|
CPM Seminar
B. Heinrich
Physics Department In this talk I will review our recent studies on crystalline tunneling junctions. Single crystal epitaxial MgO thin films were grown directly onto high quality Fe single crystal and Fe whisker substrates and covered by Au/Fe layers. RHEED and LEED patterns and STM images showed that the growth of MgO proceeded pseudomorphically in a nearly layer by layer mode up to 6 atomic layers(AL). A misfit dislocation network is formed for MgO layers thicker than 6AL. Scanning Tunneling Spectroscopy (STS) showed that the MgO films were high quality stochiometric crystals. The tunneling barrier in MgO depends on the layer thickness. A small fraction of the scanned area showed random localized spikes in the tunneling current. Tunneling I-V curves at the defects showed a lower tunneling barrier than that in the good parts of the MgO film. However the total STM current over a large area was not dominated by defects. Defects in the MgO barrier were neither related to atomic steps in the Fe whisker substrates nor to the formation of misfit dislocations. Magnetic anisotropies and exchange coupling inFe/MgO(001) andFe/MgO/Fe(001) structures were studied using Ferromagnetic resonance (FMR) and Brillouin Light Scattering (BLS). The magnetic properties of Fe films grown on MgO(001) substrates show a large uniaxial interface anisotropy Ks=1.1 ergs/cm2 with the easy axis perpendicular to the film surface. The BLS spectra show no evidence of exchange coupling between the Fe film and the Fe-whisker substrate. The MOKE images of domain structures showed that the magnetic moment of the Fe film is coupled to the Fe whisker only through the stray fields of magnetic domain walls resulting in a wide variety of magnetization patterns. The recent theoretical calculations of intrinsic tunneling electron transport provide challenging predictions. The tunneling conductance in Fe/MgO/Fe(001) is governed by the majority spin electrons. The tunneling resistance for the parallel orientation of the magnetic moments is estimated to be 100 times larger than that for the antiparallel orientation of the magnetic moments. The I-V characteristics of crystalline Fe/MgO/Fe(001) structures are dominated by the intrinsic properties allowing, in principle, to test the wealth of predictions based on first principle calculations. Our present STM and AFM systems have not allow us to investigate magnetoresistance. The purpose of my visit is to establish with the group of Peter Gruetter a collaborative effort to investigate the magnetoresistance using their AFM facilities.
Tuesday, December 12th 2000, 15:00 |