小沢 文智

Abstract Redox mediators (RMs) suppress the charging overpotential to enhance the cycle performance of lithium-air batteries (LABs), but inappropriate RM incorporation can adversely shorten cycle life. In this study, three typical organic RMs; tetrathiafulvalene (TTF), 2,2,6,6-tetramethylpiperidine 1-oxyl (TEMPO), and 10-methylphenothiazine (MPT), were incorporated into the air-electrode (AE) of the LAB (RM-on-AE), rather than dissolving them in the electrolyte (RM-in-EL), to maximize the RM effect throughout the cycle life. The discharge/charge cycle test confirmed that the cells with RM-on-AE prevented the reductive decomposition of RM with the lithium anode, deriving the RM effect for a longer cycle life than the cells with RM-in-EL. The measurement of AE deposits revealed that the TTF- and TEMPO-on-AE cells failed to generate a quantitative amount of Li2O2 discharge product. In contrast, the MPT-on-AE provided a 96% yield of Li2O2 after the first discharge because of the reductive tolerance of the MPT as organic RM. The quantitative analysis also revealed an accumulation of Li2CO3 on the AEs, along with the generation of carboxylate, as the side products of irrelevant battery reactions. This study provides a practical methodology for selecting RMs and their incorporation for developing long-life LABs.

Four ethers were compared as solvents of lithium naphthalenide (Li-NTL) solutions to pre-dope Li into Si electrodes. The solvents of the Li-NTL solutions affected the stability and equilibrium potential (V (eq)). X-ray diffraction, thermodynamic characterization and ultraviolet-visible (UV-vis) spectroscopy were used to clarify the effects of the solvation structure, the lowest unoccupied molecular orbital (LUMO) energy of the solvent molecule and the ion pair structure between Li+ ions and naphthalenide radical anions ([NTL](center dot-)) on doping capacity. A Li-NTL solution having a low V (eq) and sufficient stability under potentials as low as that of Li metal was found to provide the highest pre-doping capacity. In particular, a 2-methyltetrahydrofuran (MeTHF) solution exhibiting the lowest V (eq) showed a pre-doping capacity as high as 3250 mAh g(-1) after 24 h. UV-vis spectra and Walden plots indicated that a Li-NTL solution using MeTHF provided less dissociation than a tetrahydrofuran (THF) solution. The doping capacity is evidently determined by the V (eq) of the Li-NTL solution as a consequence of the dissociation equilibrium of the ion pair of the solvated Li+ ion and [NTL](center dot-) radical ions.