Ion transport pathways in molecular dynamics simulated lithium silicate glasses

The xLi2O–(1 − x)SiO2 (x = 0.1, 0.15, 0.2, 0.25, 0.3, 0.33, 0.4, 0.45, and 0.5) glass systems have been studied by constant volume molecular dynamics (MD) simulations, using empirical pairwise Morse-type potentials. Bond valence (BV) method is applied to the equilibrated configurations to analyse the structural variation in these glass systems with increasing network modifier content, its consequence for Li+ ion mobility, as well as the distribution of bridging and non-bridging oxygen atoms and the variation of the Qi values. The contribution of non-bridging oxygen atoms to the Lithium BV sums exhibits a transition around x = 1/3. The observed Qi variation can be explained by a bond order model. Despite slight deviations of the interatomic distances in the MD-simulated glasses, their BV analysis reveals features of the ion transport pathway (such as volume fraction and local dimensionality of the percolating pathway). Results are compared to pathway models for related glassy solid electrolytes based on reverse Monte Carlo modelling of diffraction data. For complex disordered systems with low ion mobilities the bond valence analysis of the pathway characteristics for the mobile ion is thus a viable method to extract ion transport properties even if the mobilities are too low to be directly analysed from the mean square displacements over the simulated period.