Ion migration mechanisms in glassy solid electrolytes at low temperatures
Professor Donald Siegel, University of Michigan, USA
Sulphur-based glasses are promising candidates for use as solid electrolytes in Li-based batteries. Nevertheless, due to their amorphous structure, the atomic-scale mechanisms that underlie Li-ion conductivity in these systems are challenging to characterize. The present study employs ab initio molecular dynamics to predict the local structure and migration processes in the prototype Li-ion conducting glass, 75Li2S–25P2S5. A model of the amorphous structure was generated and shown to closely match the measured neutron pair distribution function. Lithium migration is observed to occur via a complex mechanism that combines concerted motion of lithium ions with large, quasi-permanent rotational displacements of the PS43- tetrahedra. This latter effect, commonly referred to as the ‘paddlewheel’ mechanism, is most commonly observed in lower-density crystalline phases that are stable only at elevated-temperatures. Unlike these crystalline analogues, in the glass, the present calculations indicate that paddlewheel dynamics contribute to Li-ion mobility at temperatures as low as 300 K. Paddlewheel contributions are confirmed through analyses of spatial, temporal, vibrational, and energetic correlations with Li motion. Furthermore, the dynamics in the glass are shown to differ from those in the stable crystalline phase (-Li3PS4), where contributions from anion reorientations are negligible and the conductivity is much smaller. These data imply that glasses based on complex anions, and in which covalent network formation is minimized, have the potential to exhibit paddlewheel dynamics at low temperature. Glasses that satisfy these requirements may be fertile ground in the search for new solid electrolytes.
Interplay of site-disorder, interplay and ionic conductivity of superionic conductors: insights from atomistic computer simulations
Professor Karsten Albe, Technische Universität Darmstadt, Germany
Glassy, glass–ceramic, and crystalline lithium thiophosphates have attracted interest in their use as solid electrolytes in all-solid-state batteries. Despite similar structural motifs, including PS43–, P2S64–, and P2S74– polyhedra, these materials exhibit a wide range of possible compositions, crystal and amorphous structures, as well as ionic conductivities. Calculations based on density functional theory can be a helpful tool for understanding diffusion pathways and Li+ ionic conductivity and interface stabilities.
This contribution will include a discussion of recent results on the defect chemistry and conductivity of the solid electrolyte Li4P2S6 as well as its interfacial instability with respect to Li. Then, molecular dynamics simulations of crystalline and amorphous Li4PS4I, will be shown, which unravel the diffusion mechanism and can be explained by a rate-equation model based on superbasins. Finally, results on the Lithium argyrodites of the type Li6PS5X (X = Cl, Br, I) are presented, where the influence of S2-/Br- site-disorder was studied. The simulations reveal that local “Li cages” trap Li ions in the ordered material. At higher degrees of site-disorder the cage structures dissolve and long-range low energy pathways are established. The analysis of pair distribution functions (PDF) and Li-density maps elucidates the correlation between structural disorder and ionic conductivity.
Paradigms of structural, chemical, and dynamical frustration in superionic conductors
Dr Brandon Wood, Lawrence Livermore National Laboratory, USA
Rationally motivated computational discovery and optimization of solid electrolytes require the development of reliable descriptors for fast solid-state ionic conductivity. However, many of the fundamental motivations for superionic behaviour in solids remain enigmatic, which has generally slowed progress in screening new candidates or tuning existing materials to maximize ionic conductivity. Dr Wood will discuss the use of high-performance computer simulations and advanced analytical techniques to unravel various mechanisms of ionic conductivity in model classes of solid electrolytes. Using computational “experiments”, the simulations systematically isolate factors such as stoichiometry, strain, composition, crystal structure, and local environment in the determination of ionic conductivity. Collectively, the results point to the importance of a frustrated energy landscape in promoting ultrafast diffusion. Different types of frustration in model superionic conductors will be discussed, arising from factors such as off-stoichiometry, competition between interstitial site occupancies, symmetry incompatibilities between local bonding character and lattice geometry, and dynamical frustration coupled to anharmonic lattice motion. Dr Wood will explore the physicochemical relevance of these factors for understanding and promoting cation mobility, with a view towards developing design rules for engineering faster ionic conductors. Among the topics to be discussed is the dependence of the different frustration paradigms on the fundamental nature of the lattice-forming ions, which suggests there may be no single universal descriptor for ionic conductivity, but rather classes of superionic conductors with similar underlying motivations. Specific examples will be drawn from recent results on superionic materials based on oxides, halides, and polyatomic anions.