Electrolyte Composition in Li/O₂ Batteries with LiI Redox Mediators: Solvation Effects on Redox Potentials and Implications for Redox Shuttling

Abstract

The use of LiI as redox mediator for the charge reaction in nonaqueous Li/O 2 cells has been widely studied recently, as a possible means to fulfill the great promise of the Li/O 2 system as a high energy density "beyond Li-ion" battery. In this work, we highlight the importance of considering the redox potential for both the I − /I 3 − and I 3 − /I 2 redox couples and how the electrolyte solvent (here tetraglyme (G4) and dimethyl sulfoxide (DMSO)) and concentration (here 1.0 and 2.8 M) have a profound influence on these potentials. Through a combination of galvanostatic cycling, electrochemical mass spectrometry, and cyclic voltammetry, we thus consider the influence of solvent and electrolyte concentration on both the redox mediation and redox shuttle processes and suggest that this important aspect must be considered for further studies with mediators in Li/O 2 and related systems. We demonstrate that, in our system, 100 mM LiI in 1.0 M Li[TFSA]/DMSO provides the most effective redox mediation among the electrolytes we have studied but conversely exhibits the highest degree of redox shuttling (in the absence of O 2 ). The balance between effective limitation of redox shuttle and ease of mediator diffusion to discharge products is of great importance and should be considered in any future cell design utilizing a mediator.

Liquid‐Based Janus Electrolyte for Sustainable Redox Mediation in Lithium–Oxygen Batteries

“…Because DMSO is expected to be a more solvating medium with a higher ε (46.7) than TEGDME (ε = 7.8), the potential crossover of RM + from DMSO to TEGDME in the Janus electrolyte would be subsequently suppressed, restricting the shuttle effect. As a proof of concept, we employed LiI as an RM, [ 51–54 ] dissolved it selectively in the cathode‐side electrolyte, i.e., 1 m LiTFSI DMSO, and monitored its diffusion over time. Figure 1e shows that liquid‐based Janus electrolyte of 1 m LiTFSI DMSO/2.3 m LiTFSI TEGDME successfully confined the oxidized RM (16.7 × 10 −3 m I 3 − ) within the DMSO phase without any apparent crossdiffusion toward the TEGDME phase.…”

Electrode Protection in High-Efficiency Li–O2 Batteries

“…Thereby, it enhances the stability of the carbon cathode and the Coulombic efficiency, rate capability, as well as durability of Li–O 2 batteries. 2 For example, the use of tetrathiafulvalene (TTF) and LiI as OER RMs could reduce the charge potential down to 3.4 and 3.3 V, 46 , 50 respectively, as a result, much-improved battery performance was achieved. Despite their advantages, the mobile nature of RM can make it cross over to the Li anode, leading to rapid failure of RM, active Li loss of the Li anode, self-discharge or even short circuit of the cell.…”

Towards Reversible and Moisture Tolerant Aprotic Lithium-Air Batteries

“…The signal at m/z= 34 (Fig. 3d -orange) is assigned to either to 17 O- 17 O or 18 O- 16 O. Since the signal intensity of the m/z= 33 and 34 signals are similar in magnitude, this indicates that the m/z= 33 signal is not due to the 17 O in O 2 gas, since the natural abundance of 17 O and 18 O are 0.04 and 0.2 %, respectively.…”

“…It is now established that the (I -/I 3 -) redox potential is solvent-dependent, and therefore that the oxidizing power of I 3can be tailored by means of adjusting the physicochemical parameters (i.e., dielectric constant, ionic strength, Gutmann acceptor/donor numbers, etc.) of the electrolyte [9,[16][17][18]. Ionic liquids (IL) have been shown to alter the solvation/coordination environments of this redox system, therefore substantially affecting the (I -/I 3 -) redox potential [16], while combining high ionic conductivity with low volatility [19][20][21][22][23][24][25][26].…”