Structure and lithium-ion dynamics in fluoride-doped cubic Li7La3Zr2O12 (LLZO) garnet for Li solid-state battery applications
journal contributionposted on 2018-12-05, 11:34 authored by Stephen Yeandel, Benjamin J. Chapman, Peter R. Slater, Pooja GoddardPooja Goddard
The lithium-stuffed garnet Li7La3Zr2O12 (LLZO), when suitably doped, is a promising candidate material for use as a solid-state electrolyte within advanced Li-ion batteries. It possesses the thermal and mechanical stability of many inorganic ceramics, while exhibiting high Li+ ionic conductivities often associated with conventional liquid electrolytes, making it an ideal component for large-scale energy storage. However, only the high-temperature cubic phase has any meaningful Li-ion conductivity. Typically the formation of this phase is achieved through cation doping (e.g., Al3+ on the Li site) to lower the Li content and so disrupt Li ordering. However, Li-site doping, in particular, may potentially lead to some disruption of the Li-ion conduction pathways and suboptimal ionic conductivities. Consequently, other novel doping strategies involving the anion site are gaining traction, for example, F– for O2– as an alternative strategy to lower the Li content without directly blocking the lithium-diffusion pathways. For the first time, classical potential-based simulations have been employed to simulate the incorporation of fluoride anions into LLZO. Low incorporation energies have been calculated, suggesting fluoride anions are stable on the oxygen sites with a compensating lithium-ion vacancy defect. Molecular dynamics calculations suggest a definitive phase transition to the more desirable cubic phase of LLZO when doped with fluoride at temperature significantly lower than that for the tetragonal–cubic phase transition found for pure LLZO. Remarkably, the lithium-ion transport properties are shown to improve in the fluoride-doped samples particularly at low temperatures due to the stabilization of the cubic phase, suggesting anion doping of garnet systems may be a compelling alternative route to optimize the ionic conductivity.
P.G. and S.R.Y. acknowledge the support of EPSRC SUPERGEN grant, EP/N001982/1. P.R.S. acknowledges the support of EPSRC grant EP/R024006/1: ICSF Wave 1: GENESIS: Garnet Electrolytes for New Energy Storage Integrated Solutions. This paper recognizes the use of the “Hydra” High-Performance System at Loughborough University. Via our membership of the U.K.’s HEC Materials Chemistry Consortium, which is funded by EPSRC (EP/L000202), this work used the ARCHER U.K. National Supercomputing Service (http://www.archer.ac.uk).