Crystal structure and oxygen stoichiometry in LiMn 1.5 Ni 0.5 O 4−␦ , a potential lithium-battery cathode, vary with temperature, as observed in samples quenched from different temperatures and by in situ diffraction and thermogravimetry techniques. When prepared in high O 2 pressure, this cation-ordered spinel is oxygen-stoichiometric, ␦ = 0, space group P4 3 32. Upon heating between 650 and 680°C, increasing oxygen deficiency occurs exclusively in MnO 6 octahedra and Mn-O-Mn bonds, which induces a volume increase of the 12d octahedra, a reduction of Mn as shown by X-ray absorption near-edge structure, equalization of Mn-O and Ni-O bond lengths, and disordering of Mn, Ni on octahedral sites. Hence, the transformation to space group Fd3 ¯m, shown by Rietveld refinement of variable-temperature neutron diffraction data, is a direct consequence of oxygen loss from the structure. On further oxygen loss, a second phase transformation occurs to give a cation-deficient cubic rock salt phase, ␦ ϳ 0.65, at 950°C, which loses more oxygen at higher temperatures until, at 1100°C, the material is essentially a stoichiometric, single-phase cation-disordered rock salt, space group Fm3 ¯m. A second spinel phase persists in small amounts from 950 to 1100°C. Differences in electrochemical behavior depend on sample preparation and correlate with the oxygen content of LiMn 1.5 Ni 0.5 O 4−␦ when used as a cathode in Li test cells.

No abstract

Li 2 MnO 3 loses oxygen, reversibly, on heating above ∼600 °C. By 1100 °C, 1% of the oxygen has been lost, giving a stoichiometry of Li 2 MnO 2.97 . Depending on the synthesis conditions, materials with a range of oxygen contents may be produced. Associated with the variable oxygen content, partial reduction of Mn 4+ to Mn 3+ occurs, as demonstrated by a direct correlation between electrochemical capacity, when used as a cathode in a rechargeable Li test cell, and oxygen content.

A study of the lithium ion conductor Li(3x)La(2/3-x)TiO(3) solid solution and the surrounding composition space was carried out using a high throughput physical vapor deposition system. An optimum total ionic conductivity value of 5.45 × 10(-4) S cm(-1) was obtained for the composition Li(0.17)La(0.29)Ti(0.54) (Li(3x)La(2/3-x)TiO(3)x = 0.11). This optimum value was calculated using an artificial neural network model based on the empirical data. Due to the large scale of the data set produced and the complexity of synthesis, informatics tools were required to analyze the data. Partition analysis was carried out to determine the synthetic parameters of importance and their threshold values. Multivariate curve resolution and principal component analysis were applied to the diffraction data set. This analysis enabled the construction of phase distribution diagrams, illustrating both the phases obtained and the compositional zones in which they occur. The synthetic technique presented has significant advantages over other thin film and bulk methodologies, in terms of both the compositional range covered and the nature of the materials produced.

LiCoMnO 42d is one of the materials first reported as a 5 V cathode in prototype lithium batteries, but as-prepared material has a residual 4 V plateau in addition to the main 5 V plateau. A combination of thermogravimetry on samples prepared under various conditions, high temperature X-ray powder diffraction and electrochemical testing, shows the 4 V plateau to depend on the extent of oxygen deficiency, d, and to, therefore, correlate with the Mn 3+ content. A complex reaction sequence occurs upon oxygen loss, including formation of oxygen-deficient spinel, precipitation of Li 2 MnO 3 and change in cation composition of the spinel but remarkably, these reactions show a high degree of reversibility upon oxygen uptake.