Controlling Defects to Achieve Reproducibly High Ionic Conductivity in Na3SbS4 Solid Electrolyte

Abstract

The ability to reproducibly synthesize highly conductive solid electrolytes (SEs) is a prerequisite for the widespread usage of solid-state batteries. However, reported ionic conductivities of SEs exhibit significant variation even in materials with same nominal composition. In this study, the thermodynamic origin of such sample-dependent variations are discussed using sodium-ion conducting Na3SbS4 as a model SE. The impact of uncontrolled variations in elemental chemical potentials on the ionic conductivity is investigated with theory and experiments. The elemental chemical potentials are uniquely defined when the system is constrained to have zero thermodynamic degrees of freedom. First, we establish the relationship between the chemical potentials and sodium-ion conductivity in Na3SbS4 by computing the phase diagram and native defect formation energies. From these calculations, we identify two distinct three-phase equilibrium regions (zero degrees of freedom) with the highest ratio of sodium-ion conductivity, which are then experimentally probed. Transport measurements reveal an abrupt change in the bulk ion transport of the phase-pure samples, with room-temperature ionic conductivity of 0.16 − 1.2 mS cm−1 with a standard deviation of 50% when the elemental chemical potentials are not controlled i.e., uniquely defined. In contrast, we show that by controlling the chemical potentials and therefore, the defect formation energies through the experimental concept of phase boundary mapping, the sample-dependent variation is reduced to 15% with a high average ionic conductivity of 0.94 mS cm−1. This study highlights the existence of “hidden” thermodynamic states defined by their chemical potentials and the need to precisely control these states to achieve reproducibly high ionic conductivity.