3, The Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley, California, United States
1, Department of Materials Science and Engineering, Stanford University, Stanford, California, United States
2, Stanford Synchrotron Radiation Lightsource, SLAC National Accelerator Laboratory, Menlo Park, California, United States
5, The Molecular Foundry, Lawrence Berkeley National Laboratory, Berkeley, California, United States
6, Materials Science Division, Lawrence Berkeley National Laboratory, Berkeley, California, United States
7, Department of Materials Science & Engineering, University of California, Berkeley, Berkeley, California, United States
Oxygen redox has garnered intense interest as a means to increase the energy density of transition metal (TM) oxide positive electrodes in lithium ion batteries, as it enables additional lithium (de)intercalation capacity at high voltages beyond the usual TM redox capacity. However, most oxygen-redox-active materials discovered to date suffer from large charge-discharge voltage hysteresis and irreversible voltage fade over extended cycling, limiting their practical use. Several hypotheses have been proposed to explain the nature of the oxidized oxygen species in order to guide improvements to the electrochemical properties of oxygen redox. The general consensus is that in more ionic oxides (e.g. those with certain 3d TMs), reactive and unstable O– species are created when oxygen is oxidized, whereas in more covalent systems (e.g. those with 4d and 5d TMs) the increased hybridization allows for the formation of stabilizing, long (~ 2.3 Å) O2n– dimers. Accordingly, recent work has attempted to tune the covalency of the TM–O interactions to improve the reversibility of oxygen redox in 3d materials, but has found limited success. In this presentation, I propose that consideration of only electronic structure properties is insufficient to explain the electrochemical behaviors associated with oxygen redox, and that the intrinsic coupling of oxygen redox to defect formation and cation disordering is in fact more significant in determining its reversibility. I first show that in the commercially promising lithium-rich Ni/Mn/Co oxides, oxygen oxidation occurs simultaneously with migration of TMs into Li sites, forming TMLi defects in the material bulk. By drastically altering the local oxygen coordination environment, these defects lower the oxygen redox voltage, which we observe experimentally to fall by ~ 1V after the first charge. This redox voltage shift is a major driver of the charge-discharge voltage hysteresis in oxygen-redox-active materials and cannot be explained without considering the link between oxygen redox and structural evolution. By then investigating two model systems, I reveal the origin of the correlation between oxygen redox and TMLi formation, wherein the oxidation of oxygen is not sufficiently stabilized by either long (~ 2.3 Å) O2n– dimers or O– species, and instead promotes the formation of short (~ 1.4 Å) O2n– dimers and short TM=O π bonds. These species require the decoordination of oxygen to single-TM-coordinate, which is realized in the layered structure through the formation of TMLi defects. I show spectroscopically that these defect-localized oxidized oxygen species exist in 3d, 4d, and 5d systems, suggesting that the electrochemical properties of oxygen redox in TM oxides may depend more on the structural mechanism of oxygen decoordination than the identity of the TMs. Finally, I use this understanding to propose a new pathway to achieving reversible oxygen redox by employing new structures outside the layered framework.