Unlocking Li superionic conductivity in face-centred cubic oxides via face-sharing configurations

Abstract

Oxides with a face-centred cubic (fcc) anion sublattice are generally not considered as solid-state electrolytes as the structural framework is thought to be unfavourable for lithium (Li) superionic conduction. Here we demonstrate Li superionic conductivity in fcc-type oxides in which face-sharing Li configurations have been created through cation over-stoichiometry in rocksalt-type lattices via excess Li. We find that the face-sharing Li configurations create a novel spinel with unconventional stoichiometry and raise the energy of Li, thereby promoting fast Li-ion conduction. The over-stoichiometric Li–In–Sn–O compound exhibits a total Li superionic conductivity of 3.38 × 10−4 S cm−1 at room temperature with a low migration barrier of 255 meV. Our work unlocks the potential of designing Li superionic conductors in a prototypical structural framework with vast chemical flexibility, providing fertile ground for discovering new solid-state electrolytes.

Design principles for anode stable solid-state electrolytes

“…The essential difference between Li-DRX and Li-PRX with the same metal species is the Li content per formula unit. The prototype formula for Li in Li-DRX electrolytes, 5 Li 1+ x + δ (M 1 M 2 ) 1− x O 2 , contrasts with that for Li-PRX 1,32 electrolytes, Li x (A,Vac) 1− x (M 1 M 2 ) 1 O 3 , indicating a significant variation in Li fraction per formula, regardless of chemical formulae. Typically, Li-PRX will have maximum one Li per five sites or per three anions, 33 while Li-DRX will have at least one Li per two anions.…”

“…The systematic exploration of the metal-dependent reduction limits involves a comprehensive examination of three promising categories of metal-containing electrolytes: perovskite-based metal oxides (Li-PRX or Na-PRX), 1,2,30,31 disordered rocksalt-type metal oxides (Li-DRX or Na-DRX), 5–7 and metal halides (Li-MH or Na-MH). 8–13 To ensure our screening encapsulates all typical stoichiometries and compositions capable of maintaining charge balance within these three structures, we employed grid enumeration, as detailed in the ESI Section S1 †.…”

“…The advancement of solid-state electrolytes has been significantly propelled by the computational and experimental identification of novel superionic conductors. Substantial efforts have been dedicated to the discovery of metal contained solid-state electrolytes, particularly metal oxides 1–7 and metal halides 8–13 as both of them show rich chemical space and structural form. It is of no doubt that both metal oxide and metal halide based solid-state electrolytes can be designed to show very competitive ionic conductivity, even approaching the conductivity of sulfide electrolytes 14–16 and liquid electrolytes.…”

“…Furthermore, we propose a set of generalized compositional and structural design principles aimed at developing stable electrolytes at the anode. Specifically, through a high-throughput screening of 16 644 compositions across three typical structures 1,2,5–13 – 3846 compositions for Li(Na) perovskite-based metal oxides, 1238 compositions for Li(Na) disordered rocksalt-type metal oxides, and 11 560 compositions for Li(Na) metal halides – we have discovered that lithium-rich compounds, such as overlithiated disordered rocksalt types, generally exhibit significantly better anode stability than lithium-deficient compounds, such as Li perovskites or Li metal halides, even when the same metal species are involved. Additionally, we found that halides tend to be more unstable against anodes than oxides, primarily due to the increased likelihood of forming reduced metal halides.…”

Rapid Mining of Fast Ion Conductors via Subgraph Isomorphism Matching

“…Another shortcoming of this graph-based method is that it cannot contain nodes of partial occupancy of non-Li cations at Li sites. We note that the cation-disordered structures (e.g., solid-solution rocksalt compounds) − have recently gained increasing attention in the field of ISSEs. The disorder at lattice sites that can be assigned to either a Li diffusion tunnel or a non-Li framework precludes us from designating the corresponding subgraphs as subjects for isomorphism matching with another structure.…”