scispace - formally typeset
Search or ask a question

Showing papers in "Physica Status Solidi (a) in 1999"


Journal ArticleDOI
TL;DR: In this article, a simple and straightforward procedure is presented for the rationalisation of anisotropic strain broadening in terms of the contrast effect of dislocations or dislocation-like lattice defects, making the evaluation of particle size or coherent domain size, and the density and arrangement of dislasations rather simple with a high precision, especially in the case of untextured polycrystals or when dislocation population on the possible slip systems is uniform.
Abstract: Anisotropic strain broadening in X-ray line profile analysis means that the width of diffraction profiles of Bragg reflections is not a monotonous function of the diffraction angle. The lack of a physically sound model makes the interpretation of line broadening and the structure refinement procedure of Rietveld often difficult or even impossible. A simple and straightforward procedure is presented for the rationalisation of anisotropic strain broadening in terms of the anisotropic contrast effect of dislocations or dislocation-like lattice defects. The procedure makes the evaluation of particle size or coherent domain size, and the density and arrangement of dislocations rather simple with a high precision, especially in the case of untextured polycrystals or when dislocation population on the possible slip systems is uniform. Three illustrative examples of severely deformed microcrystalline copper, the Rb 3 C 60 f.c.c. fullerite and a ball-milled iron powder show the effectiveness of the procedure.

353 citations


Journal ArticleDOI
TL;DR: In this paper, the effects of low pressure MOVPE on epitaxial lateral overgrowth (ELO) was studied in relation to the growth temperature, and the mechanism of the morphological change was discussed based on the stability of the surface atoms.
Abstract: Effects of reactor pressure on the epitaxial lateral overgrowth (ELO) via low pressure MOVPE have been studied in relation to the growth temperature. For the ELO GaN on SiO2 stripes along the 〈11-00〉 direction of the underlying GaN, by decreasing reactor pressures from 500 to 40 Torr or by increasing growth temperatures from 950 to 1050 °C, the (0001) surfaces become broad and the side walls are varied from inclined {112-2} surfaces to vertical {112-0} surfaces. For stripes along the 〈112-0〉 direction, the shapes of ELO GaN are independent of the reactor pressures and the growth temperatures. The mechanism of the morphological change is discussed based on the stability of the surface atoms. Typical ELO GaN layers with two-step growth are demonstrated and characterized in their crystalline properties.

293 citations


Journal ArticleDOI
TL;DR: In this paper, three squeeze cast Mg-Gd binary alloys have been investigated after a solution heat treatment by isochronal annealing up to 500 °C using electrical resistivity and hardness measurements.
Abstract: High thermal stability and good mechanical properties are crucial for the wider future application of magnesium alloys. One of the most promising directions is the alloying of Mg with heavy rare earth elements such as Gd, Dy, Tb, etc. Three squeeze cast Mg–Gd binary alloys (up to 15 wt% Gd) have been investigated after a solution heat treatment by isochronal annealing up to 500 °C using electrical resistivity and hardness measurements. The microstructural development during this treatment, responsible for the observed changes, was observed by transmission electron microscopy. The decomposition of α′-Mg supersaturated solid solution in Mg–14.55 wt% Gd with increasing heating temperature is as follows: β″ (D019) metastable phase β′ (c-b.c.o.) metastable phase β(Mg5Gd f.c.c.) stable. Peak hardening is achieved by a heat treatment resulting in precipitation of β′ phase in the shape of fine plates parallel to all three 21-1-0 planes of the α-Mg matrix. At higher temperatures (above 280 °C) coarsening occurs and only one orientation of β′ plates remains. The decomposition of Mg–4.47 wt% Gd and Mg–9.33 wt% Gd alloys differs from that of Mg–14.55 wt% Gd by the absence of the β′ phase. Um das Anwendungsfeld fur Magnesiumlegierungen zu erweitern, ist die Entwicklung von Legierungen mit hoher thermischer Stabilitat bei gleichzeitig guten mechanischen Eigenschaften erforderlich. Eine der vielversprechendsten Entwicklungen ist das Zulegieren von schweren Seltenen Erden wie z. B. Gd, Dy und Tb. Nach einer Homogenisierungsbehandlung wurden drei presgegossene binare Mg–Gd-Legierungen (bis zu 15 Gew.% Gd) mit Hilfe elektrischer Widerstandsmessungen und Hartemessungen hinsichtlich ihres Verhaltens bei isochronen Gluhungen bis zu 500 °C untersucht. Die wahrend dieser Warmebehandlung eingetretenen mikrostrukturellen Veranderungen wurden mittels transmissionselektronenmiskroskopischer Untersuchungen verfolgt. Mit der Gluhungstemperaturerhohung bilden sich in einer Mg–14,55 Gew% Gd-Legierung folgende Phasen: β″ (D019) metastabile Phase β′ (c-b.c.o.) metastabile Phase β (Mg5Gd f.c.c.) stabil. Die hochsten Hartewerte sind mit einer Warmebehandlung erreichbar, die zur Bildung der β′-Phase in Form feiner Plattchen parallel zu allen drei 21-1-0-Ebenen der Mg-Matrix fuhrt. Bei hoheren Temperaturen (uber 280 °C) bleibt nur eine Orientierung der β′-Phase erhalten und es tritt eine Vergroberung ein. Die korrespondierenden Entmischungsvorgange in einer Mg–4,47 Gew.% Gd- und einer Mg–9,33 Gew.% Gd-Legierung ist durch die Abwesenheit der β′-Phase gekennzeichnet.

