Abstract

Nonaqueous lithium-oxygen (Li-O2) batteries have attracted intensive research interest owing to their potential to provide gravimetric energy density 3-5 times that of conventional Li-ion batteries.1-2 The practical application of this technology has been hindered by poor cycle life (<100 cycles) and low round-trip efficiency (65-70%). These challenges are related to sluggish reaction kinetics and side reactions at the oxygen electrode.3-4 Applying redox mediator has been demonstrated an effective strategy to improve the reaction kinetics of Li-O2 reactions.5-8 However, limited understanding is available about the effect of redox mediator on the reaction stability of Li-O2 batteries. In this work we will study the effect of redox mediator on the reaction stability of the Li-O2batteries. We first studied the instability in cycling a Li-O2 cell without redox mediator. Applying high-temporal resolution on-line electrochemical mass spectrometry (OEMS), we show that the charging process of the Li-O2 cell consists of several oxygen-evolving stages, followed by CO2 evolution. First, the oxygen gas is evolved at ~2e-/O2 between 3.0 – 3.5 V vs. Li (VLi). As the charge voltage reaches 3.5 VLi, the rate of oxygen evolution starts to decrease accompanied with hydrogen evolution, as shown in Figure 1a. We further characterize the cycled O2-electrode and the electrolyte with 1H NMR. As shown in Figure 1b, the formation of side products including formic acid, acetic acid and acetone is confirmed. It was reported that generation of acetone can be mitigated by cycling with a redox mediator.9 Therefore, it is critical to investigate if and how the redox mediator influence the reaction stability of Li-O2reactions. The effect of redox mediator on charging and cycling instabilities will be discussed. Origin responsible for the instabilities as well as the mitigation mechanism of redox mediator will be discussed. Figure caption: Fig. 1. (a)Voltage and gas evolution profiles during charging of a Li-O2 battery without redox mediator after discharged to 1000 mAh/g. (b) 1H NMR spectrum of the carbon cathode and separator after 5 cycles of galvanostatic discharge and charge without redox mediator. Acknowledgements This work is substantially supported by a grant from the Research Grants Council (RGC) of the Hong Kong Special Administrative Region (HKSAR), China, under Theme-based Research Scheme through Project No. T23-407/13-N, and partially supported by a RGC project No. CUHK24200414.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call