Modelling Messy Materials—Amorphous and Polycrystalline Solid Electrolytes

Using case study examples, we will demonstrate how computational techniques can be used to model messy solid electrolyte materials which are rich in defects and disorder. These results are connected to experiment to gain an understanding of ionic diffusion mechanisms, thus informing the material design.<br/><br/>Disordered systems pose unique challenges for computational modelling due to the size of the simulation cells required. This is especially vexing, given that solid electrolytes are usually highly defective and disordered materials. This may be by design, by introducing disorder as a means to enhance ionic diffusivity. Alternatively, it may be a side effect of synthesis; for example, the pressing of powders will produce a polycrystalline sample with many grain boundaries which can impede diffusion.<br/><br/>Sulfide solid electrolytes are a particularly promising technology. We show that an amorphous thiosilicate with composition 5Li₂S-3SiS₂ shows excellent ionic conductivity. By simulating NMR, we confirm the structure of the polyanions present in the material, allowing us to determine that the geometry of the polyanion is the origin of the low barriers to diffusion. [X. Hao, J. A. Quirk et. al., <i>Adv. Energy Mater. </i>2024, 2304556.] Beyond this, we have produced models of polycrystalline and glassy Li₆PS₅Cl. These models agree well with experimental observations and show encouraging tolerance towards defects.