On the Solvation of Redox Mediators and Implications for their Reactivity in Li-Air Batteries

Abstract

<jats:p>Lithium-air batteries are a promising energy storage technology for transport applications, given their exceptionally high energy density. However, their development is significantly hampered by high overpotentials, which lead to poor efficiency and short lifetimes. Redox mediators provide a solution to this problem by shuttling electrons from the electrode to the active species at just above the redox potential of the mediator. Thus, knowing the redox potential and having the ability to tune it are critical to electrochemical performance. We focus on LiI as a model mediator—given its additional role in controlling LiOH vs Li<jats:sub>2</jats:sub>O<jats:sub>2</jats:sub> chemistry—and use cyclic voltammetry (CV), NMR, UV/Vis spectrometry, and molecular dynamics (MD) simulations to monitor the effects of electrolyte composition on solvation. Li<jats:sup>+</jats:sup> and I<jats:sup>–</jats:sup> solvation in common Li-air solvents, the electrochemical implications, and the applicability of each technique to probe the nature of the solvation shell and its effect on the electrochemical properties are explored. Starting with a simple thermodynamic model, we then used UV/Vis spectrometry to probe I<jats:sup>–</jats:sup> solvation, <jats:sup>1</jats:sup>H NMR spectroscopy to study water solvation and <jats:sup>31</jats:sup>P of the probe molecule triethylphosphine oxide (TEPO) to explore Li<jats:sup>+</jats:sup> solvation; we find that no single descriptor can provide an accurate description of the solvation environment. Instead, we use all these methods in combination with the MD results to help rationalise the CV data. We find that the I<jats:sup>–</jats:sup> solvation improves significantly in tetraglyme (G4), with increasing salt and water concentration, but minimal effects on changing salt/water concentrations are seen in DMSO. In contrast, increasing salt concentration increases the Li<jats:sup>+</jats:sup> activity in DMSO but not in G4. Furthermore, a simple model considering the equilibria between the different species was used to explain the <jats:sup>1</jats:sup>H NMR data.</jats:p>