Views: 0 | Downloads: 0
Lithium metal batteries (LMBs) represent a highly promising technology for the future generation of energy storage, due to their very high theoretical energy density. However, their implementation is hindered by the growth of dendrites; needle-like structures, that pose a safety concern and degrade the performance and lifespan of the battery. This thesis covers the fundamentals of dendrite growth and explores new dendrite prevention strategies by merging experimental methods with computational modelling.
The main goal of this thesis was to study the dendrite growth process and to find potential mitigation approaches. This was attempted by combining magnetic resonance imaging (MRI) and diffusion-limited aggregation algorithm (DLA) based computational modelling. MRI is a technique for non-invasive operando imaging of the electrodeposition process, while the DLA-based model is suitable for capturing the random nature of the dendrite growth process.
MRI provided a way to noninvasively monitor the electrodeposition/dendrite growth in symmetric lithium-electrolyte-lithium cells. The acquired images showed complex structures with different morphologies, depending on the operating conditions of the cell. Comparative experiments on simple, parallel and series electrode topologies showed that parallel promote runaway filament growth, whereas series configuration imposes uniform planar deposition. Additionally, electrolyte-depletion fronts and the accumulation of passivated “dead” lithium was monitored. MRI enabled not only the imaging of dendrites (via signal voids), but also gas formation and formation of “dead lithium”, which, when combined, provides a comprehensive view of the nucleation and growth process of lithium dendrites.
To complement MRI observations, a field-biased 3D diffusion-limited aggregation (DLA) algorithm, coupled with a Laplace solver for the electric potential, reproduced deposition patterns and showed that ohmic drop is the primary selector of deposition sites. A two-dimensional multibody DLA (MBDLA) model extended the analysis to include the electrolyte ion concentration and interfacial kinetics, generating a morphological phase diagram that maps transitions from sparse fractals to dense deposits.
This research aligns well with the current electrochemomechanical knowledge on electrodeposition; dendrite morphology is governed by a complex interaction of electrolyte concentration gradients, local current-density heterogeneities as well as multitude of chemical and mechanical interfacial phenomena. Based on the insights of our studies, the thesis proposes mitigation approaches that focus on current distribution homogenization, high concentration, high diffusion electrolytes and engineered interphases that regulate interfacial kinetics.
The developed methodologies, presented in this work, offer robust tools for predicting dendritic behaviours and designing safer and more efficient LMBs. Future research directions include expanding the use of MRI as a non-destructive analysis method for battery research, as well as ion-transport based dendrite mitigation methods.