A new magnetic field sensor is presented, based on perpendicular hot electron transport in a giant magnetoresistance (Co/Cu)4 multilayer, which serves as a base region of an n-silicon metal-base transistor structure.A 215% change in collector current is found in 500 Oe (77 K), with typical characteristics of the spin-valve effect. The in-plane magnetoresistance was only 3%. The transistor structure allows the investigation of energy resolved perpendicular transport properties, and in particular spin-dependent scattering of hot electrons in transition-metal as well as rare-earth-based multilayers.PACS numbers: 72. 15. Gd, 73.40.Vz, 75.50.Rr, 85.70.Kh The discovery of giant magnetoresistance in magnetic multilayers [1] (also called the spin-valve effect [2]) has led to a large number of studies on giant magnetoresistance systems. Usually, the resistance of the multilayer is measured with the current in plane (CIP). This is the easiest experimental approach of electrical transport in magnetic multilayers. Devices exhibiting CIP giant magnetoresistance are under development as magnetic field sensors, for instance, in read-back magnetic heads used in magnetic recording technology. However, from a fundamental point of view, the CIP configuration suffers from several drawbacks; the CIP magnetoresistance (MR) is diminished by shunting and channeling [2,3].In particular, uncoupled multilayers or sandwiches with thick spacer layers suffer from this problem, whereas the saturation field in such systems is usually small. Moreover, diffusive surface scattering reduces the MR for sandwiches [2] and thin multilayers [4]. Finally, fundamental parameters of the effect, such as the relative contributions of interface and bulk spin-dependent scatterings, are difficult to obtain using the CIP geometry [5]. Measuring with the current perpendicular to the planes (CPP) solves most of these problems, mainly because the electrons cross all magnetic layers, but a practical difficulty is encountered; the perpendicular resistance of the ultrathin multilayers is too small to be measured by ordinary techniques. The first CPP-MR experiments were reported on Co/Ag multilayers [6], where the multilayer was sandwiched between superconducting Nb leads. In this way, CPP experiments could be performed, albeit only at liquid helium temperatures. The use of microfabrication techniques for CPP measurements from 4.2 to 300 K was first shown for Fe/Cr multilayers [7], where the multilayers were etched into micropillars to obtain a relatively large resistance (a few mA). Both types of measurements have confirmed the larger MR effect for the CPP configuration, but they suffered from the general complexity of realization and measurement techniques. Experiments using electrodeposited nanowires showed CPP-MR up to 15% at RT [8].In this Letter, we present the design, prospects, and experimental results of a new magnetic field sensor and measurement tool based on perpendicular hot electron transport in a spin-valve multilayer: the spin-valve transistor. Here...

Spin reorientation has been observed in CoFe 2 O 4 thin single crystalline films epitaxially grown on ͑100͒ MgO substrate upon varying the film thickness. The critical thickness for such a spin-reorientation transition was estimated to be 300 nm. The reorientation is driven by a structural transition in the film from a tetragonal to cubic symmetry. At low thickness, the in-plane tensile stress induces a tetragonal distortion of the lattice that generates a perpendicular anisotropy, large enough to overcome the shape anisotropy and to stabilize the magnetization easy axis out of plane. However, in thicker films, the lattice relaxation toward the cubic structure of the bulk allows the shape anisotropy to force the magnetization to be in plane aligned. The importance of magnetic anisotropy is well recognized in many technical applications such as magnetic and magneto-optic recording. The large interest for high anisotropies is motivated by technological demands such as increasing the magnetic recording density. With large anisotropy, the superparamagnetic limit can be pushed down, and a stable magnetization can be promoted in ultrasmall nanosized magnetic structures, which are needed in advanced media for ultrahigh density recording. Besides the intrinsic anisotropy of the bulk, other sources of anisotropy may be enhanced in artificial structures and contribute to their magnetic properties. Depending on their relative orientations and magnitudes, the involved anisotropies may compete between each other, leading to spin-reorientation phenomena in the system. For example, the broken symmetry at the interfaces in ultrathin films generates a perpendicular anisotropy, which overcomes the shape anisotropy.1 However, increasing the layer thickness reduces the ratio between the surface and the volume atoms, leading to an in-plane alignment of the easy axis.2 In obliquely sputtered metallic thin films, we established the existence of an in-plane reorientation of magnetic anisotropy.3 Depending on the film thickness and due to the shadow effect during the growth, the layer can develop columns or nuclei able to confine the anisotropy parallel or perpendicular to the longitudinal direction ͑projection of the incident beam in the film plane͒.Ferrites cover a large family of oxides, including soft as well as hard magnetic materials. Hard ferrites such as the hexagonal ͑BaFe 12 O 19 ͒ and the spinel ͑CoFe 2 O 4 ͒ are particularly attractive for magnetic and magneto-optic recording applications due to their large magnetocrystalline anisotropy and high chemical stability. Recent studies demonstrated that integrating cobalt ferrite as a pinning layer in the spin valve architecture can strongly enhance the magnetoresistance effect of the sandwiched structure. 4 In epitaxial hexaferrite thin films, the uniaxial magnetocrystalline anisotropy is strong enough to dominate all the other sources of anisotropy and to keep the spin alignment constant regardless the film thickness and the preparation conditions. 5 However, cobalt ferrite rep...

