The Effects of Water Concentration on Lithium Deposition/Dissolution Toward Practical Operation of Lithium-Air Batteries

Abstract

Lithium (Li)–air batteries are promising devices for the use in sustainable energy management systems as they have the potential to achieve significantly higher energy density than current state-of-the-art Li-ion batteries.1 However, there are many problems for the application of Li-air batteries. Low coulombic efficiency (C.E.) of Li deposition/dissolution on a Li metal anode caused by various side reactions is one of the critical issues to be overcome.2 Moisture contaminated from air is known to have influence on the anode reactions and hence on the C.E..1 For instance, Zaghib et al. reported the apparent increase of discharge capacity and the formation of a porous lithium hydroxide (LiOH) layer as water (H2O) concentration (C H2O) increased.3 Osaka et al. also examined the effects of H2O on the C.E. by using carbonate-based electrolytes containing carbon dioxide. The C.E. drastically increased to a maximum of 88.9% with increasing C H2O up to 35 ppm.4 They concluded from X-ray photoelectron spectroscopy (XPS) results that the solid electrolyte interface (SEI) composed of lithium carbonate and lithium fluoride (LiF), which were formed by reactions involving H2O, effectively suppressed side reactions. However, for the purpose of evaluating the C.E. toward the practical application of Li-air batteries, ether-based electrolytes, which are often used in Li-air batteries owing to their higher stability against reactive oxygen species, should be employed. Therefore, in this work, we examined the effects of C H2Oon the C.E. by using ether-based electrolytes. In our experiments, coin-type cells composed of Ni foil (as working electrode) and Li foil (as counter electrode) were used with tetraethylene glycol dimethyl ether (TEGDME) containing 1 M lithium bis(fluorosulfonyl)imide (LiFSI) and various concentration of H2O as electrolytes. The C.E. was estimated using the cycling protocol proposed by Koch.5 Figure 1 shows the relationship between the C.E. and C H2O. Although the C.E. was in the range between 40% and 60% in the absence of additional H2O, it increased with increasing C H2O and reached a maximum at 80 % at C H2O of 1000 ppm. Then, the C.E. turned to decline when C H2O was further increased. It is known that the morphology of Li deposits significantly affects the C.E..2 Therefore, we inspected the morphology of the Li deposits after the initial deposition by scanning electron microscope (SEM). The homogeneity of Li deposits was improved as C H2O increased to 1000 ppm, whereas Li metal was deposited locally and formed aggregates with further increase of C H2O. Importantly, the degree of homogeneity estimated from the SEM images correlated with the C.E., which was in good agreement with the literature.2 It is known that the SEI composition is one of the crucial factors affecting the Li deposition processes and hence the cycling performance.2 Therefore, we next investigated the chemical composition of SEI by XPS with the aim of elucidating the mechanisms underlying the relationship between the C.E. and the morphology of Li deposits. It was revealed that the upper part of the SEI was dominated by LiF and organic compounds, irrespective of the presence or the absence of H2O. In contrast, the inner part was mainly composed of inorganic species, and the amount of lithium oxide (Li2O) and LiOH was revealed to increase with increasing the C H2O. Taken SEM and XPS results together, it can be suggested that Li metal was consumed by the reactions with H2O, converting to Li2O and LiOH.