Unveiling Electrolyte-Dependent Coordination Site Dynamics for Redox Mediator Design in Lithium-O ₂ Batteries: Exchange vs Rearrangement

Developing multifunctional redox mediators (RMs) is critical for mitigating cathode passivation, parasitic reactions, and high discharge overpotential in Li–O 2 batteries. While the computational studies of RMs relied on gas-phase electronic descriptors, here, we investigate through explicit-solvent ab initio molecular dynamics (AIMD) that the solution-phase mechanism is governed by previously unrecognized solvation-dependent coordination site exchange dynamics. Using benzo[1,2-b:4,5-b′] dithiophene-4,8-dione (BDTD) as a model RM across varying electrolyte environments with dimethyl ether (DME), Dimethyl sulfoxide (DMSO), and acetonitrile (ACN) solvents, we uncover a unique mechanistic divergence. In bidentate solvents like DME, a flexible coordination rearrangement enables the spontaneous formation of a BDTD:Li complex via a transient trigonal-bipyramidal intermediate. Conversely, this process is sterically hindered in monodentate solvents (DMSO and ACN), forcing the reaction to proceed via a significantly slower solvent-reorganization pathway. The flexibility is absent in monodentate solvents, such as DMSO and ACN. The complex formation in the DME environment proceeds through coordination rearrangement site exchange phenomena, whereas in DMSO and ACN, it proceeds through solvent reorganization phenomena. The descriptor for controlling redox activity evaluated by screening 18 BDTD derivatives with various substituents (-NMe 2, -NH 2, -Me, -OMe, −OH, -Cl, −CN, -NO 2, and −COOH) at two positions is systematically evaluated. Structure–property relationships show that the nature of the substituent, positional effects, hydrogen bonding, and reduced state stability collectively govern the reduction potential and discharge overpotential. One of the positioned −COOH functionalized BDTD provides bidentate coordination, stabilizing the reduced state and leading to a more favorable redox shift and lower overpotential. RM activity cannot be predicted solely by LUMO energy; instead, a combined descriptor involving electronic effects, substitution position, and reduced-state stabilization is required. These results establish a new framework for RM chemistry, shifting the design paradigm from simple electronic tuning to the active management of the dynamic solvation shell, providing a blueprint for the next generation of high-efficiency electrochemical energy storage.