Using mechanosynthesis to produce possible cathodes with a disordered rock-salt phase

The use of mechanosynthesis methods to prepare a wide range of single phase materials with a disordered rock salt structure, and their subsequent characterisation using a range of techniques, is reported. Phase studies by X-ray diffraction, thermal stability by thermogravimetry, evolved gas analysis by mass spectrometry MS and differential scanning calorimetry, chemical analysis by inductively coupled plasma MS and X-ray fluorescence, were carried out to analyse the products and confirm successful synthesis. Contamination caused by ball-milling reactants and products using three milling media, stainless steel, tungsten carbide and zirconia and choice of suitable reactants to obtain the target products, have been studied systematically and results are summarised. All three milling media introduced corresponding contamination, i.e. Fe, W and Zr. Transition metal oxides were highly contaminated (10-40 wt%) when separately ball-milled. However, the products using suitable reactant mixtures usually contained little contamination (2-3 wt%). Li2O and Li2O2 were possible Li sources to produce Li2MnO2 when milled with suitable manganese oxides. Products of mechanosynthesis showed high hygroscopicity. They may absorb water and probably CO2 during synthesis and storage. The loss and absorption of water, CO2 and oxygen was studied by thermogravimetry profiles with evolved gas analysis on heat/cool cycles. A wide range of Li-Mn-O compositions were prepared by mechanosynthesis and many of the products had a disordered rock-salt structure. Compositions with a fixed Li/Mn ratio but different oxygen stoichiometry, such as Li2MnOδ and Li4Mn5Oδ, and with a varied Li/Mn ratio, but a fixed cation/anion ratio, such as LixMn1-xO and LixMn3-xO4, were prepared and characterised. Some had a rock-salt stoichiometry and structure whereas others had spinel stoichiometry. The doping of Li2MnOδ by partially substituting O for F was studied. Four possible charge compensation mechanisms and the extent of doping for each was determined. Some F doped solid solutions which cannot be produced by high temperature synthesis were successfully produced by mechanosynthesis and their thermal stability studied. Ball-milled products were often metastable and underwent phase transformation or decomposition during heating. A wide range Li-TM-O (TM=Nb, Ti, Al, Fe, Co, V, and Ni) compositions was prepared by mechanosynthesis and obtained with disordered rock-salt structure, such as Li3NbO4, LiTiO2, Li2TiO3, LiAlO2, LiFeO2, LiNiO2, Li2NbO3 and Li2NbO2F. However, their thermal stabilities were poor. During heating, phase transformation from disordered to ordered structures often occurred. Li loss might also occur, as for example, disordered Li3NbO4 which transformed to a mixture of ordered Li3NbO4 and LiNbO3 during heating. The F doped solid solution, Li2MnO3-xFx, is attractive as a possible cathode, in which fluorine replaces oxygen, accompanied by reduction of Mn4+ to Mn3+ to form the redox couple Mn4+/Mn3+. Ball-milled Li2MnO3-xFx martials were nanosize and had an initial charge capacity of 215 mAh/g, much higher than that of undoped Li2MnO3. Most ball-milled rock-salt phases were metastable but several F doped solid solutions on the Li-TM-O-LiF join had a good thermal stability, where Li-TM-O was Li2MnO3, Li3NbO4 and Li2TiO3. No structure change or F loss from those compositions occurred up to 800 °C.