Ion diffusion is important in a variety of applications, yet fundamental understanding of the diffusive process in solids is still missing, especially considering the interaction of lattice vibrations (phonons) and the mobile species. In this work, we introduce two formalisms that determine the individual contributions of normal modes of vibration (phonons) to the diffusion of ions through a solid, based on (i) Nudged Elastic Band (NEB) calculations and (ii) molecular dynamics (MD) simulations. The results for a model ion conductor of $rm{Ge}$-substituted $rm{Li_3PO_4}$ ($rm{Li_{3.042}Ge_{0.042}P_{0.958}O_4}$) revealed that more than 87% of the $rm{Li^+}$ ion diffusion in the lattice originated from a subset of less than 10% of the vibrational modes with frequencies between 8 and 20 THz. By deliberately exciting a small targeted subset of these contributing modes (less than 1%) to a higher temperature and still keeping the lattice at low temperature, we observed an increase in diffusivity by several orders of magnitude, consistent with what would be observed if the entire material (i.e., all modes) were excited to the same high temperature. This observation suggests that an entire material need not be heated to elevated temperatures to increase diffusivity, but instead only the modes that contribute to diffusion, or more generally a reaction/transition pathway, need to be excited to elevated temperatures. This new understanding identifies new avenues for increasing diffusivity by engineering the vibrations in a material, and/or increasing diffusivity by external stimuli/excitation of phonons (e.g., via photons or other interactions) without necessarily changing the compound chemistry.
Download