 0.2, 0.3 and 0.4) was prepared by a combustion method to investigate the effect of valence states of Ni and Mn on the structure and electrochemical performances. A certain amount of Mn³⁺ in Li-rich cathode materials can be a useful approach to obtain the higher capacity and superior cyclability after the complete activation of Li₂MnO₃, while it will cause serious voltage decay. For example, Li_1.2Ni_0.1Mn_0.7O₂ reaches the highest discharge capacity of 231.5 mA h g⁻¹ from the initial value of 76 mA h g⁻¹ after 75 cycles and still keeps 100% of the capacity at 0.1C after 100 cycles, while the operating voltage drops from 3.60 V to 2.85 V after 100 cycles. In addition, Ni³⁺ plays an important role in improving the stability of capacity and voltage. And operating voltage increases with the content of Ni³⁺ because of the higher reduction potential of Ni⁴⁺/Ni³⁺ compared with Ni⁴⁺/Ni²⁺ and Mn⁴⁺/Mn³⁺. But Ni³⁺ will give rise to a lower discharge capacity for a little oxygen release. For instance, Li_1.2Ni_0.4Mn_0.4O₂ delivers the discharge capacity of 192.5 mA h g⁻¹ in the first cycle at 0.1C with the capacity retention of 92.13%, and initial operating voltage of 3.89 V with the voltage drop of 0.25 V after 100 cycles. Therefore, it is proposed that the amounts of Mn³⁺ and Ni³⁺ need to be well controlled during synthesis of Li-rich layered oxides to reach a reasonable integrated performance between high capacity, high operating voltage and excellent stability of capacity and voltage.

Acknowledgements

This work was supported by the China Postdoctoral Science Foundation (no. 2012M521760) and the Fundamental Research Funds for the Central Universities (no. xjj2014052).

