Speaker
Description
Polarons play a central role in mediating Li-ion diffusion
in the Li4Ti5O12 (LTO) anode. The presence of polarons reduces the energy barrier for Li-ion diffusion and influences both electronic and ionic mobility. However, mobility estimation remains a significant challenge for purely ab initio methods, since the relevant timescales exceed by orders of magnitude those accessible via ab initio molecular dynamics. Machine learning interatomic potentials (MLIPs) provide a promising alternative, combining the accuracy of first-principles methods with substantially reduced computational cost, thus enabling simulations on the nanosecond timescale. In this work, we employ the recently developed MLIP architecture LEOPOLD (Learning of Polaron Dynamics) to estimate the mobility of a single polaron in bulk LTO. This represents a crucial first step toward a broader understanding of polaron dynamics and their impact on Li-ion transport in LTO for battery applications.