Rechargeable batteries will be essential for allowing for the necessary sustainability transition society must undergo. Looking beyond current Li-ion batteries, the next-generation of batteries may employ a 'metal anode', where we cycle the active ion directly to and from its metallic state, rather than intercalating it into a graphite host. This would allow for far better battery performance. However, sodium and lithium metals are highly reactive, so rapidly form a solid-electrolyte interphase (SEI) layer across their surface due to reacting with the battery electrolyte. This sub-micron SEI layer is the rate limiting step for the transmission of ions from the electrolyte to the anode, and thus its properties are crucial for governing the stability and performance of the metallic anode.

The SEI's structure and composition can be tailored by modifying the electrolyte, as this leads to different electrolyte decomposition products forming the SEI. If we can understand the link between the nature of the SEI and the metal anode cycling behaviour we can then tailor the electrolyte such that it optimises the SEI that is formed. This prospect of the informed design of the SEI via electrolyte engineering, leading to a more stable and durable anode robust to dendrite formation and other sources of battery capacity fade, is one of the most promising approaches for realising practical high performance metal anodes.

Unfortunately, diagnosing the link between SEI properties and anode cycling behaviour is highly challenging; the SEI is a 'buried' solid-liquid interface that is not easily accessible; it is a highly reactive chemical environment, so cannot be exposed to ambient atmosphere; and the interface is fragile, so cannot be easily extracted. And on top of these concerns, the electrode cycling itself is a dynamic process that is difficult to map accurately through simple post-mortem characterisation. Operando imaging the of the interface and electrode in-situ are necessary.

In our recent work we have used operando liquid-cell TEM to understand the relationship between electrolyte, SEI, and electrode cycling performance for the cases of Li and Na metal anodes [1][2]. For the Li metal anode, we explored how a fluoride-rich interphase layer can encourage the uniform dissolution of lithium during discharge, allowing for more reliable repeated cycling. In the fluoride-poor condition, the formation and detachment of Li dendrites is clearly observed. With the Na metal anode, we identifed how the type of electrolyte solvent enables high cycling performance, with ether solvents suppressing gas evolution localised at the SEI on discharge. These gas bubbles block ionic cycling, preventing Na dissolution back into the electrolyte and thus inhibit good cycling performance.

[1] C. Gong., S. D. Pu, ... , A. W. Robertson. Adv. Energy Mat. (2021) DOI:10.1002/aenm.202003118

[2] C. Gong., S. D. Pu, ... , A. W. Robertson. Energy & Environ. Sci. (2023) DOI:10.1039/D2EE02606F