References

1. M. S. Whittingham, Chem. Rev., 2004, 104, 4271–4302.
2. G.-Y. Kim, S.-B. Yi, Y. J. Park and H.-G. Kim, Mater. Res. Bull., 2008, 43, 3543–3552.
3. J. Yan, X. Liu and B. Li, RSC Adv., 2014, 4, 63268–63284.
4. P. R. Ilango, T. Subburaj, K. Prasanna, Y. N. Jo and C. W. Lee, Mater. Chem. Phys., 2015, 158, 45–51.
5. J. Zheng, M. Gu, A. Genc, J. Xiao, P. Xu, X. Chen, Z. Zhu, W. Zhao, L. Pullan and C. Wang, Nano Lett., 2014, 14, 2628–2635.
6. M. M. Thackeray, C. S. Johnson, J. T. Vaughey, N. Li and S. A. Hackney, J. Mater. Chem., 2005, 15, 2257–2267.
7. J. Zheng, P. Xu, M. Gu, J. Xiao, N. D. Browning, P. Yan, C. Wang and J.-G. Zhang, Chem. Mater., 2015, 27, 1381–1390.
8. M. Hou, J. Liu, S. Guo, J. Yang, C. Wang and Y. Xia, Electrochem. Commun., 2014, 49, 83–87.
9. Z. Lu and J. R. Dahn, J. Electrochem. Soc., 2002, 149, A815–A822.
10. Y. Liu, Q. Wang, X. Wang, T. Wang, Y. Gao, M. Su and A. Dou, Ionics, 2015, 21, 2725–2733.
11. T. Zhao, L. Li, R. Chen, H. Wu, X. Zhang, S. Chen, M. Xie, F. Wu, J. Lu and K. Amine, Nano Energy, 2015, 15, 164–176.
12. D. Wang, Y. Huang, Z. Huo and L. Chen, Electrochim. Acta, 2013, 107, 461–466.
13. D. Wang, I. Belharouak, G. Zhou and K. Amine, Adv. Funct. Mater., 2013, 23, 1070–1075.
14. L. Xiao, J. Xiao, X. Yu, P. Yan, J. Zheng, M. Engelhard, P. Bhattacharya, C. Wang, X.-Q. Yang and J.-G. Zhang, Nano Energy, 2015, 16, 143–151.
15. X. Zhang, D. Luo, G. Li, J. Zheng, C. Yu, X. Guan, C. Fu, X. Huang and L. Li, J. Mater. Chem. A, 2013, 1, 9721–9729.
16. J. C. Knight and A. Manthiram, J. Mater. Chem. A, 2015, 3, 22199–22207.
17. L. Peng, Y. Zhu, U. Khakoo, D. Chen and G. Yu, Nano Energy, 2015, 17, 36–42.
18. B. Li, C. Li, J. Cai and J. Zhao, J. Mater. Chem. A, 2015, 3, 21290–21297.
19. M. E. Spahr, P. Novák, B. Schnyder, O. Haas and R. Nesper, J. Electrochem. Soc., 1998, 145, 1113–1121.
20. Z.-L. Gong, H.-S. Liu, X.-J. Guo, Z.-R. Zhang and Y. Yang, J. Power Sources, 2004, 136, 139–144.
21. M. Xu, Z. Chen, H. Zhu, X. Yan, L. Jun Li and Q. Zhao, J. Mater. Chem. A, 2015, 3, 13933–13945.
22. J. Lin, D. Mu, Y. Jin, B. Wu, Y. Ma and F. Wu, J. Power Sources, 2013, 230, 76–80.
23. K. Shaju, G. S. Rao and B. Chowdari, Electrochim. Acta, 2002, 48, 145–151.
24. H. Zhang, Q. Qiao, G. Li, S. Ye and X. Gao, J. Mater. Chem., 2012, 22, 13104–13109.
25. B. Hwang, Y. Tsai, D. Carlier and G. Ceder, Chem. Mater., 2003, 15, 3676–3682.
26. K. A. Jarvis, Z. Deng, L. F. Allard, A. Manthiram and P. J. Ferreira, Chem. Mater., 2011, 23, 3614–3621.
27. S.-H. Kang, P. Kempgens, S. Greenbaum, A. Kropf, K. Amine and M. Thackeray, J. Mater. Chem., 2007, 17, 2069–2077.
28. L. Xu, P. Hou, Y. Zhang, H. Zhang, D. Song, X. Shi, X. Wang and L. Zhang, J. Mater. Chem. A, 2015, 3, 21219–21226.
29. J.-S. Kim, C. S. Johnson, J. T. Vaughey, M. M. Thackeray, S. A. Hackney, W. Yoon and C. P. Grey, Chem. Mater., 2004, 16, 1996–2006.
30. M. Thackeray, A. De Kock, M. Rossouw, D. Liles, R. Bittihn and D. Hoge, J. Electrochem. Soc., 1992, 139, 363–366.
31. D. Ye, K. Ozawa, B. Wang, D. Hulicova-Jurcakova, J. Zou, C. Sun and L. Wang, Nano Energy, 2014, 6, 92–102.
32. J. R. Croy, D. Kim, M. Balasubramanian, K. Gallagher, S.-H. Kang and M. M. Thackeray, J. Electrochem. Soc., 2012, 159, A781–A790.
33. L. Wang, G. Liu, W. Wu, D. Chen and G. Liang, J. Mater. Chem. A, 2015, 3, 19497–19506.
34. L. Li, B. Song, Y. Chang, H. Xia, J. Yang, K. Lee and L. Lu, J. Power Sources, 2015, 283, 162–170.
35. S. Shi, J. Tu, Y. Zhang, Y. Zhang, C. Gu and X. Wang, Electrochim. Acta, 2013, 109, 828–834.
36. X. Liu, Q. Su, C. Zhang, T. Huang and A. Yu, ACS Sustainable Chem. Eng., 2015, 4, 255–263.
37. B. Li, C. Li, J. Cai and J. Zhao, J. Mater. Chem. A, 2015, 8, 3308–3318.

---