The magnetic moments in $\mathrm{Ni}/\mathrm{Pt}$ multilayers are thoroughly studied by combining experimental and ab initio theoretical techniques. SQUID magnetometry probes the samples' magnetizations. X-ray magnetic circular dichroism separates the contribution of Ni and Pt and provides a layer-resolved magnetic moment profile for the whole system. The results are compared to band-structure calculations. Induced Pt magnetic moments localized mostly at the interface are revealed. No magnetically “dead” Ni layers are found. The magnetization per Ni volume is slightly enhanced compared to bulk NiPt alloys.

X-ray magnetic circular dichroism measurements are reported at the beginning (W) and at the end (Ir, Pt) of the $5d$ series of the periodic table. Considerable induced magnetic moments of about $0.2{\mu }_{\mathrm{B}}/\mathrm{atom}$ were probed for the nonmagnetic W and Ir and compared to previous data for the Pt induced moments in multilayers. W was found to couple antiferromagnetically and Ir ferromagnetically to the $3d$ layers. Finally, the W spin and orbital magnetic moment couple in parallel, contrary to what is expected from the third Hund's rule. This remarkable finding shows that the induced magnetic behavior of $5d$ layers may be radically different than that of impurities and alloys.

We report on a detailed structural study of ultrathin Ni films on Cu(100) applying quantitative low-energy electron diffraction. The analysis of films with different thickness shows that the film growth is pseudomorphic throughout the coverage regime investigated, 1–11 monolayers. The vertical spacings of the Ni layers, which are unreconstructed, are largely independent of both film thickness and depth. They are by about 3.2% contracted compared to the bulk spacings of Ni (5% compared to the bulk spacings of Cu), so that all films correspond to homogeneous tetragonal phases. Due to the tetragonal distortion, the volume of Ni atoms in the films is practically the same as in bulk Ni. The layer spacing at the Ni-Cu interface is slightly contracted compared to the value resulting from a hard sphere model, and the same holds for the top Cu layer distance. In particular, within the accuracy limits (1–2% or less) we detect no structural differences for the change from in-plane to vertical magnetization with varying coverage or temperature or for the change from ferro- to paramagnetism at the Curie temperature.

In the last decade enormous effort has been made in research and investment to study magnetic properties of thin films because of their obvious practical applications. Coincidentally, it happens that theory has made enormous progress. Ab initio calculations and microscopic theories allow us for the first time in the history of magnetism to study and manipulate the magnetism on an atomic scale. In contrast to bulk magnetic materials, ultrathin films allow us to manipulate magnetism via the thickness and, by use of artificial structure growth, to produce structures which do not appear in nature. Here we discuss the fundamental magnetic observables, i.e. magnetization, Curie temperature, magnetic moment per atom, susceptibility and magnetic anisotropy, for idealized prototype thin films like Fe, Co, Ni on metal substrates such as Cu, W, Re. Finally, we present studies on trilayers, i.e. magnetic thin films separated by a spacer, like Cu. These trilayers present prototypes of interlayer coupling relevant for practical use of multilayer structures.

An unusual change of direction of the magnetization $\mathrm{}\left(M\right)\mathrm{}$ in Ni/Cu(001) monolayers from in-plane at low temperature to perpendicular at high temperature is measured by in situ ferromagnetic resonance and the magneto-optic Kerr effect. Two boundaries for the appearance of the perpendicular component and for the vanishing of the parallel component of the magnetization are determined as a function of temperature and film thickness. In between, stable intermediate directions of $M$ are measured. The reorientation occurs with a continuous rotation of $M,$ similar to a second-order phase transition. It is shown that the temperature-dependent volume contribution to the magnetic anisotropy must be included in the interpretation of the spin reorientation transition.