Abstract
Lithium metal is often deemed the "Holy Grail" of rechargeable battery technology, due to its high theoretical capacity and low electrochemical potential. However, the cascading heterogeneity from the formation of the solid-electrolyte interphase (SEI) up to the non-uniform current distribution leads to both poor efficiency via rapid consumption of active materials as well as safety issues from treacherous dendrite growth. An SEI that is ionically conductive while electrically resistive, viscoelastic, electrochemically stable, and possesses good mechanical properties is paramount for hindering these failure mechanisms. The discovery of novel approaches to producing an ideal SEI is a crucial piece to solving the lithium metal anode puzzle.
One route of controlling the SEI is through interface engineering by producing a stable artificial SEI. In this work, a mixed artificial SEI was produced with aim toward improving ionic conductivity while retaining electrochemical stability and mechanical properties. This mixed SEI embraces the unavoidable surface heterogeneity, all the while preventing dendrite growth that plagues lithium metal batteries. A mixed artificial SEI, as well as a single-component artificial SEI, were produced through a simple and rapid deposition method and tested to observe their failure mechanisms compared to bare lithium.
Symmetric cells (i.e., Li||Li) were utilized to isolate the failure mechanisms of these different anode treatments. The failure mechanism of the single-component artificial SEI cells shifted to failure via electrolyte consumption rather than short circuiting seen with bare lithium. This revealed that artificial interphases led to reduced dendrite propagation. Additionally, a reduction in cell potential range was observed with the mixed artificial SEI when compared to the single-component artificial SEI, as well as improved cyclability. Upon further investigation, it appears that the added component of the mixed artificial SEI, which was meant to improve the ionic conductivity, dissolved into the electrolyte upon initial exposure. This dissolution left behind a new SEI structure, theoretically producing easier ionic pathways and thus resulting in decreased cell overpotential. Dissolution was confirmed with high-resolution surface characterization techniques. Additionally, symmetric cell electrodes rinsed with solvent prior to testing retained superior cell performance. Though this dissolution should lead to increased available "hot spots", the growth of dendrites was hindered. This work discusses the effect an artificial SEI has on failure mechanisms, as well as provides insight into the methods of producing stable, long-life lithium metal batteries. This includes a focus on how nanoscale heterogeneity effects cell-scale performance via high-resolution characterization and computational simulations.
Different surface to sub-surface characterization techniques were utilized to measure properties of artificial SEIs and bare lithium. Scanning electron microscopy with energy dispersive X-ray spectroscopy (SEM/EDS) provided elemental make-up and distribution, as well as morphological characteristics. X-ray photoelectron spectroscopy (XPS) provided the chemical state of surfaces, as well as sub-surfaces with the use of depth profiling. Different atomic force microscopy (AFM) techniques were used to spatially measure nanoscale properties such as work functions with the Kelvin probe technique (SKPFM) and quantifiable nanomechanical properties (QNM), as well as observe kinetic mechanisms with in-situ electrochemical methods (EC-AFM). Lastly, computational simulations from the atomistic scale via density functional theory (DFT) up to molecular dynamics (MD) were utilized to support assumptions on the reaction mechanisms derived from experimental observations.
Figure 1