Magnetic and Electronic Phenomena in Oxide Interfaces, Thin Films and Heterostructures

Abstract

The miniaturisation and reduction of power consumption in modern integrated circuitry and active compo- nents such as sensors and memory demands the development of novel materials. These materials must not only have adaptable physical properties, such as tunable magnetic and electronic properties, but also good coupling with other materials for the integration of heterostructures which leverage these new materials in concert, as well as new phenomena arising at the interfaces. This thesis addresses the growth and characterisation of a number of technologically relevant oxides, with a view to addressing the major causes preventing them from being adapted for use in functional devices. Films of conductive ferromagnetic and insulating ferrimagnetic, antiferromagnetic, and ferroelectric oxides are grown by pulsed laser deposition (PLD) and characterised by structural, electrical and magnetic measurements. The magnetic and electronic dead layer thicknesses in La0:7Sr0:3MnO3 (LSMO) thin films on different sub- strates are deduced from SQUID magnetometry and resistance measurements as a function of film thickness, respectively. The magnetic dead layers of LSMO films on La0:26Sr0:76Al0:61Ta0:37O3 (LSAT), SrTiO3 and LaAlO3 are found to be 0.8, 1.5 and 3:0nm respectively, and the corresponding electrical dead layers are 3, 4 and 6nm. The nature of the dead layers was investigated by X-ray dichroism measurements. They are found to have anti- ferromagnetic character with no apparent strain induced orbital order, which supports the theory of electronic phase separation. The exchange constants of NiO thin films are determined by mapping the spin-wave dispersions in three dimensions by resonant inelastic X-ray scattering (RIXS) measurements. The nearest and next nearest neighbour exchange constants are determined by fitting the dispersions using linear spin-wave theory. They are J J 1meV and J0 = 18:1meV respectively and are responsible for the ferro- and antiferromagnetic order in NiO. The growth of NiFe2O4 and CoFe2O4 is achieved, and the strain effects of a number of different substrates are determined in the case of CoFe2O4. Optimal conditions which permit the growth of NiFe2O4 on SrRuO3-buffered SrTiO3, for the purpose of constructing a spin filtering junction, also lead to good magnetic characteristics which are shown by SQUID magnetometry. Highly anisotropic CoFe2O4 films are demonstrated to be extremely sensitive to substrate induced strains. In-plane anisotropy with anisotropy fields of up to 14T are obtained on highly strained MgAl2O4 substrates, and similar large anisotropies showing bulk-like magnetisations are seen in films deposited on MgO substrates. The results of investigations into the properties of PLD grown ferroelectric BaTiO3 show that it can be reliably polarised in thin film form down to a thickness of 9nm. EFM and DPFM have been employed to show unambiguous electrostatic interaction in the former and piezoelectric response in the latter. The results of this experiment are encouraging, and provide a basis for experiments on the ferroelectric field effect on magnetic oxides. Devices based on BaTiO3 patterned to form artificially multiferroic tunnel junctions have shown properties that closely match those of theoretical antisymmetric ferroelectric tunnel juntions. In conclusion, it is believed that a ferroelectric tunnel junction has indeed been realised during this work and that it only remains to be shown that there is coupling between the magnetic and ferroelectric properties to realise a full artificial multiferroic tunnel junction.