We present a transfer-function approach to calculate the force on a magnetic force microscope tip and the stray field due to a perpendicularly magnetized medium having an arbitrary magnetization pattern. Under certain conditions, it is possible to calculate the magnetization pattern from the measured force data. We apply this transfer function theory to quantitatively simulate magnetic force microscopy data acquired on a CoNi/Pt multilayer and on an epitaxially grown Cu/Ni/Cu/Si͑001͒ magnetic thin film. The method described here serves as an excellent basis for ͑i͒ the definition of the condition for achieving maximum resolution in a specific experiment, ͑ii͒ the differences of force and force z-derivative imaging, ͑iii͒ the artificial distinction between domain and domain wall contrast, and finally ͑iv͒ the influence of various tip shapes on image content.

Functional integration between semiconductors and ferromagnets was demonstrated with the spin-valve transistor. A ferromagnetic multilayer was sandwiched between two device-quality silicon substrates by means of vacuum bonding. The emitter Schottky barrier injected hot electrons into the spin-valve base. The collector Schottky barrier accepts only ballistic electrons, which makes the collector current very sensitive to magnetic fields. Room temperature operation was accomplished by preparing Si-Pt-Co-Cu-Co-Si devices. The vacuum bonding technique allows the realization of many ideas for vertical transport devices and forms a permanent link that is useful in demanding adhesion applications.Ten years after its discovery (1), giant magnetoresistance (GMR) or, designated more appropriately, the spin-valve effect, has already shown its strength in applications such as read heads and magnetic random access memories (MRAMs). Driven by such lowfield applications, a search for higher sensitivities is continuing. In the spin-valve effect, majority carrier electrons with long mean free paths can travel with low resistance through a multilayer when a magnetic field aligns magnetizations of adjacent magnetic layers. In ordinary four-point resistance measurements with the current in plane (CIP), channeling, shunting, and diffusive surface scattering diminish and complicate the effect. Experiments with currents perpendicular to the planes (CPP) are very useful for fundamental studies of the electron transport process (2), yet application of the larger effect to sensors is cumbersome because of the very low resistances involved. We introduced the solidstate spin-valve transistor (SVT) structure as a spectroscopic tool to investigate transport properties of the CPP spin-valve effect and found a large, perpendicular, hot-electron spin-valve effect (3). In addition, we proposed to employ nonmagnetic and ferromagnetic tunnel barriers, of which experimental results were reported later (4). Such experiments had to be conducted at 77 K to decrease the collector leakage current with respect to the hot-electron current. Room-temperature (RT) operation requires a large emitter current density, reduction of the collector barrier area, increase of the collector barrier height, and enhancement of the base transport factor. In order to obtain sufficiently large injection current densities, a device-quality semiconductor layer would need to be grown on top of the metallic base. In practice, such layers grow poorly, so we instead turned to bonding the base between two crystalline substrates in air. This provided spontaneous adhesion based on van der Waals forces and hydrogen bonds between adsorbed water molecules. This bond was not strong enough, however, to allow lithographic processing. Our development of vacuum metal bonding during sputtering provided a superior bond and hot-electron transport in a nonmagnetic Si-Au-Ge metal base transistor (5). We have used this technique to prepare SVTs operating at RT. Because the RT operation now also p...

The magnetic anisotropy of epitaxial La 0.67 Sr 0.33 MnO 3 ͑LSMO͒ thin films on vicinal, TiO 2-terminated SrTiO 3 substrates is investigated. Atomic force microscopy shows a regular step-terrace structure on the LSMO surface which is a replication of the surface of the substrate. The films show in-plane uniaxial magnetic anisotropy at room temperature, with the easy axis along the step direction. At low temperature the films show biaxial crystalline anisotropy with easy axes along ͓110͔, and hard axes along the ͓100͔ direction of LSMO.