Atomic Defect Mediated Li-Ion Diffusion in a Lithium Lanthanum Titanate Solid-State Electrolyte

Abstract

Lithium lanthanum titanium oxide (LLTO) as a fast Li-ion conductor is a promising candidate for future all-solid-state Li batteries. Fundamental understanding of the microstructure of LLTO and its effect on the Li+ diffusion mechanism, especially across different length scales and interfaces, is a prerequisite to improving the material design and processing development of oxide-based solid electrolytes. Herein, through detailed structural analysis of LLTO ceramic pellets by aberration-corrected transmission electron microscopy, we discovered previously unreported intrinsic planar defects in LLTO single-crystal grains. These planar defects feature an antiphase boundary along specific crystal planes with a “rock-salt” structure enriched by Li within a few atomic layers. Corroborated by density-functional-theory-based calculations, we show an increased diffusion barrier across these planar defects inevitably lowers the bulk Li+ diffusivity of the oxide electrolyte.

Probing the Phase Transition during the Formation of Lithium Lanthanum Zirconium Oxide Solid Electrolyte

“…However, these phase transitions often take time to reach thermodynamic equilibrium, where non-stoichiometric reaction and phases may happen and are vital to determining the total Li-ion conduction. Recent works on lithium lanthanum titanium oxide (LLTO) reveal the existence of a network of planar defects in the single crystalline LLTO grains, which were formed during the ceramic processing of LLTO pellets . This indicates that structural heterogeneity is critically dependent on the local chemical stoichiometry, and the phase transformation process is always accompanied by the change of the chemical environment.…”

“…Recent works on lithium lanthanum titanium oxide (LLTO) reveal the existence of a network of planar defects in the single crystalline LLTO grains, which were formed during the ceramic processing of LLTO pellets. 57 This indicates that structural heterogeneity is critically dependent on the local chemical stoichiometry, and the phase transformation process is always accompanied by the change of the chemical environment. Hence, we conduct a complete survey for different growth stages (the Ar-ion sputtering time is 30 min before each spectrum was obtained).…”

Atomic-scale study clarifying the role of space-charge layers in a Li-ion-conducting solid electrolyte

“…In recent years, although more and more powerful techniques are demonstrated as effective tools for the characterization of SCLs 4,18 , their atomic configurations remain unexplored. Therefore, despite the large number of insightful computational studies in recent years [19][20][21][22][23][24] , the absence of experimental verification still prevents the conclusive, precise comprehension of the critical interfaces involving SCLs.…”

“…Here, based on direct observation of the atomic configuration, we clarify the role of SCLs in Li 0.33 La 0.56 TiO 3 (LLTO), a prototype solid electrolyte plagued by the large grain-boundary resistance 12,[19][20][21]25 , through a combined experimental and computational investigation. The study discloses a scenario that is completely different from the previous understanding.…”

Lithium-Ion Diffusion in Perovskite-Type Solid Electrolyte Lithium Lanthanum Titanate Revealed by Pulsed-Field Gradient Nuclear Magnetic Resonance

“…Recent advances in density functional theory (DFT) and molecular dynamics (MD) allow simulation of Li + -ion diffusion in solid electrolytes. , DFT calculations can predict local diffusion pathways and activation barriers. − In MD simulations, the tracer diffusion coefficient ( D MD ) is calculated from the mean-squared displacement of Li + ions, − whereby ab initio MD calculations (AIMD) afford highly accurate diffusion coefficients. ,,, The simulated D MD is usually converted to ionic conductivity via the Nernst–Einstein equation, assuming the carrier number and the correlation factor. However, these assumptions make it difficult to compare results from simulations and experiments, and thus the best way to bridge the gap is to determine the diffusion coefficient experimentally.…”