Citation

Xiaohui Rong, Fei Gao, Yaxiang Lu, Kai Yang, Yongsheng Hu. P2-type Na_0.6[Mg(Ⅱ)_0.3Mn(Ⅳ)_0.7]O₂ as a new model material for anionic redox reaction[J]. Chin. Chem. Lett., 2018, 29(12): 1791-1794  

Xiaohui Rong^a,b,1, Fei Gao^a,1, Yaxiang Lu^b,*, Kai Yang^a, Yongsheng Hu^b     

With the growing demand for the higher energy density of lithium-ion batteries (LIBs), the anionic redox reaction (ARR), which could further improve the capacity even beyond the theoretical capacity based on the transition metal redox, has become a hot topic in recent years. Lithium-rich materials, delivering the capacity even more than 300 mAh/g, are the most representative cathode materials taken advantage of ARR during initial de-lithiation [1-3]. The long plateau during the initial delithiation at ~4.5 V is the typical electrochemical behavior of ARR for lithium-rich cathodes [1]. However, the irreversibility of ARR in most lithium-rich materials with only 3d transition metals hinders their application. The origin of the irreversibility is due to the irreversible phase transition from layered to spinel phase caused by the migration of transition metals [4]. In contrast, materials with 4d or 5d transition metals, offering stronger M—O bonds to stabilize the oxidized oxygen species, have relatively better ARR reversibility [5, 6]. In the case of Na-ion batteries (NIBs), which have attracted great attention for large-scale energy storage [7-9], ARR is also found in many layered oxide cathode materials, such as, P2-type Na_5/6[Li_0.25Mn_0.75]O₂, P2-type Na_0.67[Mg_0.28Mn_0.72]O₂ reported by Komaba et al. [10, 11], P3-type Na_0.6[Li_0.2Mn_0.8]O₂ reported by Goodenough et al. [12, 13], P2-type Na_0.6[Li_0.2Mn_0.8]O₂ reported by Nazar et al. [14], P2-type Na_0.78[Ni_0.23Mn_0.69]O₂ reported by Meng et al. [15], Na-rich Na₂[Ru_1-ySn_y]O₃ and Na₂IrO₃ reported by Tarascon et al. [16, 17]. To be noticed that the oxygen redox behavior in NIBs is different from that in LIBs, for some of the materials with only 3d transition metals have highly reversible ARR behavior [10, 12]. However, the present materials are still not enough for researchers to fully understand the ARR mechanism. One significant reason is that most synthesized materials, such as Na_5/6[Li_0.25Mn(Ⅲ)_0.08Mn(Ⅳ)_0.67]O₂ and Na_0.67[Mg_0.28Mn(Ⅲ)_0.11Mn (Ⅳ)_0.61]O₂, contain both Mn³⁺ and Mn⁴⁺, which will make the study of charge compensation more complicated as Mn³⁺ will also participate in the first initial charge process. Here, a layered oxide Na_0.6[Mg(Ⅱ)_0.3Mn(Ⅳ)_0.7]O₂ with only Mg²⁺ and Mn⁴⁺ in the transition metal layer is synthesized and investigated. The Mg²⁺ and Mn⁴⁺ are always considered unable to be further oxidized [13], thus oxygen should be the only element that possibly participates in the charge compensation process for the initial charge.

P2-type Na_0.6[Mg_0.3Mn_0.7]O₂ was prepared by a solid-state reaction using precursors of Na₂CO₃ (99.9%, Alfa), MgO (99%, Alfa), MnO₂ (99.9%, Alfa), and 2% more Na₂CO₃ was used for the compensation of volatilization loss during high temperature. Then the mixture was heated at 700 ℃ for 24 h in air and then naturally cooled to room temperature.

The structure was characterized using an X-ray diffractometer (XRD) (D8 Bruker) with Cu-K_α radiation (λ = 1.5405 Å) in the scan range (2θ) of 10°-80°. The morphologies of the samples were investigated by scanning electron microscopy (SEM) in combination with energy-dispersive X-ray spectroscopy (EDS)- mapping (Hitachi-S4800). The specific ratios of Na, Mg and Mn in the samples were measured by inductively coupled plasma atomic emission spectrometry (ICP-AES).

