Novel cathode and anode materials for rechargeable lithium-ion batteries.

A number of compositions from the Li2O-CoOx-MnOy ternary phase diagram have been studied. Samples have been studied by X-ray powder diffraction. Property characterisation has been conducted using a variety of techniques (Chapter 2): thermal analysis has been carried out using thermogravimetry (TG), differential thermal analysis (DT A) and differential scanning calorimetry (DSC), electrical measurements have been made using ac impedance, battery testing has been carried out using galvanostatic cycling with potential limitation (GCPL).
The pseudo-binary join between cobalt oxide and manganese oxide has been studied (Chapter 3); the phase diagram has been established from room temperature to - 1400 0c. It was found that single phase samples with the spinel structure (with either cubic (0.00≤x:≤1.30) or tetragonal (for 1.30 < x≤3.00) symmetry could be isolated. These phases were studied using GCPL (Chapter 4) for possible application as anodes in Liion batteries, and were found to be able to reversibly accept Li by a reductionlreoxidation process. On first discharge, the spinel was fully reduced to a discrete mixture of Co and Mn metal particles; these could be re-oxidised to CoO and MnO on subsequent charging. The best performance found was for Co0.4Mn2.6O4, which showed good capacity retention, better specific capacity than the graphite anode materials currently used, and the lowest potential vs. Li metal, while also replacing most of the expensive and toxic Co with Mn.
The ternary Li2O-CoOx-MnOy phase diagram has also been studied (Chapter 5), with particular attention paid to three solid solutions:
(i) The reported cubic spinel solid solution between LiMn2O4 and LiCoMnO4. (i.e. LiCoxMn2-xO4).
(ii) A previously unreported join between LiCoMnO4 and Co3O4. (i.e. Li1-xCo1+2xMn1-xO4)
(iii) A previously unreported join between Li2CoMn3O8 and Co1.7Mn1.3O4. (i.e. Li1-xCo1/2+6x/5Mn1.5-x/4O4).
It is most likely that these joins are, in fact, three sections through a solid solution area. It was noted that upon heating these cubic spinels, oxygen volatilization occurred. For a limited range of cubic spinel compositions, the equivalent of - 1 oxygen atom could be lost from the structure, resulting in the formation of a high temperature rock salt solid solution area.
Battery testing (Chapter 6) was conducted on phases from two of these cubic spine I solid solutions (Li1-xCo1+2xMn1-xO4 and Li1-xCo1/2+6x/5Mn3/2-x/5O4). LiCoMnO4 has the greatest potential applicability as a cathode material for the next generation of Li-ion batteries. Two plateaux are observed in the electrochemical data; a short plateau at 4.0 V (possibly due to a Mn3+ ↔Mn4+ redox couple brought about by a small degree of oxygen non-stoichiometry) and at - 5.0 V (most likely due to the C03 + ↔ C04 + redox couple). It has also been found that phases with compositions on the Li1-xCo1+2xMn1-xO4 and Li1-xCo1/2+6x/5Mn3/2-x/5O4 solid solutions performed well as anode materials. The best anode performance was observed for x = 0.51 on the Li1-xCo1+2xMn1-xO4 solid solution, where specific capacities of - 750 mAh g-1 were obtained. This is acceptable for use as an anode material.
The low temperature phase diagram of LiCoMnO4 was investigated by DSC (Chapter 7); two low temperature phase transitions were observed. An 'α→β’ transition, at -23°C on heating, has been widely reported, though the symmetry changes involved are not yet satisfactorily understood. A second transition(‘β→y’) occurred at - -140°C on cooling, and - 15°C on heating. This second transition could be stabilised over the Li1+xMn2-xO4 solid solution for compositions with 0.04≤ ᵡ≤0.12 by introducing oxygen vacancies such that the average Mn valence fell to ≤3.5+. This second transition has not been clearly identified in the literature, and has not been studied by DSC previously.