Na-ion Batteries. Laure Monconduit

of charge/discharge curves for Li//LiCoO₂ and Na//NaCoO₂ cells. A schematic illustration of the crystal structure of isostructural LiCoO₂ and NaCoO₂ is also shown. Reproduced with permission from Yabuuchi et al. (2014). Copyright 2014, American Chemical Society. For a color version of this figure, see www.iste.co.uk/monconduit/batteries.zip

Since our group first demonstrated hard carbon//NaNi_1/2Mn_1/2O₂ full cells exhibiting acceptable cycle stability in 2009 (Komaba et al. 2009) and the results were published in Advanced Functional Materials in 2011 (Komaba et al. 2011), yearly numbers of scientific papers on sodium batteries have rapidly increased (Kubota et al. 2018a) and significant efforts have been devoted to attaining high-voltage and high-capacity positive electrode materials for Na-ion batteries (Delmas et al. 1982; Pan et al. 2013; Slater et al. 2013; Xu et al. 2013; Yabuuchi et al. 2014; Clement et al. 2015; Kundu et al. 2015; Fang et al. 2016; Kim et al. 2016; Deng et al. 2017; Hwang et al. 2017b; Ortiz-Vitoriano et al. 2017).

      Figure 1.2. Average voltage (V) and energy density (Wh kg⁻¹) versus gravimetric capacity (mAh g⁻¹) for selected positive electrode materials for Na-ion batteries. Energy density was calculated with the hard carbon (reversible capacity of 350 mAh g⁻¹ with E_ave = 0.3 V vs. Na metal) as negative electrode materials. Reproduced with permission from Kubota et al. (2018b). Copyright 2018, Wiley-VCH. For a color version of this figure, see www.iste.co.uk/monconduit/batteries.zip

Figure 1.2 shows a comparison of reversible capacity and working voltage among the positive electrode materials. The positive electrode materials for Na-ion batteries can be basically categorized into layered oxides, polyanionic compounds, and Prussian blue analogues. Polyanion materials seem to exhibit relatively higher working voltage and smaller capacities compared to the others. However, the high-voltage polyanion materials typically contain costly cobalt (Nose et al. 2013) or vanadium (Jian et al. 2012; Kang et al. 2012; Lim et al. 2014; Zhang et al. 2016) except for Na₂Fe₂(SO₄)₃ (Barpanda et al. 2014). On the other hand, Na-containing layered transition metal oxides exhibit relatively lower working voltage but larger capacities on the basis of redox activity of Fe³⁺/Fe⁴⁺ or manganese Mn³⁺/Mn⁴⁺. Because of the high-performance redox of iron and manganese, abundant mineral resources of iron and manganese are available, and the resultant cost-reduction will be thus advantageous for stationary applications of Na-ion batteries. We emphasize that the layered transition metal oxides meet this important demand for future application. Of course, higher energy density (i.e. large capacities and high working potential), long-term cycle life, high coulombic and energetic efficiencies, thermal-stability, and ease in handling and production, toxic-free chemistry and so on are desired for the practical use as positive electrode materials for Na-ion batteries.

      In this chapter, developments of Na-containing layered 3d transition metal oxides are reviewed for the application as active materials of Na-ion batteries based upon the authors’ experience since 2003 (Komaba 2019). The electrochemical performances, phase transitions during the charge/discharge, surface chemistry in the batteries, key factors influencing the battery performances and future prospective are discussed mainly based on our leading studies on the layered oxides since 2005.

1.2. Crystal structures of layered materials

      1.2.1. Crystal structures of synthesizable Na_xMO₂

An iron based oxide, α-NaFeO₂, is one of the most well-known Na-containing layered transition metal oxides in inorganic chemistry and in the battery research community. This is because LiCoO₂, LiNi_0.8Co_0.15Al_0.05O₂ and LiNi_1/3Mn_1/3Co_1/3O₂ used as positive electrode in commercialized Li-ion batteries are isostructural to α-NaFeO₂. That is, the layered rock salt–type structure with space group of R-3m is called α-NaFeO₂ type. α-NaFeO₂-type materials are found in NaMO₂ (M = Co, Cr, Fe, Ti, Sc, etc.), as shown in Figure 1.3. In contrast, α-NaFeO₂-type LiM’O₂ ones are crystallized only for M’ = Co, Ni, Cr and V, which can be explained by the difference in ionic radii between Li⁺ and Na⁺ ions (Shirane et al. 1995; Kanno et al. 1997). The ionic radius of a Li⁺ ion at an octahedral site of sixfold coordination is 0.76 Å and is quite close to those of transition metal ions (Shannon 1976). Because of the similar ion size, lithium ion is often mixed with transition metal ions during a solid-state synthesis reaction of LiMO₂, resulting in the formation of a cationordered rock salt–type (γ-LiFeO₂ type) or a cation-disordered rock salt–type (NaCl type) phase as seen in Figure 1.3 (Hoffmann Hoppe 1977; Shirane et al. 1995; Kanno et al. 1997). An ionic radius of Na⁺ (1.02 Å) is larger than those of Li⁺ and transition metal ions, leading to the obvious separation of the Na⁺ and transition metal layers. A variety of transition metals can be accommodated in α-NaFeO₂ type. This fact implies that cation mixing between Na⁺ and transition metal ions is suppressed and avoided in the synthesis of NaMO₂ and various transition metals are simultaneously adopted to form α-NaFeO₂-type solid solutions, which is advantageous for optimizing composition of Na_xMO₂ representing good electrochemical properties for Na-ion batteries.

      Figure 1.3. Structure field map of ABO₂ compounds. Modified with permission from Kanno et al. (1997). Copyright 1997 Elsevier. For a color version of this figure, see www.iste.co.uk/monconduit/batteries.zip

      A systematic notation system for layered transition metal oxides containing alkali metal was proposed by Delmas et al. (1977, 1980). Layered oxides of α-NaFeO₂ and α-NaCoO₂ are categorized into O3-type materials, and β- and γ-Na_xCoO₂ are P’3- and P2-type materials, respectively. Schematic illustrations of typical layered structures for sodium transition metal oxides were drawn using the program VESTA (Momma and Izumi 2011) and are shown in Figure 1.4 (Kubota et al. 2014).

In O3-type (α-NaFeO₂-type) structure, MO₂ slabs consisting of edge-shared MO₆ octahedra stack along c-axis with cubic close-packed oxygen as AB CA BC array and alkali metal ions are accommodated at octahedral sites in the interslab space. The number of MO₂ slabs is three in the hexagonal unit cell. Namely, O in O3-type means the octahedral site accommodating alkali metal ions and following 3 is the number of MO₂ slabs included in a hexagonal unit cell. When the hexagonal lattice is distorted into monoclinic or orthorhombic lattice, a prime symbol is added between the alphabet and number, but the number of MO₂ layers is counted in pseudohexagonal unit cells, such as O’3-type NaMnO₂ with a monoclinic lattice (space group [S.G.], C2/m) (Parant et al. 1971), P’3-type Na_xCoO₂ (S.G., C2/m) (Fouassier et al. 1973) and P’2-type Na_xMnO₂ with an orthorhombic lattice (S.G., Cmcm) (Parant et al. 1971). O3-type NaCoO₂ reversibly transforms into P3-type Na_xCoO₂ by electrochemical Na extraction during the charging process (Braconnier et al. 1980).

      Figure 1.4. Schematic illustrations of the crystal structures of O3-, P3-, and P2-type A_xMO₂.

Скачать книгу