Atomistic simulation studies of Li2MnO3 Nanoarchitectures

Abstract

High-energy lithium-ion batteries (LIBs) are in high demand for the establishment of electric vehicles and hybrid vehicles in the automotive sector. Considering that the cathode is the limiting factor, it is of significant importance to discover cathode materials with a large number of lithiums, as this has a positive effect on the voltage, capacity and potential difference required for the advancement of the next-generation LIBs. Lithium manganite (Li2MnO3), owing to its huge capacity (458 mAh.g-1), nontoxicity and affordability has been considered an active cathode material for high-capacity LIBs. However, its capacity degradation and phase transformation during the cycling process are the main limiting factors to its commercialisation. The mechanisms behind these degradations are not yet fully understood to enable a complete diagnosis of the drawback. Sodium-ion doping at the Li-sites of Li2MnO3 can, however, mitigate structural transformation during the charge/discharge process by prohibiting the migration of manganese ions from the transition metal (TM) layers to lithium layers. In this work, the simulated amorphisation and recrystallisation (A+R) technique was employed during the high-temperature synthesis of Li2MnO3 nano-architectures i.e. nanosphere, nanoporous and the bulk material consisting of a large number of atoms (32148). Nanostructuring, due to the advantage in particle size reduction, has been considered, owing to the large surface exposure and shortened diffusion path. Thus facilitating effective activation of the Li2MnO3 electrode material. The nano-architectures were investigated between the temperature range of 1500 K and 1900 K as guided by the melting point, which was found to be around 1845 K, from the total energy graph. Recrystallisation of the Li2MnO3 models was confirmed by the radial distribution functions (RDFs) which depicted multiple sharp peaks due to the long-range ordering of atoms and strong bonds. Structural validation through the x-ray diffraction plots (XRD) compared very well with experimental work in literature, displaying broadening, splitting and partial shifting of peaks resulting from the mixing of Li and Mn layers. However, the nanosphere depicted broader peaks compared to both the nanoporous and bulk systems. The microstructural features that evolved during the recrystallisation of all the models contained cation mixing and vacancies resulting in structures constituting mixed layers Li-ion kinetics for the nanosphere model was found to be higher (approximately 1.54 nm2s-1) than in the nanoporous and bulk systems which were below 0.3 nm2s-1. The Li2MnO3 nanosphere model was then used to dope sodium ions into the Li-sites. Accordingly, a series of Li2-xNaxMnO3 (0≤x≤2) nanosphere models of different lithium and sodium content were generated via the amorphisation and recrystallisation strategy. The results thereof showed structural collapse upon amorphisation for systems containing a large dosage of Na-ions. However, systems with 25% Na content and below, were successfully amorphised and recrystallised at 1700 K, while retaining their structural morphology, with the Li1.75Na0.25MnO3 system resulting in a polycrystalline structure. The microstructural analysis also revealed the mixing of layers and improved structural arrangement for the Li1.95Na0.05MnO3 system, which showed a large portion of a wellordered Li2MnO3 phase. Li-ion mobility for the Na-incorporated systems was found to be higher compared to the undoped systems with the system containing the lowest Na-ions (Li1.975Na0.025MnO3) diffusing better. This proves that incorporating sodium ions into the Li2MnO3 lattice aid in improving the ionic movement. XRD analysis revealed peaks with a broad character. The Peaks belonging to 2Θ~38 slightly shift to the right as a result of the enlarged Li-layers which are occupied by Na ions. The enlargement of the Li-layers facilitates Li-ion kinetics. Again, superlattice peaks were observed for the Li1.95Na0.05MnO3 at lower angles due to the ordering of Li/Mn ions in the transitional metal layers. Charging was achieved by simultaneously removing Li+ and O- ions from the outer surfaces of the Li2MnO3 and Li2-xNaxMnO3 nanoarchitectures. All the Li/O deficient systems crystallised into multi-grained crystals with grain boundaries increasing with a decrease in Li/O content for the undoped systems. For the Na-incorporated systems, the Li1.95Na0.05MnO3 revealed minimal grain boundaries as Li/O was varied, with the Li1.70Na0.05MnO2.75 and Li1.45Na0.05MnO2.50 concentrations showing no formation of grain boundaries. Characterisation of the XRD for the charged systems also resulted in broad peaks, accompanied by the growth of peaks 2Θ~18-25⁰ peak and the 2Θ~29⁰ associated with the spinel phase. Contrarily, the nanoporous system (Li1.25MnO2.25) depicts the narrowing and decreasing of diffraction peaks at 2Θ~18-25 and 2Θ~29⁰. The Li2- xNaxMnO3 systems also depict the narrowing and decreasing of the peak at approximately 2Θ~38⁰. In addition, the complex internal microstructures of the Li/O deficient models (Li2- xMnO3-x) revealed structures governed by defects. These defects contribute to the evolution of distorted LiMn2O4, Li2MnO3 and LiMnO2 polymorphs. The spontaneous migration of manganese into the lithium layers triggers the transformation from layered to spinel-type configurations. The charging of the Li1.95Na0.05MnO3 system resulted in improved structural stability, with structures revealing a high concentration of the layered Li2MnO3 structure. Again, the diffusion has been improved by doping sodium into the lithium sites of the Li2MnO3 nanosphere upon charging. Our results accentuate the process or mechanism behind the charging of the Li2MnO3 and Li2-xNaxMnO3 constituting complex structures. This study will help guide the optimisation of high-capacity cathodes for advanced LIB technologies