197 citations


Journal ArticleDOI
TL;DR: In this paper, the authors present experimental data on the bonding of CO and NO to vacuum-cleaved MgO(100) and NiO( 100) and compare them with theoretical results.
Abstract: This paper presents experimental data on the bonding of CO and NO to vacuum-cleaved MgO(100) and NiO(100) and compares them with theoretical results. In the case of CO and NO on NiO(100) we find that the bonding energies obtained for the vacuum-cleaved single crystals agree well with results of recent studies on thin NiO(100) films grown by oxidation of Ni(100) whereas they are at variance with theoretical results. On the other hand, for CO on MgO(100) the experimental data fit well to recent theoretical studies while they contradict studies of adsorption on MgO(100) films grown on Mo(100). The experimentally determined values for the adsorption energies are 0.30 and 0.57 eV for adsorption of CO and NO on NiO(100). respectively, and 0.14 and 0.22 eV for adsorption on MgO(100). We suggest that the stronger bonding to NiO(100) as compared to MgO(100) is due to the influence of the 3d electrons of NiO.

95 citations


Journal ArticleDOI
TL;DR: In this article, a model for threading dislocation flow in relaxed graded SiGe buffers is used to determine the potential lower limit of threading dislocations in relaxed SiGe buffer layers on Si.
Abstract: Recent advances in the understanding and control of threading dislocations in substantially relaxed SiGe buffer layers on Si are presented. A model for threading dislocation flow in relaxed graded SiGe buffers is used to determine the potential lower limit of threading dislocation density in relaxed SiGe buffers. Greater densities than expected from the model are seen in relaxed graded alloys with final concentrations greater than 50%. We show that the culprits of the higher threading dislocation density are threading dislocation pile-ups. Observation of early development of pile-ups confirms that inhomogeneous misfit dislocation densities in the graded buffer form regions of more severe crosshatch on the surface that impede dislocation flow. By using chemomechanical planarization (CMP), deleterious pile-up formation can be avoided, and previously formed pile-ups can be destroyed by regrowth of a graded layer. Experiments with CMP and regrowth of graded layers suggest that dislocation annihilation can be effective at reducing threading dislocation densities to densities of the order expected by the model. High quality Ge on Si layers created with the CMP process were used as templates to grow high quality GaAs on Si with strong room temperature photoluminescence and record minority carrier lifetime.

92 citations


Journal ArticleDOI
TL;DR: In this paper, X-ray diffraction patterns indicated the presence of a single spinel phase and the lattice parameter was found to be nearly independent of the doped rare earth ion type.
Abstract: Samples of Cu-Zn ferrites doped with different rare earth ions (La, Sm, Nd, Gd, and Dy) are investigated X-ray diffraction patterns indicated the presence of a single spinel phase The lattice parameter is found to be nearly independent of the doped rare earth ion type The density is measured and the porosity is calculated for each sample The magnetization is measured on toroidal samples using the induction method The initial permeability μ i is measured as a function of temperature at a constant frequency of 10 kHz and the Curie temperature T C is determined The variations of magnetization, relative permeability and the initial permeability as a function of the effective magnetic moment of the doped rare earth ion are discussed

