Fast ionic conductivity is a key requirement for solid-state electrolyte (SSE) in solid-state batteries. There has been growing interest in halide-based SSEs in the A_yMX₆ (A = Li or Na, M = transition metal or rare earth, X=Cl, Br or I) family, for both Li and Na solid-state batteries. Since the initial study by Asano et al., [1] in which fast ionic conductivity was demonstrated in Li₃YCl₆ and Li₃YBr₆ materials, numerous studies have focused on halides materials with hexagonal (HCP) or cubic (CCP) closed packed anion lattices, in which the metal, M, cation is replaced with other transition metal or rare earth elements. [2] Chloride based systems have been shown to give the highest stability against high voltage cathodes, but typically demonstrate lower ionic conductivities than bromide or iodide systems. Although numerous chloride systems have been studied, achieving ionic conductivities > 3 mS/cm at room temperature appears to be a significant challenge. In contrast, Ag and Cu conducting halides based on the RbAg₄I₅ structure such as RbCu₄Cl₃I₂ have long been known to exhibit exceptional conductivities > 300 mS/cm at room temperature, falling into the class of ‘advanced superionic conductors’.[3,4] An understanding of how rapid ion transport originates in halide materials is fundamentally lacking, which is holding back the design and optimization of future chloride-based advanced superionic conductors.

In this work, we explore the role that A-site cations play on the ionic conductivity of four isostructural chloride SSE systems, A₂ZrCl₆ where A=Li, Na, Cu and Ag. All four structures adopt the HCP structure when synthesised mechanochemically, allowing for a direct comparison of the impact of the A-site cation. Cu₂ZrCl₆ and Ag₂ZrCl₆ display exceptional ionic conductivities (> 3 mS/cm) that are 1–3 orders of magnitude higher than their Li and Na analogues. Using state-of-the-art DFT calculations coupled with transition state searching (TSS), a diverse range of previously unknown ionic conduction mechanisms in the A₂ZrCl₆ systems are captured.

We demonstrate a new universal transition state model using simple binary halide structures (LiCl, NaCl, CuCl and AgCl) to show that the activation barrier is dominated by the connectivity and relative energies of different octahedral, tetrahedral and trigonal planar coordination at various lattice sizes. We show that an intrinsic ‘lower bound’ activation barrier is always present when there is more than one coordination change along a pathway. We demonstrate that conductivity in the Cu₂ZrCl₆ system approaches the optimum criterion for the HCP A_yM_zX₆ family of materials. This model is universal to all ionic conductors, including oxide and sulfide SSEs. The results strongly suggest that new halide structural families are required to obtain advanced superionic chloride conductors, and we provide guidance about how these materials may be discovered.