The transmission electron microscope (TEM) is one of the most useful tools available to the materials scientist. Yet both the complexity and expense of the equipment, and the huge investment in time necessary to become proficient in specimen preparation and image acquisition and analysis, mean that it is difficult for most industrial institutions to maintain a state-of-the-art TEM facility. How can industry overcome this problem? One solution is to set up a collaboration with a university, an industrial partner, or a government research laboratory. Such collaborations can be extremely valuable to the company, which gains access to microscopes, specimen-preparation equipment and the expertise of professional microscopists, and to the research laboratory, which benefits from the industrial perspective and the private sector's proficiency in materials preparation and processing.

Such collaborations exist, and they can produce excellent results. In this article, we present three case studies in which successful collaboration has occurred between industry and one of the Department of Energy's scientific user facilities, the National Center for Electron Microscopy (NCEM-see sidebar). Our aim is not only to describe results that we hope will be of scientific interest but also to encourage industrial researchers to consider collaborations with institutes such as NCEM.

Giant Magnetoresistance, GMR, in thin metal films elicits attention due to its technological potential as well as its relevance to theory of exchange coupling. Epitaxial, phase-segregated ferromagnet/paramagnet Mixtures have been grown by UHV evaporation. Such films show spontaneous formation of ferromagnetic clusters, leading to large values of GMR (40% at room temperature) as grown. The growth of Co-Cu, Co-Ag, Fe-Ag and Permalloy-Ag films are described. Structural analysis by grazing-incidence small angle X-ray scattering (GISAXS) provides a measure of cluster size and characteristic spacing. Effects of growth temperature and subsequent annealing on GMR and film structure are described. Preliminary results of TEM examination of (001) Fe-Ag and Co-Ag granular films are presented for the first time.

The Magneto-optical Kerr effect of CoPt3 ordered compounds and alloys is investigated experimentally and analyzed theoretically using first principles band structure calculations.

The calculations confirm the experimental results and allow for the first time a detailed, bandstructure based interpretation of the electronic transitions leading to the well known strong uv peak at around 4eV. This feature can be unambiguously attributed to Magneto-optical transitions within the Pt p and d bands.

A quantitative determination of interface (Ks) and volume anisotropy {?ψ) constants of MBE and sputtered CO/Pt Multilayers is reported. Torque and VSM Magnetometry were used to determine the total average anisotropy and the room temperature magnetization of four different series of films with varying Co thickness and nearly constant Pt thickness. All films were characterized with X-ray diffraction and X-ray fluorescence, allowing the determination of the “Magnetic” volume with good accuracy. Both Ks and Jeff are found to be orientation dependent. We find the following results for MBE films grown on Ag buffered GaAs substrates and highly < 111 > textured films, grown on etched SiNx buffers:

(111) Ks = 0.97mJ/m2, Kveff =-0.74MJ/m3 MBE

(111) Ks = 0.92mJ/m2, Kveff =-l.lIMJ/m3 sputtered

(110) Ks = 0.42mJ/m2, Kveff =-l.95MJ/m3 MBE

(001) Ks = 0.59mJ/m2, Kveff =-5.98MJ/m3 MBE

The [110]-oriented MBE films show in addition a large (intrinsic) in-plane anisotropy constant K‖0≃-3MJ/m3 which is found to be independent of the Co thickness. [100] is the easy and [110] the hard in-plane direction.

We have performed x-ray diffraction experiments on MBE grown 〈111〉 and 〈001〉 oriented Co (3 Å)/Pt (18 Å) superlattices with the scattering vector Q covering a plane that includes a major component perpendicular and a minor component parallel to the plane of the sample. The superlattice diffraction peaks were found to have a large width parallel to the plane of the film, indicating that there is strong in-plane disorder. These broad peaks were integrated to evaluate more accurately the amount of interdiffusion at the interface, which was found to be similar in both films, contrary to what the conventional x-ray diffraction experiment suggests.

