Lithium superionic conductors with corner-sharing frameworks

Abstract

Superionic lithium conductivity has only been discovered in a few classes of materials, mostly found in thiophosphates and rarely in oxides. Herein, we reveal that corner-sharing connectivity of the oxide crystal structure framework promotes superionic conductivity which we rationalize from their distorted lithium environment and reduced interaction between lithium and non-Li cations. By performing a high-throughput search for materials with this feature, we discover 10 novel oxide frameworks predicted to exhibit superionic conductivity-from which we experimentally demonstrate LiGa(SeO3)2 with a bulk ionic conductivity of 0.11 mS/cm and activation energy of 0.17 eV. Our findings provide new insight into the factors that govern fast lithium mobility in oxide materials and will accelerate the development of novel oxide electrolytes for all solid-state batteries.

Unconventional Ferroelectricity with Quantized Polarizations in Ionic Conductors: High-Throughput Screening

“…Herein, we propose a new type of ferroelectricity that may exist in ionic conductors even with a centrosymmetrical lattice, which may possess giant quantized polarizations. In ionic conductors − (e.g., lithium ion battery materials − ), akin to ferroelectric switching, ions can also be displaced under an electric field but for much longer distances (trans-unit cell displacement). Because ionic bondings are nondirectional and nonsaturating, usually the coordination around the cations (like alkali and alkali earth metal ions) in most ionic conductors is octahedral, yielding highly symmetrical structures.…”

High‐Throughput Data‐Driven Prediction of Stable High‐Performance Na‐Ion Sulfide Solid Electrolytes

“…To the best of our knowledge, the largest DFT-MD n data for SEs was given by Sendek et al ( n data = 19), [36] while ours (n data = 170) exceeds the existing SE-related studies. [36][37][38][39][40][41][42][43][44] Among n data = 170, we chose n data = 44 samples with various magnitudes of D acc,Na, 300K and σ acc,Na, 300K (including extrema) from the different M-M′ combinations to achieve a representative sample distribution. (Hereafter, D acc,Na, 300K and σ acc,Na, 300K denote the "accelerated"-DFT-MD D Na,300K and σ Na,300K given in the single-temperature "short-time" diagnosis, respectively; see Table S1 1).…”

“…It is noted that the electrostatic, the free space, the geometrical, and the lattice-dynamics descriptors were suggested in the literature for σ Na,T (σ Li,T ) and/or E a (see Table S4, Supporting Information). [36][37][38][39][40][41][42][43][44][51][52][53][54][55][56][57] For the statistical validation of the "important" descriptors, we leveraged on our D Na,300K -values accumulated from the multi-stage DFT-MD sampling workflow. Our descriptors are readily accessible from any given cell structures.…”

Degradation of a Li1.5Al0.5Ge1.5(PO4)3-Based Solid-State Li-Metal Battery: Corrosion of Li1.5Al0.5Ge1.5(PO4)3 against the Li-Metal Anode

“…However, the highly covalent nature of the PO 4 tetrahedron makes it hard to tolerate distortion of the local structure. The extra interstitial Li ion results in the distortion of the local structure. , Furthermore, the additional interstitial Li ion also leads to the weakening of the P–O bond . Therefore, the PO 4 sublattice is destructed by local distortion resulting from the extra Li-ion injection.…”