Increasing demands for high energy density Li-ion batteries as a source for power supply are demanding an optimization of cathode active materials (CAMs) and of the overall cell chemistry of batteries. Thus, the main interest of researchers has focused on nickel-rich (>80%) layered transition metal oxides such as NCMs (LiNixCoyMnzO2, with x+y+z=1) or NCAs (LiNixCoyAlzO2, with x+y+z=1) as CAMs, owing to their high capacity. However, as a result of the increase in Ni-content, the structural stability of these CAMs is significantly decreased at high upper cut-off potentials compared to those with lower Ni-content, resulting in capacity fading, which restricts their practically achievable capacity.[1] One possible culprit behind this capacity fading at high state-of-charge (SOC) is the release of reactive lattice oxygen, which is known for layered TM-oxides to occur at high SOC (i.e., starting at approx. 80% SOC).[2] The released oxygen is proposed to lead to the chemical oxidation of ethylene carbonate (EC), resulting in HF formation, which in turn is proposed to trigger the dissolution of TMs.[3] Furthermore, the electrochemical oxidation of the electrolyte is known to occur at high potentials, which as a result of its reactive reaction products, can induce further transition metal (TM) dissolution from the CAM.[4,5] The proposed chemical oxidation of the electrolyte by lattice oxygen suggests that this process should be strongly correlated to the dissolution of TMs from the CAM. However, as shown by operando X-ray absorption spectroscopy (XAS) and on-line electrochemical mass spectroscopy (OEMS) studies, the onset for the TM-dissolution, depending on the CAM under investigation, occurs at ~300 mV higher potentials compared to the onset for oxygen evolution.[2,3] This suggests that the dissolution of metals is rather induced by the electrochemical oxidation of the carbonate-based electrolyte, which was reported to initiate at potentials above 4.7 VLi at room temperature.[5] Both the electrochemical oxidation of battery electrolytes, as well as the dissolution of TMs from the CAM were reported to be strongly affected by temperature, while the onset potential for lattice oxygen loss is essentially unaffected, which further hints towards the significance of former.[5-7] Within this work we therefore conduct a detailed study on the temperature dependency of both the gas formation due to (electro)chemical electrolyte oxidation and the dissolution of TMs by means of OEMS and operando hard XAS, respectively. By applying a measurement protocol consisting of constant voltage-holds in steps of 50 mV at elevated potentials (see Figure 1), an excellent time as well as potential resolution enables the deconvolution of the onset for lattice oxygen release and the (electro)chemical oxidation of LP47 (1 M LiPF6 in EC:DEC 3:7 by weight) for a LiNi0.80Co0.15Al0.05O2 (NCA) CAM via OEMS. It will be shown that the onset for oxygen evolution coincides with a significant evolution of CO2 (see Figure 1b), indicating chemical oxidation of the electrolyte. However, the onset for POF3/PF5 evolution (marked in Figure 1b as “POF3”, since both gasses appear on the same mass channel)[8], which is a reaction product from the decomposition of LiPF6 caused by electrochemical electrolyte oxidation, is shifted to 300 – 450 mV higher potentials (depending on the temperature). It suggests a chemical oxidation mechanism of the electrolyte, which proceeds without significant H2O/H+ production. A change in temperature strongly affects this onset of “POF3” evolution, while O2 is released at the same degree of de-lithiation, regardless of temperature. To correlate this effect observed in the temperature-dependent gassing behavior to the dissolution of transition metals for this CAM, an operando hard XAS study at different temperatures will be conducted.
Read full abstract