All the electrochemical tests were conducted using coin cells (CR2032), assembled in an argon filled glove-box. The working electrode was prepared by rolling out the mixture of active material (80 wt%), Super P (15 wt%) and polytetrafluoroethylene (PTFE 5 wt%) into thin films, followed by cutting into equal-area pieces and drying at 120 ℃ in vacuum for 10 h. The loading mass of active material on positive electrode was controlled to 6.0-7.0mg/cm². The electrolyte wasa solutionof1 mol/LNaClO₄ inethylene(EC), dimethyl carbonate (DMC), propylene carbonate (PC) (1:1:1 in volume) and fluoroethylene carbonate (FEC) (2% involume). Sodium foil was used as the counter electrode and glass fiber was used as the separator. The charge and discharge tests were carried out using a Land CT2001 A battery test system (Wuhan, China) in a voltage range of 1.5-4.4 V at various C-rates at room temperature.

The XRD pattern of the as-prepared material is shown in Fig. 1a. Na_0.6[Mg_0.3Mn_0.7]O₂ is crystallized in P2 phase, similar to previously reported P2-Na_0.67[Mg_0.28Mn_0.72]O₂, in which sodium ions are accommodated between MO₂ slabs as shown in Fig. 1a. All the diffraction peaks can be indexed to a hexagonal lattice with space group of P6₃/mcm, where the (100) and (102) peaks are superlattice line, suggesting the formation of in-plane ordering between Mg²⁺ and Mn⁴⁺. The SEM image in Fig. 1b shows that the crystallites of the compound are highly agglomerated with a wide size range of irregular particles.

Figure 1

	Download: JPG larger image
Fig. 1. Crystal structure and particle morphology. (a) X-ray diffraction pattern and (b) typical SEM image of as-prepared Na0.6[Mg0.3Mn0.7]O2 sample.

The quantitative analysis of the as-prepared Na_0.6[Mg_0.3Mn_0.7] O₂ sample was then carried out using ICP-AES, and the results are shown in Table 1. The experimental value is in good agreement with the designed stoichiometric of Na_0.6[Mg_0.3Mn_0.7]O₂, thus confirming that the valence of almost all manganese in the compound is +4. EDS mapping of Na_0.6[Mg_0.3Mn_0.7]O₂ was also conducted to confirm the elemental distribution. The elemental mappings of Na, Mg, and Mn are displayed in Fig. 2, exhibiting a uniform distribution of sodium, magnesium, and manganese in the sample.

Table 1

Table 1 Results of ICP-AES analysis of Na0.6[Mg0.3Mn0.7]O2 compound where Mg + Mn was normalized to be unity.

Table 1 Results of ICP-AES analysis of Na0.6[Mg0.3Mn0.7]O2 compound where Mg + Mn was normalized to be unity.

Figure 2

	Download: JPG larger image
Fig. 2. EDS mapping of Na0.6[Mg0.3Mn0.7]O2 sample. (a) SEM image; (b–d) corresponding elemental mapping of the Na, Mg, and Mn elements, respectively, with the selected area marked by the yellow rectangle in (a).

The electrochemical performance of the Na_0.6[Mg_0.3Mn_0.7]O₂ electrode was evaluated in sodium half cells at room temperature. Fig. 3a shows galvanostatic charge (Na deintercalation)/discharge (Na intercalation) curves of the first 20 cycles in the voltage range of 1.5-4.4 V versus Na⁺/Na at a current rate of 0.05C (10 mA/g). The initial charge capacity is ~130 mAh/g, showing a long plateau at ~4.2 V, corresponding to 0.44 e^- transfer. The initial discharge capacity is ~210 mAh/g, corresponding to 0.72 e^- transfer, with the average storage voltage of 2.5 V for the first discharge. The Coulombic efficiency of the second cycle is 97.9% and then increases to ~99.5% in the following cycles. The overall electrochemical behavior of Na_0.6[Mg_0.3Mn_0.7]O₂ is alike with that of Na_5/6[Li_0.25Mn_0.75]O₂ and Na_2/3[Mg_0.28Mn_0.72]O₂. For the initial charge, there is not a strong evidence so far to prove that Mn⁴⁺ can be further oxidized to Mn⁵⁺ in NIBs, thus indicating the 4.2 V voltage plateau during the initial charge is probably induced by ARR. The long slope region during discharge involves both ARR (partly reversible) and Mn redox reaction (Mn⁴⁺ → Mn³⁺) at lower voltage. For the following cycles, the 4.2 V voltage plateau still exists, indicating the oxygen redox is reversible. Notable voltage hysteresis of the material is observed after the initial charge, which is a typical character of many Li-rich materials with 3d transition metals [4-6].

