Nickel-rich layered lithium nickel cobalt manganese oxides (NCMs, Li1+δNixCoyMnzO2, x+y+z+δ = 1 and δ > 0) are increasingly used as cathode active materials (CAMs) for lithium-ion batteries due to their high specific capacity; they are also expected to play a major role in future battery technologies, such as all-solid-state batteries. With increasing nickel content, however, the gained capacity is often accompanied by shorter cycle life due to undesired side reactions, resulting e.g. from the decomposition of residual lithium salts and surface contaminants that are present in greater amounts on Ni-rich materials,[1] as well as from the reaction of evolved lattice oxygen with the electrolyte,[2] which occurs at lower potentials for higher nickel content.[3] These processes degrade the NCM active material upon charge-discharge cycling and dramatically increase the impedance of the entire cell, eventually leading to cell failure. One way to minimize side reactions is to reduce the specific surface area through a greater NCM crystallite size, being known in the literature under the name of single crystals, which are expected to maintain their pristine surface area upon cycling.[4,5] By this approach, however, the NCM surface area normalized current of the CAM is significantly increased, possibly leading to critical overpotentials and compromising its rate capability.In this study, we investigate the electrochemical characteristics of two µm-sized NCM622 materials with a similar pristine BET surface area of ~0.3 m²/g: a polycrystalline material (PC) with secondary agglomerate diameters of ~5-10 µm comprised of primary crystallites with particle diameters on the order of ~10-1 µm, and a single crystalline material (SC) with individual primary crystallites of ~5 µm, both shown in the SEM images in Figure 1. Due to the highly unequal particle structure of the two material classes, one would expect significant differences in rate capability, long-term cycling behavior, extent of gassing at high state of charge (SOC), and the change of CAM surface area over the course of extended charge/discharge cycling.Calendered electrodes with PC or SC NCM622 with electrode loadings of ~12 mgAM/cm² were assembled in half-cells with a lithium metal reference electrode and then subjected to charge and discharge rate tests with potential cutoffs controlled by the reference electrode. Since clear differences in the rate-dependent discharge capacity of the two materials are observed, the impact of the particle morphology and the cause on the overpotential during (dis)charge are then further illuminated. In order to show that the difference in rate capability is owed to differences in the CAM surface increase upon cycling due to particle cracking that is induced by the NCM volume change upon (de-)lithiation, the Kr-BET surface area of pristine and charged electrodes of both PC and SC NCM622 is quantified. By this method, it is proven that, depending on the active material type, the CAM surface area may strongly increase, already during the very first charge. Due to this fundamental property, the amount of gas released from both CAMs is expected to differ significantly. We quantify the evolved O2 and CO2 during the first four cycles to >80 %SOC using On-Line Electrochemical Mass Spectrometry (OEMS). It is shown that the ratio of the surface area of both materials in charged state correlates well with the ratio of the amounts of gas released during electrochemical cycling. The obtained results on the fundamental properties of the two material classes on rate capability and gassing are complemented with long-term cycling experiments in full-cells at different upper cutoff potentials, chosen to be either below and or above the limit of oxygen release of NCM622 in order to analyze the evolution of the NCM622 particle morphology over extended charge/discharge cycling.In summary, this study gives comprehensive insights into the fundamental properties of poly- and single crystalline NCMs and into the impact of typical CAM failure mechanisms on the long-term cycling performance of these two material classes. Reference s : [1] J. Paulsen and J. H. Kim, USA Pat. US2014/0054495A1 (2014).[2] A. T. S. Freiberg, M. K. Roos, J. Wandt, R. de Vivie-Riedle, and H. A. Gasteiger, J. Phys. Chem. A, 122, 8828 (2018).[3] R. Jung, M. Metzger, F. Maglia, C. Stinner, and H. A. Gasteiger, J. Electrochem. Soc., 164, A1361 (2017).[4] Yongseon Kim, ACS Appl. Mater. Interfaces, 4, 2329 (2012).[5] G. Liu, M. Li, N. Wu, L. Cui, X. Huang, X. Liu, Y. Zhao, H. Chen, W. Yuan, and Y. Bai, J. Electrochem. Soc., 165, A3040 (2018). Figure 1
Read full abstract