Non-Arrhenius Ionic Conductivity Transitions in Sodium Antiperovskite Ionic Conductors

Abstract

Anti-perovskites materials have recently attracted great interest as a family of solid state electrolytes with high ionic conduction.[1] Typical solid-state ionic conductors have temperature-dependent ionic conductivity following Arrhenius behavior over a substantial range of temperature, corresponding to a constant activation energy Ea. In the Na3OCl antiperovskite (Fig. 1), there are regimes of Arrhenius behavior separated by broad transitions, all while the structure remains cubic, Figure 2. Such non-Arrhenius ionic conductivity of Na3OCl has been observed in previous literature. [2][3] However, there is no clear explanation for this non-Arrhenius transition. Differential scanning calorimetry shows thermal transitions in the vicinity of these conductivity transitions, Figure 3, suggesting order/disorder phase transitions. Reitveld refinement on temperature dependent synchrotron XRD of Na3OCl suggests changes in octahedral tilt systems, which is CmCm in the low temperature region, Pnma in the medium temperature region, and P4/mbm in the high temperature geion, Figure 4. In this work, we use multiple temperature-resolved characterization methods including synchrotron XRD, transport measurements, and calorimetry, and modeling by AIMD and DFT, to understand this behavior and to gain insight into tuning antiperovskites for high ionic conductivity.This work was supported as part of the Joint Center for Energy Storage Research, an Energy Innovation Hub funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences. Figure 1

On the Origin of the Non‐Arrhenius Na‐ion Conductivity in Na3OBr

“…Previous studies on Li, Na and K antiperovskites also showed similar endothermic peaks (K 3 OI: 245 °C, Li 2 OHBr: 254 °C, Na 3 OBr: 255 °C, and Na 3 OCl: 191 °C). Some studies attributed these peaks to melting of the antiperovskite phases and others to anion disordering [14,16,17,19] . Except for this peak, no other peaks were observed in the DSC plot during heating to 300 °C.…”

“…Some studies attributed these peaks to melting of the antiperovskite phases and others to anion disordering. [ 14 , 16 , 17 , 19 ] Except for this peak, no other peaks were observed in the DSC plot during heating to 300 °C. Additionally, the bulk material remains solid up to 300 °C, which is confirmed by variable temperature synchrotron XRD, Figure 2b , meaning that the peak does not correspond to the melting of Na 3 OBr.…”

“…for Na 3 O(BH 4 ); however, attempts to reproduce their results have been unsuccessful [10,11,15] . Interestingly, non‐Arrhenius behaviour resulting in a decrease in activation energy for ion conduction above 250 °C has been observed across different antiperovskite compositions (Li, K and Na) [16–23] . The origin of this behaviour is not yet well understood.…”

“…[ 10 , 11 , 15 ] Interestingly, non‐Arrhenius behaviour resulting in a decrease in activation energy for ion conduction above 250 °C has been observed across different antiperovskite compositions (Li, K and Na). [ 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 ] The origin of this behaviour is not yet well understood. For antiperovskite compositions with cluster anions like Na 3 O(NO 2 ) and Na 2 (NH 2 )(BH 4 ), deviation from Arrhenius behaviour is often assigned to activation of the cluster anion rotations and correlated motion between these rotations and the translating cation.…”

On the Origin of the Non‐Arrhenius Na‐ion Conductivity in Na3OBr

“…Some studies attributed these peaks to melting of the antiperovskite phases and others to anion disordering. [ 14 , 16 , 17 , 19 ] Except for this peak, no other peaks were observed in the DSC plot during heating to 300 °C. Additionally, the bulk material remains solid up to 300 °C, which is confirmed by variable temperature synchrotron XRD, Figure 2b , meaning that the peak does not correspond to the melting of Na 3 OBr.…”