Perovskite Na-ion conductors developed from analogous Li3xLa2/3−xTiO3 (LLTO): chemo-mechanical and defect engineering

Abstract

Na-ion conducting solid electrolytes can enable both the enhanced safety profile of all-solid-state-batteries and the transition to an earth-abundant charge-carrier for large-scale stationary storage. In this work, we developed new perovskite-structured Na-ion conductors from the analogous fast Li-ion conducting Li_3xLa_2/3−xTiO₃ (LLTO), testing strategies of chemo-mechanical and defect engineering. Na_xLa_2/3−1/3xZrO₃ (NLZ) and Na_xLa_1/3−1/3xBa_0.5ZrO₃ (NLBZ) were prepared using a modified Pechini method with varying initial stoichiometries and sintering temperatures. With the substitution of larger framework cations Zr⁴⁺ and Ba²⁺ on B- and A-sites respectively, NLZ and NLBZ both had larger lattice parameters compared to LLTO, in order to accommodate and potentially enhance the transport of larger Na ions. Additionally, we sought to introduce Na vacancies through (a) sub-stoichiometric Na : La ratios, (b) Na loss during sintering, and (c) donor doping with Nb. AC impedance spectroscopy and DC polarization experiments were performed on both Na_0.5La_0.5ZrO₃ and Na_0.25La_0.25Ba_0.5ZrO₃ in controlled gas environments (variable oxygen partial pressure, humidity) at elevated temperatures to quantify the contributions of various possible charge carriers (sodium ions, holes, electrons, oxygen ions, protons). Our results showed that the lattice-enlarged NLZ and NLBZ exhibited ∼19× (conventional sintering)/49× (spark plasma sintering) and ∼7× higher Na-ion conductivities, respectively, compared to unexpanded Na_0.42La_0.525TiO₃. Moreover, the Na-ion conductivity of Na_0.5La_0.5ZrO₃ is comparable with that of NaNbO₃, despite having half the carrier concentration. Additionally, more than 96% of the total conductivity in dry conditions was contributed by sodium ions for both compositions, with negligible electronic conductivity and little oxygen ion conductivity. We also identified factors that limited Na-ion transport: NLZ and NLBZ were both challenging to densify using conventional sintering without the loss of Na because of its volatility. With spark plasma sintering, higher density can be achieved. In addition, the NLZ perovskite phase appeared unable to accommodate significant Na deficiency, whereas NLBZ allowed some. Density functional theory calculations supported a thermodynamic limitation to creation of Na-deficient NLZ in favor of a pyrochlore-type phase. Humid environments generated different behavior: in Na_0.25La_0.25Ba_0.5ZrO₃, incorporated protons raised total conductivity, whereas in Na_0.5La_0.5ZrO₃, they lowered total conductivity. Ultimately, this systematic approach revealed both effective approaches and limitations to achieving super-ionic Na-ion conductivity, which may eventually be overcome through alternative processing routes.