Rate and Voltage Dependence of the Electrode-Electrolyte Interface in Ni Containing Lithium-Ion Cathode Materials

Abstract

Performance, safety, lifetime, and cost of lithium-ion batteries (LIB) are strongly influenced by the active materials used. As a result, cell chemistries are selected with the end goal in mind. For applications requiring superior power and safety, cells containing LiFePO4 cathodes and graphite anodes with an organic liquid electrolyte are frequently used. The major drawback of LIB cells designed for high-power is a relatively low energy density compared to high-energy cells, which conversely have poor rate capability. While graphite anodes remain a constant in most commercial LIB, recent research has identified cathode materials which would potentially mitigate the tradeoff between power, energy, and safety. In particular, spinel structured “LNMO” (LiNi0.5Mn1.5O4) and layered nickel-rich “NCM” compounds (LiNixMnyCozO2, X ≥ 0.5) have reportedly high energy densities relative to LiFePO4 while maintaining good power capability and thermal stability (1) (2). While LNMO and NCM compounds have similar composition and produce Ni4+ when fully charged, several reports have highlighted variations in surface chemistry resulting from electrolyte-electrode interactions as a function of cathode stoichiometry (3; 4; 5). One issue known to influence the surface chemistry and associated cycling stability of transition metal oxide cathode materials is the upper voltage cutoff needed to extract lithium during charge. Layered compounds can typically be reversibly delithiated within a voltage range of 4.2 to 4.6 V vs. Li/Li+ depending on the specific stoichiometry, while the spinel must be charged in excess of 4.8 V vs. Li/Li+, which historically has been a major hurdle to application. These upper voltage cutoff are already within the range of expected electrolyte oxidation, which is exacerbated by over potentials produced when charging at high current densities which further increase the upper voltage cutoff required to fully delithiate the cathode. This presentation will focus on analysis of the electrode-electrolyte interface, which is known to be a critical indicator of cell safety and lifetime, as a function of high-rate cycling and upper voltage cutoff. Composite cathodes containing spinel LNMO and layered NCM523 were cycled in a conventional lithium-ion electrolyte at various rates of charge and discharge, then disassembled in an inert atmosphere for analysis by X-ray Photoelectron Spectroscopy (XPS). Capacity and electrochemical impedance measurements were correlated with XPS analysis. Our results indicated that even when cycled in identical electrolytes spinel and layered compounds have notably different responses to an increased charge rate, specifically high rate charging (5C rate) is actually found to benefit LNMO, leading to lower impedance rise and associated formation of LiF, while the opposite is true for NCM523. One complicating factor in the XPS analysis of composite electrode is the presence of inactive components, specifically the binder and conductive additive. Both of these compounds dilute the overall signal detected from the active material, and some reports suggest may even contribute to the specific surface chemistry observed. To remove these effects, sol-gel films of identical composition to the composite electrode active materials were also synthesized and subjected to electrochemical cycling in the same electrolyte. [1] Review-Li-Rich Layered Oxide Cathodes for Next-Generation Li-Ion Batteries: Chances and Challenges. Rozier, P and Tarascon, JM. 14, 2015, Journal of the Electrochemical Society, Vol. 162, pp. A2490-A2499. [2] Spinel LiNi0.5Mn1.5O4 and its derivates as cathodes for high-voltage Li-ion batteries. Liu, G Q, Wen, L and Liu, Y M. 2010, Journal of Solid State Electrochemistry, Vol. 14, pp. 2191-2202. [3] Coupled LiPF6 Decomposition and Carbonate Dehydrogenation Enhanced by Highly Covalent Metal Oxides in High-Energy Li-Ion Batteries. Yu, Y, et al. 2018, Journal of Physical Chemistry C, Vol. 122, pp. 27368-27382. [4] XPS Studies of Surface Chemistry Changes of LiNi0.5Mn0.5O2 Electrodes during High-Voltage Cycling. Quinlan, RA, et al. 4, 2013, Journal of the Electrochemical Society, Vol. 160, pp. A669-A677. [5] Probing the Electrode-Electrolyte Interface in Cycled LiNi0.5Mn1.5O4 by XPS Using Mg and Synchrotron X-rays. Mansour, AN, et al. 14, 2016, Journal of the Electrochemical Society, Vol. 163, pp. A2911-A2918.