patterns that coulombic attractive interactions between the slabs are enhanced by the existence of ordered alkali vacancies, i.e. the [Ru2/3□1/3]O2 honeycomb structure formed by Na extraction during charge, leading to reduction of stacking faults and progressive ordered stacking of the [Ru2/3□1/3]O2 slabs upon charging. The cooperatively ordered vacancies generate the electro-active nonbonding 2p orbitals of neighboring oxygen anions and stabilize the phase transformation for highly reversible oxygen-redox reactions (Mortemard de Boisse et al. 2019). Although the operation voltage of O3-Na2RuO3 is relatively low as a positive electrode material and further studies are required to enhance the oxygen-redox capacity for Na2Mn3O7, the findings provide a compelling future research direction toward reversible oxygen-redox positive electrode materials for high energy density batteries (Mortemard de Boisse et al. 2018).
Figure 1.11. (a) Potential profile (second cycle) of Na2Mn3O7 upon (de)intercalation in 1.5–4.7 V versus Na/Na+. The dashed gray box highlights the high voltage region, where O is the active redox process. (b) Galvanostatic charge/discharge curves in 3.0–4.7 V versus Na/Na+ at a rate of C/20. The inset shows a cyclic voltammetry curve at the second cycle at a scan rate of 0.1 mV s−1. Reprinted with permission from Mortemard de Boisse et al. (2018). Copyright 2018, Wiley-VCH. For a color version of this figure, see www.iste.co.uk/monconduit/batteries.zip
1.3.1.4. O3-NaFeO2
Another elementally attractive material in terms of abundant resources is O3-NaFeO2. Although O3-NaFeO2 is well known as a typical structural type of α-NaFeO2, the electrochemical properties were first reported by Takeda et al. in 1994 (Takeda et al.1994) and O3-NaFeO2 delivers a discharge capacity of ca. 120 mAh g−1 based on Fe3+/4+ redox in the Li cell. The electrochemical Fe3+/4+ redox activity was a surprising fact because O3-LiFeO2 delivers a reversible capacity of <10 mAh g−1 and is electrochemically inactive unlike O3-NaFeO2. Moreover, O3-LiFeO2 cannot be directly synthesized by a solid-state reaction and is prepared by Li+/Na+ ion-exchange from O3-NaFeO2 (Ado et al. 1997) due to the relatively large ionic radius of Fe3+ as shown in Figure 1.3. As is the case in the Li cell, O3-NaFeO2 is electrochemically active in a Na cell and delivers a reversible capacity of ca. 80 mAh g−1 based on Fe3+/4+ redox with a flat potential plateau at ca. 3.3 V versus Na+/Na in the voltage range of 2.5–3.4 V (Figure 1.8) (Takahashi et al. 2004; Okada et al. 2006; Yabuuchi et al. 2012b; Zhao et al. 2013; Li et al. 2018). Fe3+/4+ redox was confirmed with 57Fe Mössbauer spectroscopy (Takeda et al. 1994; Zhao et al. 2013; Lee et al. 2015) and oxidation of oxide ions was recently revealed with soft XAS at O K-edge (Li et al. 2018). The mean working voltage of 3.3 V is the highest among single 3d transition metal O3 and O’3 systems. When Na+ ions are extracted by >0.5 mole from NaxFeO2 upon a charging process, the discharge capacity significantly deteriorates due to an irreversible structural change accompanied by Fe migration into the interslab space (Yabuuchi et al. 2012b) as seen in NaTiO2, NaVO2 and NaCrO2. Detailed phase transitions during charge/discharge were studied with ex situ 57Fe Mössbauer spectroscopy and operando synchrotron XRD by Lee et al. and the results revealed non-equilibrium phase-transition behavior among original O3, secondary O3 and O’3 phases in which iron ions simultaneously migrate into interslab space even in x ≤ 0.5 in NaxFeO2 (Lee et al. 2015). Furthermore, reduction of Fe4+ accompanied by electrolyte decomposition was also observed during storage for 2 days after charging to 3.6 V. Suppression of the electrolyte decomposition and of the migration of iron ions is required to utilize Fe3+/4+ redox in layered oxides for Na-ion batteries.
1.3.1.5. O3-NaCoO2
Electrochemical properties of O3-NaCoO2 in Na cells were first reported by Braconnier et al. in 1980 (Braconnier et al. 1980) at almost the same time when those of O3-LiCoO2 in Li cells were first reported by (Mizushima et al. 1980). O3-NaCoO2 delivers a reversible capacity of ca. 140 mAh g−1 in the voltage range of 2.5–4.0 V (Figure 1.8) (Yoshida et al. 2013; Lei et al. 2014). As discussed above on O’3-NaMnO2, O3-NaCoO2 represents a stepwise voltage profile that is attributed to multiple phase transitions associated with CoO2-slab gliding and in-plane Na+/vacancy ordering in the interslab spacing as described in section 1.2.2.
1.3.1.6. O’3-NaNiO2
Electrochemical properties and phase transitions of O’3-NaNiO2 in Na cells were first reported by Braconnier et al. in 1982 (Braconnier et al. 1982) and re-investigated by Vassilaras et al. in 2013 (Vassilaras et al. 2013). O’3-NaNiO2 delivers a reversible capacity of ca. 100 mAh g−1 with stepwise voltage profile in the voltage range of 1.25–3.75 V (Figure 1.8). Detailed phase transitions during charge-discharge were reported by Han et al. (2014) and Wang et al. (2017a) with operando XRD measurements. The results revealed that O’3-NaNiO2 transforms into at least six phases of O’3 and P’3 during a charging process. Both the authors found that no original O’3 phase was observed on the discharging process and the irreversible phase transition was thought to cause capacity degradation during charge–discharge cycles. Unlike Ti, V, Cr, Mn and Fe, but like Co system, O’3-NaNiO2 delivers a large discharge capacity of ca. 130 mAh g−1 even after charging to the high voltage of 4.5 V versus Na (Vassilaras et al. 2013; Wang et al. 2017a). Wang et al. revealed from in situ synchrotron XRD that the Na-extracted phase of Na0.17NiO2 is irreversibly formed upon charging to 4.5 V and is partly remained in the core of the particles even after discharging to 2.0 V versus Na. Migration of nickel ions into interslab space was often observed for O3-LiNiO2 at the end of charging to 4.45 V versus Li (Croguennec et al. 2001). However, Li et al. mentioned no migration of nickel ions for O’3-NaNiO2 after charging to 4.5 V versus
Na from the unpublished HRTEM image in the literature (Li et al. 2016). The reaction mechanism of O’3-NaNiO2 might be different to that of O3-LiNiO2.
These fundamental studies are very helpful to understand the electrochemical properties of the complicated multiple transition metal systems. Transition metals usually dominate redox potential and phase transitions of O3-type layered materials. However, those of multiple 3d transition metal systems are not simple and are influenced by difference in oxygen orbital contribution to the redox reaction (Nanba et al. 2016). Not only nominal valence of transition metals but also hybridization between oxygen 2p and transition metal 3d orbitals plays a critical role in determining the redox potential of layered transition metal oxides in Na batteries. Redox potential and electrochemical activity of nickel ions vary on the other transition metal elements in NaMO2 (Nanba et al. 2016). Selection of the transition metals is, therefore, important to synthesize ternary and quaternary 3d transition metal systems exhibiting excellent
