Achieving Desired Lithium Concentration in Garnet Solid Electrolytes; Processing Impacts on Physical and Electrochemical Properties

Lithium-ion-conducting garnets are one of the most promising solid-state electrolytes to replace liquid organic electrolytes in lithium batteries, resulting in a tremendous and rapidly increasing number of garnet publications and commercial activity. However, these publications are typically based on a nominal lithium content with the addition of excess lithium to qualitatively supplement lithium losses during calcining and sintering, resulting in an inexact lithium composition that directly influences the sintered garnet’s physical and electrochemical properties. In this study, we systematically varied the lithium content during the processing of Li6.75±XLa2.75Ca0.25Zr1.75Nb0.5O12 (LLCZN), then measured the actual lithium content, and determined density, secondary phase formation, and the impact on electrochemical properties. The molar lithium content after sintering ranged from 5.5 to 7.3 lithium per mol for a desired 6.75 stoichiometric LLCZN Li-composition, depending on the starting lithium content and sintering conditions utilized. Secondary phases were detected when lithium content was outside of the designed stoichiometry, hindering sintering. Samples sintered in MgO crucibles obtained high bulk conductivities (>10–4 S cm–1). However, they exhibited low grain boundary conductance due to secondary phases, resulting in total conductivities on the order of 10–5 S cm–1. In comparison, when sintered in Al2O3 crucibles, bulk conductivity (2.0 × 10–4 S cm–1) and grain boundary conductance (1.07 × 10–3 S cm–1) reached their maximum at 6.37 lithium per mol of LLCZN for a total conductivity of 1.68 × 10–4 S cm–1. While similar bulk conductivities were attained when sintered in MgO crucibles, the highest total conductivities were achieved in samples contaminated with parts per million levels of aluminum from the Al2O3 crucible.