Lithium metal batteries (LMBs) are highly attractive for weight-limited applications due to their high theoretical gravimetric energy density. Advancements in recent years have yielded huge improvements in their cycle life and safety. However, LMBs still show capacity degradation and premature cell death due to the formation of dead lithium and electrolyte dry out in realistic cell design and testing. Current collectors with 3-dimensional structures have been proposed and studied extensively in recent years as an attempt to alleviate these issues. Three dimensional structures are claimed to improve performance by reducing local current density, restricting lithium volume expansion, stabilizing the solid-electrolyte interphase (SEI), improving the kinetics of lithium nucleation by providing additional lithium nucleation sites, and other justifications. However, despite the expansive literature published on the topic, no consensus has been reached as to what physical, chemical, and electrochemical properties of a 3D current collector are most likely to result in improved performance. This presentation focuses on two studies of 3D current collectors: (1) a meta-study reviewing literature data and analyzing the impact of key anode properties on the cycle life and Coulombic efficiency (CE); and (2) an experimental study on the electrochemical performance of four commercially available anode in combination with four different electrolytes.Analysis of published data using statistical methods and data science approaches (linear correlation coefficient calculations, self-organizing maps) show almost no consistent improvements of CE or cycle life based on frequently utilized strategies. Preliminary results suggest that there is no significant improvement with increasing surface area, altering pore diameter size, use of a lithiophilic coating or additive, nor any particular morphology. The only significant increase in CE noted is seen when a conductive coating or additive is utilized in the 3D structure compared to when an insulative coating is applied, though cycle life does not show a similar increase.Based on these insights, an experimental study is presented that compares four commercially available copper current collectors with different structures (foil, foam, punched foil, and layered punched foil on unaltered foil). These structures were chosen and left unaltered due to the lack of evidence in the literature data review of benefits arising from alterations. In addition to varying anode structure, four electrolytes are utilized, one conventional electrolyte and three state-of-the-art electrolytes, to determine if performance trends hold regardless of electrolyte used. Average CE values are determined over the course of a cycling protocol consisting of plating/stripping from a lithium reservoir, mimicking cycling in high energy density pouch cells. Changes in cell impedance are evaluated at multiple stages of cycling using electrochemical impedance spectroscopy (EIS), and the morphology of the SEI and the plated lithium metal is examined using cryo-focused ion beam scanning electron microscopy (cryo-FIB SEM) to give further insight into the behavior of lithium on the current collectors. Average CEs calculated from this testing show no statistically significant differences between any of the structures. Significant differences in CE are observed, however, when the varied electrolytes are employed. Evaluation of cell impedance changes during cycling shows similar bulk and interfacial charge transfer resistances between structures with consistent electrolyte. The conventional electrolyte demonstrates lower bulk resistance due to a slightly higher ionic conductivity than the state-of-the-art electrolyte yet greater charge transfer resistance, predicting worse long term performance. Morphological changes demonstrated increased lithium thickness, resulting in greater percentages of mossy lithium growth and more SEI formation over cycling for the conventional electrolyte, aligning with the observed charge transfer resistance. Differences in plated lithium density are also observed between substrates regardless of electrolyte, informing on their application for lean electrolyte conditions.In all, the experimental study indicates that electrolyte chemistry has a greater effect on the efficiency and morphology of lithium plating and stripping than current collector architecture when excess lithium is used in cell design. Based on the results presented, punched copper foil is proposed to be the ideal current collector for quick scale-up practical high energy density LMBs due to high CE in newer generation electrolytes and lower mass than the other tests substrates.
Read full abstract