Half metal oxides: In search for 100% spin polarised materials

点击翻译按钮获取中文摘要

Spintronics is a rapidly developing field that utilises the electron's spin in addition to its charge to create new devices combining logic, data storage and sensor applications. The huge potential of spintronics has stimulated a wide range of research from spin transport, spin injection/accumulation and spin manipulation to device fabrication such as spin valves and magnetic tunnel junctions. One of the main challenges in the spintronics field is to find/create highly spin polarised materials that are compatible (lattice match, conductivity match, thermodynamically stable, high Curie temperature (Tc), etc.) with CMOS technology. In this proposal magnetite (Fe3O4) is proposed as the optimum highly spin polarised material for spintronics and by understanding the material at the atomic level seeks to solve the challenges in its implementation.Conventional 3d ferromagnetic metals and their alloys are only 30-40% spin polarised at the Fermi level, thus material systems with better spin polarisation are essential for the next generation of spintronic devices. The existence of 100% spin polarised materials at the Fermi level has been predicted by density functional theory (DFT). Such 100% spin polarised materials, also termed half-metals, have one of the spin channels metallic while the other spin channel is insulating. A rather large number of materials including oxides (Fe3O4, CrO2, manganites), pnictides, chalcogenides, and Heusler alloys have been predicted to be half-metallic. Among these materials magnetite (Tc=855 K) is of special interest since: (i) it has a Tc in excess of 500K, the threshold temperature for device applications; (ii) it has an excellent lattice match with MgO and MgAl2O4, the two most important oxides for spintronic applications; (iii) it can form atomically sharp interfaces with relevant semiconductors (SC) such as GaAs, GaN and SiC; and (iv) its layered structure will allow interface atomic engineering at magnetite/oxide and, magnetite/SC heterojunctions. It is worth noting that no other SP materials have these properties. For example, CrO2 has Tc below 500K and Heusler/SC junctions are not abrupt due to the high annealing temperature required for Heuslers to fully order into a L21 structure that is half-metallic. In order to incorporate magnetite in device structure, growth of thin films of magnetite and heterostructures of magnetite with suitable oxides, metals and semiconductors (SC) is required.The two main challenges to overcome for successful application of Fe3O4 are: 1) growth of thin films with control of stoichiometry and structural defects; it is well known that defects such as antiphase domain boundaries (APBs) can completely determine the functionality of magnetite films, thus controlling the APBs nature and density is highly important;2) engineering the interfaces between magnetite/oxide barriers and magnetite/SC; spin transport across interfaces critically depends on the interfaces' atomic structure.These are the two crucial steps to understand the structural basis of spin-related phenomena in the magnetite films as well as some of technologically important magnetite/oxide and magnetite/SC interfaces. This knowledge would provide a path and guide for the engineering of spintronic devices based on Fe3O4. In order to achieve this goal, the direct correlation of the films' functionality and their atomic structure, in this application I propose a joint experimental and theoretical study on the growth of half-metal magnetite oxide films and the atomic and electronic structure of the film, APBs and magnetite/oxide and magnetite/SC interfaces which are of interest for spintronic devices. Film growth will be done by Molecular Beam Epitaxy, spin polarised calculations will be performed by DFT, and High Resolution Transmission Electron Microscopy, High Angle Annular Dark Field Imaging and Electron Energy Loss Spectroscopy will be used to fully characterise these systems on atomic scale.