Publication date: 10th April 2024
All solid-state batteries are a promising technology due to their employment of inorganic solid electrolytes (SEs), which are safer compared to their flammable, liquid counterparts. Recently, halide SEs have gained significant attention due to their relatively high ionic conductivity (> 1 mS/cm), wide electrochemical stability window, and compatibility with various electrode materials. However, prior to 2023, no known halide materials could reach the conductivities of the state-of-the-art sulfide SEs (> 10 mS/cm). This can be attributed to the close-packed anion arrangement of existing halide conductors, which confines lithium ions to narrow and high-energy conduction pathways. Recently, a new Li oxyhalide superionic conductor LiMOCl4 (M=Nb/Ta) was reported with extremely high ionic conductivities greater than 10 mS/cm [1]. Unlike previously found halide conductors, LiMOCl4 has a non-close-packed anion framework that provides a large conduction pathway. We present a first-principles investigation of the phase stability, electrochemical stability, and Li-ion conductivity in the LiMXCl4 (M = Ta, Nb, Sb, Ti, Zr, Hf, Sn, and X = O, F, Cl) family of superionic conductors. We identify a mixed-anion framework as key to stabilizing the oxyhalide structure and demonstrate the impact of aliovalent anion and cation substitutions on material properties. Long ab-initio molecular dynamics simulations spanning 10’s of ns confirm room temperature superionic conductivity in all systems. Probabilistic analysis of the rotational dynamics at the picosecond timescale contributes direct, quantitative evidence correlating the high Li-ion conductivity in oxyhalide systems with the soft tilting motion of MO2Cl4 octahedral groups present in the material. Our findings supply the theoretical underpinnings for the effectiveness of LiMXCl4 family of superionic conductors as catholytes compatible with 4-V-class layered cathodes and provide essential design criteria for constructing new Li halide superionic conductors with conductivity greater than 10 mS/cm.
This research was supported by Umicore, Contract Number 34586 to the Regents of the University of California, Berkeley. This material is based upon work supported by the U.S. Department of Energy, Office of Science, Office of Advanced Scientific Computing Research, Department of Energy Computational Science Graduate Fellowship under Award Number DE-SC0023112.