Monitoring the Location of Cathode Reactions in Li-O2systems

Abstract

Lithium-oxygen (Li-O2) batteries have been considered as a promising alternative for high energy storage systems due to their exceptional theoretical energy density. However, many obstacles need to be overcome before developing a safe, green and efficient commercial dispositive [1,2].One of the key issues of these batteries is the employed electrolyte. Carbonates have been inherited from Li-ion technology, becoming their use popular in the early steps of development of Li-O2 systems. However, due to their decomposition at high potentials [3,4], they are slowly being replaced by other type of electrolytes such as glymes or ionic liquids. In order to study these processes, in addition to common electrochemical characterisation techniques, X-ray photoelectron spectroscopy (XPS), impedance spectroscopy and transmission electron microscopy (TEM) have proved to be fundamental techniques for in situ monitoring of reaction products that take place during charge and discharge [5-7]In this work, it is proved that the use of organic solvents with low oxygen solubility reduces the active area for the oxygen reduction reaction (ORR) to the oxygen cathode electrolyte triple interface (OCETI). For this aim electrodes loaded with different carbon masses have been cycled in different conditions and post-mortem analyzed by means of X-ray diffraction (XRD), scanning electrode microscope (SEM) and x-ray photoelectron spectroscopy (XPS). The obtained data evidence that the most of the cathode is not active in each stage of the cycling process (Figure 1.a and 1.b). Consequently, expressing the obtained specific capacity values in mA h g-1 units or the current densities in mA g-1 becomes undesirable as it leads to non reasonable comparisons. Moreover, the analysis of the cyclability has evidenced the competition between the OCETI (Figure 1.c) and the applied intensity (Figure 1.d) as the main limiting factors for the performance of these Li-O2 systems.