Operando Visualization of Li Metal Anodes

Abstract

A pressing challenge facing the 21st century world is replacing fossil fuels with a competitive and renewable energy source. However, enabling widespread renewable energy sources requires improvement in the energy density of current battery technology, especially for transportation applications. A promising approach to increase the energy density of current state-of-the art Li-ion batteries is to replace the graphite anode with pure metallic Li. These anodes offer a specific capacity that is 10x greater than that of graphite and are a critical component of next-generation batteries, such as Li-Air and Li-Sulfur. Incorporating a Li metal anode into a rechargeable battery would be a breakthrough technology that could further aid in the electrification of vehicles. However, the commercialization of metallic Li is limited by poor cycle life and safety concerns, which are impacted by the dynamic morphological evolution of the anode surface during cycling.

The goal of this thesis is to improve our understanding of the morphological evolution of the Li surface and the relationship between surface morphology and performance. A platform for plan-view operando video microscopy is developed, which simultaneously affords optical access to the anode surface and achieves a uniform current distribution across the working electrode surface. Using this platform, various factors that govern the morphological evolution of the Li surface are explored. Nucleation is characterized as a function of current density and electrode microstructure. A positive correlation is observed between nucleation density and current density for both dendrites and pits. However pit nucleation is found to be more sensitive to current density than dendrite nucleation. Furthermore, surface grain boundaries are observed to be preferential nucleation sites for dendrite and pits during initial cycling. However in subsequent cycles, preferential nucleation is found to transition from the surface grain boundaries to the pit edges.

The degree of reversibility during Li plating and stripping is then explored. The size of individual dendrites is tracked throughout cycling by quantifying their volumetric contraction. A clear correlation between dendrite size and reversibility is not observed, however reversibility is found to be sensitive to pitting. Dendrite reversibility is observed to improve when nucleation occurs at a pit edge compared to a pristine surface. Additionally, this reversibility is found to be sensitive to the size of the pits in which nucleation occurred. Overall, dead Li formation is found to be more sensitive to nucleation than growth.

Operando focus variation microscopy is then integrated with the plan-view platform to better understand the mechanisms that link pit formation to dendrite nucleation and reversibility. Using this upgraded capability, the nucleation and growth of individual pits is mapped in three dimensions. Pit expansion is observed to be highly anisotropic, where expansion occurs more rapidly along the anode surface than into the depth of the electrode. Concurrently, facets are observed to form at the pit edges, indicating that the underlying mechanism for anisotropic stripping is determined by the properties of the electrode crystal lattice. Additionally, pit expansion is observed to preferentially occur along surface grain boundaries, highlighting the importance of electrode microstructure on anode morphology.

Overall, this thesis provides mechanistic insight into the morphological evolution and poor reversibility of Li metal anodes through the use of plan-view operando optical analyses. In the future, the understanding developed can inform rational solutions that are aimed at improving the performance of next-generation Li metal batteries.