hide
Free keywords:
-
Abstract:
Lithium ion mobility remains one of the crucial parameters for predicting the performance and the life-time of a lithium-ion battery (LIB). In this regard, having a reliable diffusion coefficient associated with the respective lithium diffusion process is necessary and mandatory for understanding the correlation of this property with the failure or with the state of health (SOH) of a LIB over time. However, the scatter of ten orders of magnitude of those atomistic processes within the measured diffusion coefficients raises the question what the main reason behind this uncertainty is. In this thesis, a reference system for the lithium ion mobility of lithium intercalated in Highly Pyrolityc Ordered Graphite (HOPG) is investigated from a theoretical and experimental point of view. The focus is on the upper boundary of the lithium intercalation such as LiC6 which is the most studied phase and can be considered as the highest state of charge (SOC) of a LIB. Therefore, it can be considered as perfectly suitable to validate a new method and/or approach. The right choice and combination of methods is not trivial. The time and length scales (easily) accessible by theory and experiments are typically different. To match those one need to perform more elaborate experiments as well as theory. Nuclear magnetic resonance (NMR) is known to be able to give access to dynamical properties of the lithium ion, within the solid state diffusion framework. In order to theoretically simulate the ion dynamics at the same time and length scales as NMR a kinetic Monte Carlo (kMC) approach is mandatory in terms of computational cost. In classical kMC, the most displaced particle in a process is often treated as an isolated neutral atom. The electrochemistry of LiBs though motivates the decision to use an implementation of kMC, with an explicit charge treatment framework in combination with input parameters obtained from the first-principles. A successful combination of the first principle charged kinetic Monte Carlo (1p-ckMC) with an advanced analysis through an inverse Laplace transform (ILT) of 7Li Spin-Alignment-Echo-Nuclear magnetic resonance (SAE-NMR) to assess lithium ion mobility is then presented. The novelty of the approach lies within the combination of those techniques that allow to assess ion mobility independently and the possibility to further use the power of the simulation to predict mobilities and/or to unravel the mechanism behind it. At the same time, while preparing the reference system using blind conditions (ambient pressure), the formation of superdense patterns was observed. Those were then further investigated through ageing and rationalized with ab initio thermodynamics (AITD) and molecular dynamics (MD). The presence of the superdense structures at those conditions reopens then the discussion on the actual highest SOC (LiC6)
