High powerlithium ion battery materialsby computational design

1 Introduction Understanding ionic motion in disordered solids requires insight into the correlation between ion mobility and the structural and energetic environment of mobile ions. Local structure models for disordered solid electrolytes such as ion conducting glasses may in principle be derived from diffraction data (via crystal structure refinements or in the case of glasses by reverse Monte Carlo (RMC) fits) or molecular dynamics (MD) simulations 1 , 2 . MD simulations yield comprehensive structural and dynamical information within the limitations imposed by the size of the simulated system, the time span covered by the simulation, and the agreement of the employed interaction potential parameters with reality. Both approaches proved to be valuable tools in obtaining insights into conduction mechanisms and their correlation to the atomic structure, though in the case of MD simulations it has to be verified that the force field chosen for the simulations leads to structure models that are consistent with experimental information 2 - 4 .

Here, we discuss how the bond valence (BV) method can be used to predict characteristics of ionic conductivity from structure models 4 - 7 , and be optimized by linking it to an absolute energy scale for a more straightforward comparison with other simulation tools.