We report results of a synchrotron X-ray diffraction study of Co-Pt multilayers, with Pt[111] as the growth axis. The multilayers were grown by MBE on Pt seedfilms on basal-plane sapphire substrates at growth temperatures of 100 and 200ºC.The multilayers comprised 15 periods of Co 3Å, Pt 15Å and both exhibited perpendicular anisotropy. Grazing-incidence radial- and phi-scans revealed a set of {110{ peaks indicating doubling of real space periodicity consistent with the presence of the ordered alloy CoPt3. This supports the results of other studies which indicate interdiffused interfaces in Co-Pt multilayers and shows that chemical ordering can occur at much lower temperatures than in bulk alloys where annealing times of several days at ∼ 700ºC are required to stimulate ordering.

Magnetic superlattices are of great current interest due to the improvement of crystal growth techniques for their preparation and their potential for practical applications in high density magnetic recording. It is apparent that a strong correlation exists between structural and magnetic properties of these supcrlattices. Careful sample preparation and structural characterization arc necessary in order to tailor their magnetic properties. Molecular Beam Epitaxy (MBE) provides the capability for controlled crystal growth and in-situ strtuctoral, chemical and physical characterization. Ex-situ techniques such as X-ray diffraction and high-resolution TEM, also provide insight into interfaces in these structures. Co layers sandwiched with Au, Cu and Pt layers of various thickness are discussed in this paper.

The magnetic properties of epitaxial magnetic metal multilaycrs, superlatticcs and ultrathin films are of intense current interest because of the novel and potentially useful properties which these structures exhibit. MBE techniques permit the growth and in-situ characterization of such structures and the control of the growth axis of films and multilayers through epitaxial seed layers on semiconductor substrates. Techniques for controlling the growth axis of Co/Cu superlattices, for magnetoresistance studies, are described. In this and other other material systems it is necessary to combine the in situ probes of RHEED, LEED and X-ray photoelectron diffraction (XPD) with post-growth HRXTEM to understand and control the formation and nature of the interfaces in the structures.

Selected orientations of epitaxial Co/Pt superlattices were grown on GaAs substrates in [001], [110], and [111] orientations using seeded epitaxy techniques. The microstructural and magnetic properties of Co/Pt superlattices are presented. The magnetic anisotropies of Co/Pt superlattices are found strongly dependent on the crystallographic orientation of the growth axis. High-resolution cross-sectional transmission electron microscopy (HRXTEM) investigation of the Co/Pt interfaces has revealed the presence of a large number of planar defects in the [111] Co/Pt superlattice. HRXTEM shows that the interface of Co/Pt is wavy and is not atomically abrupt. These findings suggest that models for the magnetic anisotropy should not be based on idealized, chemically abrupt interfaces but should incorporate magnetocrystalline anisotropy in the alloy region. We propose a model for the magnetic anisotropy based on the N6el magnetic surface anisotropy and the existence of defects in the alloy region. The easy axis of the magnetic anisotropy predicted by the model agrees with the magnetic and structural data for the superlattices.

In this review, recent developments in MBE growth of artificially-layered magnetic metal structures on semiconductor substrates are considered. It is seen that the ability to seed specific orientations of such structures and to probe their structural and chemical properties with a variety of in situ probes are key advantages of MBE over more conventional deposition techniques.

We report the first HRXTEM study of epitaxial rare earth fluoride/ GaAs(111) interfaces. Such interfaces are of interest because they are the starting point for growth of buried epitaxial rare earth / rare earth fluoride sandwich structures which exhibit interesting and non bulk-like magnetic properties. Also, the optical transitions in ultrathin epitaxial NdF3 films may be influenced by strain and defects in the NdF3 film and the nature of the interface to GaAs.

We find that the rare earth fluoride / GaAs interfaces are semi- coherent but chemically abrupt with the transition taking place within 3Å. However, the interface is physically rough and multiple monolayer steps in the GaAs surface tend to lead to tilt boundaries in the fluoride. The origin of these steps is believed to be thermal etching of the GaAs during the heat-cleaning stage prior to epitaxy. The surface of the fluoride film is much smoother than the initial GaAs surface indicating planarization during epitaxy.

A series of MBE grown CdTe films were grown on (100) InSb substrates. The substrate temperature, Ts, was varied from 170° to 285°C in eleven steps. Low temperature (∿2K) photoluminescence measurements and TEM have been combined to show a strong correlation between defect density and the details of the luminescence spectra. A natural division is obtained for samples grown with substrate temperatures from 285°C to 250°C (Region I) and from 225°C to 170°C (Region II).