Figure 3

	Download: JPG larger image
Fig. 3. Electrochemical performance of Na0.6[Mg0.3Mn0.7]O2. (a) The charge-discharge curves; (b) capacity & Coulombic efficiency Na0.6[Mg0.3Mn0.7]O2versus cycle number at a current rate of 0.05C in the voltage range of 1.5–4.4 V versus Na+/Na; and (c) rate capability of Na0.6[Mg0.3Mn0.7]O2.

Cycling and rate performance are also tested in half-cells with Na metal as the negative electrode. After 50 cycles, the capacity retention is only ~50% with Coulombic efficiency values scatted after ~40 cycles, which is a sign of electrolyte decomposition and drying up and indicates the severe side-reaction at electrode/electrolyte interface [13]. Rate capability of Na_0.6[Mg_0.3Mn_0.7]O₂ is tested from 1/20C to 2C and the corresponding specific discharge capacities are 212 mAh/g, 196 mAh/g, 156 mAh/g, 110 mAh/g and 52 mAh/g. With the increase of C-rate, voltage hysteresis becomes severer and ARR plateau degenerates obviously. The inferior rate capability could be related to the sluggish kinetics of ARR. Effective strategies should be further explored to solve problems of these type materials.

The differential capacity curves of the first and second cycle are shown in Fig. 4. There are two pairs of oxidation/reduction peaks: the first pair at 4.2 V/2.8 V and the second pair at 2.0 V/2.0 V. The first pair of oxygen redox shows large polarization, which is similar to the phenomenon in Li-air batteries, while the second Mn⁴⁺/Mn³⁺ redox pair displays small polarization. The other small peaks could be caused by the Na/vacancy ordering, which is usually found in many manganese-based NIB materials. Notably, there is only one peak in the first charge process, once again confirming that only oxygen redox participates into the charge compensation mechanism. In the second cycle the oxygen redox peak shifts slightly to lower voltage, which might be caused by the involvement of Mn⁴⁺/Mn³⁺ redox during the second charge process, indicating that the Jahn-Teller distortion introduced by Mn³⁺ could have subtle effects on the ARR activity and its reversibility.

Figure 4

	Download: JPG larger image
Fig. 4. Differential capacity curves of the first and second cycles of the Na0.6[Mg0.3Mn0.7]O2 electrode.

In conclusion, a P2-type layered oxide only containing Na⁺, Mg²⁺, Mn⁴⁺ and O^2- has been synthesized and investigated as cathode for NIBs. The Mg²⁺ and Mn⁴⁺ are always considered unable to be further oxidized, thus the charge compensation should be contributed by ARR itself during the initial charge. In the first charge, a long voltage plateau at ~4.2 V with a capacity of ~130 mAh/g is observed. In the following cycles, both oxygen and manganese redox pairs participate in the electrochemical process with a reversible capacity of ~210 mAh/g, and the oxygen redox behavior is found to be highly reversible. The simple elementary composition and highly reversible oxygen redox behavior make Na_0.6[Mg_0.3Mn_0.7]O₂ a suitable model material for the further investigation of reversible oxygen redox mechanism in NIBs.

This work was supported by funding from Science and Technology Project of the State Grid Corporation of China ("research on key technology of low-strain layered oxides for long-life Na-ion batteries", No. DG71-16-027).