Revealing the Ion Dynamics in Li₁₀GeP₂S₁₂ by Quasi-Elastic Neutron Scattering Measurements

Abstract

Understanding Li-ion conduction in superionic conductors accelerates the development of new solid electrolytes to enhance the charge− discharge performances of all-solid-state batteries. We performed a quasi-elastic neutron scattering study on a model superionic conductor (Li 10+x Ge 1+x P 2−x S 12 , LGPS), to reveal its ion dynamics on an angstrom-scale spatial range and a pico-tonanosecond temporal range. The observation of spectra at 298 K confirmed the high lithium diffusivity. The obtained diffusion coefficient was in the order of 10 −6 cm 2 s −1 at temperatures >338 K and was higher than the reported diffusion coefficient over a longer time scale, as determined by the pulse-field gradient nuclear magnetic resonance method. This difference indicates that there are impediments to ionic motion over a longer time scale. The dynamic behavior of the Li ions was compared with that observed for the Li 9 P 3 S 9 O 3 phase, which possesses the same crystal structure type, but a lower ionic conductivity. The LGPS phase possessed a high lithium mobility over a distance of ∼10 Å, as well as a larger fraction of mobile Li ions, thereby indicating that these features enhance lithium conduction over a longer spatial scale, which is important in all-solidstate batteries.

Deciphering Fast Ion Transport in Glasses: A Case Study of Sodium and Silver Vitreous Sulfides

“…The derived effective sodium diffusion coefficient D Na FPMD (eff) where the time domain is restricted by the last 10 ps, appears to be overestimated in comparison with long-range sodium migration, Figure c. This is not surprising because cation mobility is faster on a shorter spatial and time scale, see for example, and references therein. It should also be noted that the observed difference, D Na FPMD (eff)/ D σ = 7 ± 2, might be lower, while D σ corresponds to a solid glass, and D Na FPMD (eff) to a supercooled liquid.…”

Accurate description of ion migration in solid-state ion conductors from machine-learning molecular dynamics

“…D + Li values of 5.33 × 10 −6 cm 2 s −1 and 4.50 × 10 −6 cm 2 s −1 from AIMD and MLMD simulations, respectively, were extracted, which are in line with previous computational studies 99 and quasi-elastic neutron scattering measurements. 100 We assign the observed deviations to the increased stochastic character of Li migration in LGPS. Thus, large simulation timesexceeding feasible AIMD timescalesare required to predict reliable MSD values.…”

Investigating the Li+substructure and ionic transport in Li10GeP2−xSbxS12(0 ≤x≤ 0.25)

“…Additionally, Li(2) and Li(4) bridge the Li(1)–Li(3) tunnel in the ab -plane offering the possibility of three-dimensional diffusion in which the Li(1)–Li(4) pathway seems to be preferred over the Li(2)–Li(3) pathway. 20–23…”

“…13,19 Additionally, Li (2) and Li(4) bridge the Li(1)-Li(3) tunnel in the ab-plane offering the possibility of three-dimensional diffusion in which the Li(1)-Li(4) pathway seems to be preferred over the Li(2)-Li(3) pathway. [20][21][22][23] To date, several substitutions were performed within the LGPS family to study structural influences on conductivity, derive design principles, and further improve the conductivity. Higher conducting compositions of the Li-Sn-Si-P-S system showed three-dimensional diffusion pathways, while the lower conducting samples exhibited one-dimensional diffusion pathways indicated by maximum entropy method.…”