The composition, strain and surface morphology of (0001 )InGaN layers are investigated as a function of growth temperature (460-645 °C) and impinging In flux. Three different growth regimes: nitrogen-rich, metal-rich and intermediate metal-rich, are clearly identified and found to be in correlation with surface morphology and strain relaxation. Best epilayers' quality is obtained when growing under intermediate metal-rich conditions, with 1 -2 monolayers thick In ad-coverage. For a given In flux, the In incorporation decreases with increasing growth temperature due to InN thermal decomposition that follows an Arrhenius behavior with 1.84 + 0.12 eV activation energy.

The magneto-optical properties of Co microsquare-2 m edge-arrays have been investigated for different interelement separations, from 0.2 to 2.0 m. The magneto-optical response is measured both at reflected and diffracted beams, and it is compared with the results of a model that uses micromagnetic simulations and optical diffraction theory to calculate the magneto-optical response for different diffracted spots. A satisfactory agreement between the experiments and the predictions from the combined micromagnetic and optical diffraction models allows the interpretation of the experimental data and provides a way to analyze and understand the physical meaning of the magneto-optic diffracted signal. The comparison of this diffracted magneto-optical experimental data with predictions from simple reversal models allows us to monitor different element magnetization reversal mechanisms as the separation between elements in the array varies.

No abstract

Hydrogen (H) incorporation into AlGaN/GaN heterostructures used in high electron mobility transistors, grown by different methods, is studied by high-resolution depth profiling. Samples grown on sapphire and Si(1 1 1) substrates by molecular-beam epitaxy and metal–organic vapour phase epitaxy; involving H-free and H-containing precursors, were analysed to evaluate the eventual incorporation of H into the wafer. The amount of H was measured by means of nuclear reaction analysis (NRA) using the 1H(15N,αγ)12C reaction up to a depth of ∼110 nm into the heterostructures. Interestingly, the H profiles are similar in all the samples analysed, with an increasing H content towards the surface and a negligible H incorporation into the GaN layer (0.24 ± 0.08 at%) or at the AlGaN/GaN interface. Therefore, NRA shows that H uptake is not related to the growth process or technique employed and that H contamination may be due to external sources after growth. The eventual correlation between topographical defects on the AlGaN surface and the H concentration are also discussed.

The performance of axial fans in close proximity to the electromagnetic compatibility (EMC) screens was analyzed by means of an experimental parametric study. The following geometrical parameters were studied: the hub-to-tip ratio, the ratio between fan thickness and fan diameter, the porosity and thickness of the perforated plate, and finally, the distance between the perforated plate and the inlet and the outlet of the fan. Screen porosity was found to be the most important parameter. Fan performance degradation is expressed by means of two correlations: one for the deterioration in the fan pressure at the no-flow point and the other for the flow rate reduction at the free delivery point. Both correlations were formulated as functions of screen porosity and the distance between the fan and the screen. We believe that the correlations can serve as a good guide for correct fan placement in a telecommunications cabinet.

This paper describes the experimental part of an electronics cooling research project. To undertake the experiments a test rig has been designed and assembled which represents a channel made by two parallel printed circuit boards (PCBs). Different arrangements of heated prismatic bodies can be mounted on the walls of the channel simulating electronic components. These bodies are cooled combining a channel cross-flow and impinging jets issuing from the walls. The working fluid is air and the maximum Reynolds number (calculated respect to the channel height) is 13740. The test rig has been designed to take flow measurements using different experimental techniques such as hot wire anemometry (HWA), particle image velocimetry (PIV) and infrared imaging (IR). Experimental data on fluid flow features around electronic components will be used to develop and validate turbulence models that will be implemented in Computational Fluid Dynamics simulations.The presented results are two preliminary experimental studies to analyse the influence of the component height (h) and the Reynolds number on the flow structure around the component. The first study was made with the component in a cross flow without impinging jet and in the second study an impinging jet was added to the cross flow. The Reynolds numbers employed are close to those used in applications within the electronics industry. The measurement mean (U, V), rms velocity (u rms , v rms ) profiles in three main regions of the flow, namely, the wake, the upper and the side region have been obtained using the hot wire anemometry. From theses measurements the lateral flow separation, reattachment points and recirculation have been studied.

A theoretical formalism that allows analysis of the magneto-optical response of nanocorrugated ferromagnetic surfaces is presented and its validity checked with measurements in expressly fabricated structures. The formalism uses conventional scattering theory to find the expressions that account for the power scattered per unit area, and finds that for particular light-scattering directions and incidence polarizations the topographic and the magnetic contributions to the scattered light can be separated. By comparing theoretical and experimental results the magnetic state of the surfaces and its field evolution can be extracted.

The fabrication of a periodic domain structure in a ferromagnetic thin film is reported. This periodic domain structure is formed in a thin continuous magnetic film by coupling it to a periodic array of magnetic elements grown on top. When the array and the continuous film are exchange decoupled, magnetostatic interactions produce in the continuous layer a domain structure replica of the topographic pattern at selected field values. The present work reports a direct confirmation of this periodic domain structure in the flat continuous film by Kerr microscopy, which is responsible for the pure magnetooptic diffraction. The effect on the magnetization processes of one-and two-dimensional structures with different periodicities and dimensions is studied in detail and compared with micromagnetic simulations, for Co and Fe films.