Even though Na_xMO₂ (x ≤ 1, M = transition metal) materials have been studied for a long time as positive electrode candidates, the available capacity is still limited. Partial substitution of M by Na⁺ is an appealing strategy to overcome this issue: the excess Na⁺ ions in the MO₂ layer could participate in the (de)intercalation process while lowering the weight of the battery, thus greatly enhancing the gravimetric capacity. This strategy, applied to Li-ion batteries for many years, is still poorly transferred to Na-ion battery materials.

In order to establish a model for the study of Na₂MO₃ – and A₂MO₃ in general –, we prepared two polymorphs of O3-Na₂RuO₃ distinguished by the honeycomb-like ordering of the [Ru_2/3Na_1/3]O₂ layers ^[1,2]. In both samples, we found that Na in the interlayer and in the [Ru_2/3Na_1/3]O₂ layers is accessible. In ordered-Na₂RuO₃, the honeycomb ordering triggers the spontaneous formation of an ilmenite-type intermediate which facilitates highly reversible oxygen redox chemistry, associated to a large extra capacity (Figure 1) as well as noticeable structural stability upon cycling making it, to our knowledge, the first A_xMO₃ to do so.

Figure 1. Galvanostatic cycling curves for (a) disordered and (b) ordered O3-Na₂RuO₃ (first cycle highlighted in blue). The insets represent the coordination environment of Na at x = 1.