Synthesis and electrochemical properties of nanocrystalline Li[NixLi(1−2x)/3Mn(2−x)/3]O2 prepared by a simple combustion method

Abstract

Nanocrystalline Li[Ni_xLi_(1−2x)/3Mn_(2−x)/3]O₂ powders were prepared by a simple combustion method and investigated using X-ray diffraction (XRD), X-ray absorption spectroscopy (XAS), scanning electron microscopy (SEM), particle size analysis (PSA), and galvanostatic charge/discharge cycling. According to the XRD analysis, single-phase compounds with a layered structure were obtained for powders with 0 ≤ x ≤ 0.25, while mixtures were obtained for powders with 0.30 ≤ x ≤ 0.50. Rietveld analysis revealed that single-phase Li[Ni_xLi_(1−2x)/3Mn_(2−x)/3]O₂ is basically a layered rock-salt structure in which a small amount of Ni occupies the 3a sites. The initial discharge capacity of a Li/Li[Ni_xLi_(1−2x)/3Mn_(2−x)/3]O₂ cell with x = 0.20 was about 288 mA h g⁻¹, corresponding to about 91% of the theoretical value, when it was cycled in the voltage range of 4.8–2.0 V with a specific current of 20 mA g⁻¹ at 30 °C. As far as we know, charge/discharge cycling on an Li/Li[Ni_0.20Li_0.20Mn_0.60]O₂ cell gives the highest discharge capacity of 288 mA h g⁻¹ among the LiMO₂-based (M = Co, Ni, and Mn) cathode materials. A very promising factor for high-rate capability applications was an excellent rate capability in continuous cycling at specific currents ranging from 20 mA g⁻¹ to 900 mA g⁻¹, due to the nanocrystalline particle size of 80–200 nm. The origin of the 4.5 V plateau was investigated by means of weight loss measurement and XAS for the charged/discharged electrodes. The weight loss measurement for the charged electrodes gave indirect evidence that the 4.5 V plateau did not originate from the ejection of oxygen. In XAS, the Mn oxidation state of 4+ did not change during the charge/discharge process, and surprisingly the Ni did not further oxidize beyond about 3+.