(Invited) High-Throughput Screening of Solid-State Li-Ion Conductors Using Lattice-Dynamics Descriptors

Abstract

The average vibrational frequencies of Lithium sublattice or Li-phonon band centers have been correlated with the migration barriers of Li-ion in solid-state Li-ion conductors. In this talk, we present a computational screening of ~14,000 Li-containing compounds in the Materials Project database using this lattice-dynamic descriptor to identify new promising Li-ion conductors. After pre-screening the database for structural stability, electronic conductivity and compositions, we arrived at ~1200 compounds whose phonon band centers were computed using an optimized computational approach called Quickhesswhich requires only 2 DFT calculations per compounds. In addition to newly predicted Li-ion conductors, we also discussed broad correlations between the oxidation potential as well as anion p-band centers and the anion-phonon band centers, whose origin can be understood in term of covalency/ionicity of the cation-anion bonds in the structure. Finally, combining a low computed Li phonon-band center with the computed electrochemical stability window and the structural stability based on their energy above hull, 18 compounds were predicted to be promising Li-ion conductors. One of these newly predicted compounds, Li3ErCl6 was synthesized and was shown to exhibit a measured room-temperature conductivity of 0.3 mS/cm for the ball-milled amorphous sample and 0.05 mS/cm in the crystalline state, which further shows the promise of high-throughput search of Li-ion conductors based on lattice dynamics descriptors.