Atomic-scale engineering of magnetic anisotropy of nanostructures through interfaces and interlines.

The central goals of nanoscale magnetic materials science are the self-assembly of the smallest structure exhibiting ferromagnetic hysteresis at room temperature, and the assembly of these structures into the highest density patterns. The focus has been on chemically ordered alloys combining magnetic 3d elements with polarizable 5d elements having high spin-orbit coupling and thus yielding the desired large magneto-crystalline anisotropy. The chemical synthesis of nanoparticles of these alloys yields disordered phases requiring annealing to transform them to the high-anisotropy L1(0) structure. Despite considerable efforts, so far only part of the nanoparticles can be transformed without coalescence. Here we present an alternative approach to homogeneous alloys, namely the creation of nanostructures with atomically sharp bimetallic interfaces and interlines. They exhibit unexpectedly high magnetization reversal energy with values and directions of the easy magnetization axes strongly depending on chemistry and texture. We find significant deviations from the expected behaviour for commonly used element combinations. Ab-initio calculations reproduce these results and unravel their origin.

High resolution in situ magneto-optic Kerr effect and scanning tunneling microscopy setup with all optical components in UHV.

A surface magneto-optic Kerr effect (MOKE) setup fully integrated in an ultrahigh vacuum chamber is presented. The system has been designed to combine in situ MOKE and scanning tunneling microscopy. Magnetic fields up to 0.3 T can be applied at any angle in the transverse plane allowing the study of in-plane and out-of-plane magnetization. The setup performance is demonstrated for a continuous film of 0.9 monolayers (ML) Co/Rh(111) with in-plane easy axis and for a superlattice of nanometric double layer Co islands on Au(11,12,12) with out-of-plane easy axis. For Co/Au(11,12,12) we demonstrate that the magnetic anisotropy energies deduced from thermally induced magnetization reversal and from applying a torque onto the magnetization by turning the field are the same. For the presented setup we establish a coverage detection limit of 0.5 ML for transverse and 0.1 ML for polar MOKE. For island superlattices with the density of Co/Au(11,12,12), the latter limit corresponds to islands composed of about 50 atoms. The detection limit can be further reduced when optimizing the MOKE setup for either one of the two Kerr configurations.

The remarkable difference between surface and step atoms in the magnetic anisotropy of two-dimensional nanostructures.

The original magnetic properties of nanometre-sized particles are due to the distinct contributions of volume, surface and step atoms. To disentangle these contributions is an ongoing challenge of materials science. Here we introduce a method enabling the identification of the remarkably different contributions of surface and perimeter atoms to the magnetic anisotropy energy of two-dimensional nanostructures. Our method uses the generally nonlinear relationship between perimeter length and surface area. Atomic-scale characterization of the morphology of ensembles of polydisperse nanostructures, combined with in situ measurements of their temperature-dependent magnetic susceptibility, gives access to the role played by the differently coordinated atoms. We show for Co nanostructures on a Pt(111) surface that their uniaxial out-of-plane magnetization is entirely caused by edge atoms having 20 times more anisotropy energy than their bulk and surface counterparts. Identification of the role of perimeter and surface atoms opens up unprecedented opportunities for materials engineering. As an example, we separately tune magnetic hardness and moment in bimetallic core-shell nanostructures.