85 citations


Journal ArticleDOI
TL;DR: In this article, defect related states and excitonic transitions in epitaxial GaN have been studied by combining cathodoluminescence and transmission electron microscopy, and the structural origin of these defects in terms of formation of VGa−SiGa and VGa-ON complexes were discussed.
Abstract: Defect related states and excitonic transitions in epitaxial GaN have been studied by combining cathodoluminescence and transmission electron microscopy. A series of deep levels with energies at about 2.4, 2.6 and 2.8 eV has been found by low temperature cathodoluminescence on free-standing 150 μm thick epitaxial GaN. These deep levels are characterised by a high recombination efficiency. They are radiative from 5 to 70 K and undergo a nonradiative transition at 70 K. These levels completely quench the near band edge and the conventional yellow emissions. We discuss the structural origin of these defects in terms of formation of VGa–SiGa and VGa–ON complexes. The consequences of our model with respect to non radiative transitions at threading dislocations are also presented. An excitonic transition at 3.41 eV close to the near band edge line on differently grown epitaxial GaN has been correlated to stacking faults. This line can be explained by a model based on the concept of excitons bound to SFs that form a quantum well of cubic material in the wurtzite lattice of the layer.

85 citations


Journal ArticleDOI
TL;DR: In this article, a statistical nucleation and growth model using rate equations and a Focker-Planck equation is applied to simulate the oxygen precipitation process numerically, and it is found that certain types of oxygen defects reduce the lifetime of the bulk and enhance the recombination activity of dislocations.
Abstract: Oxygen and carbon are the main impurities in multicrystalline silicon for photovoltaic applications. Precipitation of oxygen and carbon occurs during crystal growth and solar cell processing. Depending on the thermal conditions and the initial oxygen and carbon content various types of SiO2, SiC precipitates and oxygen related defects are observed and investigated by IR spectroscopy and transmission electron microscopy. Topographic μ-PCD measurements are used to study the minority carrier lifetime in the material locally. It is found that certain types of oxygen defects reduce the lifetime of the bulk and enhance the recombination activity of dislocations. Quantitative measurements of the oxygen precipitation of pre-annealed specimens are carried out to study the oxygen precipitation systematically. A statistical nucleation and growth model using rate equations and a Focker-Planck equation is applied to simulate the precipitation process numerically.

84 citations



Journal ArticleDOI
TL;DR: In this paper, the intrinsic stacking fault energy in 20 tetrahedrally coordinated crystals, determined by electron microscopy from the widths of extended dislocations, was found to have correlations with the effective charge, the charge redistribution index of the crystals and with the c/ aratio of the wurtzite structure.
Abstract: The energies of the intrinsic stacking fault in 20 tetrahedrally coordinated crystals, determined by electron microscopy from the widths of extended dislocations, range from a few mJ/m2 to 300 mJ/m2. The reduced stacking fault energy (RSFE: stacking fault energy per bond perpendicular to the fault plane) has been found to have correlations with the effective charge, the charge redistribution index of the crystals and with the c/ aratio of the wurtzite structure (for many of the diamond and zincblende crystals, the wurtzite structure has also been reported to exist as a metastable phase). The implication of the correlations has been discussed from the point of the relative stability of the crystals between the cubic structure and the hexagonal structure.

81 citations


Journal ArticleDOI
TL;DR: In this article, the electronic states of boron and phosphorus in diamond have been studied by infrared absorption and photo-thermal ionisation spectroscopies, and the results suggest that the top of the valence band of diamond is different from that of silicon and germanium.
Abstract: The electronic states of boron and phosphorus in diamond have been studied by infrared absorption and photo-thermal ionisation spectroscopies. High quality boron doped synthetic diamond (p-type conductive) and phosphorus-doped CVD diamond film (n-type conductive) were used for this study. In the case of boron-doped diamond, the four main excited states of the bound hole follow a Rydberg series, suggesting that boron has a hydrogen-like behaviour, with a weak splitting of the excited states. The consistent values of the optical ionisation energy (E0 = 382 meV), of an “average” effective mass (m* = 0.74m0) and of the Bohr radius of the ground state (a* = 4.1 A) deduced from the Rydberg series support this suggestion. The comparison with the effective mass approximation, applied for acceptor states in diamond, suggests that the top of the valence band of diamond is different from that of silicon and germanium. In the case of phosphorus-doped diamond, two excited states of the bound electron have been observed for the first time, at 523 and 562 meV from the ground level. The good agreement with the effective mass approximation suggests that phosphorus is a shallow donor, and allows us to propose a first value of the optical ionisation energy of phosphorus in diamond of about 600 meV, consistent with Hall effect measurements.

Journal ArticleDOI
TL;DR: In this article, the nature of damage induced by ion-impact and its annealing is studied both experimentally and theoretically using molecular dynamics (MD) computations using the Tersoff potential, and it is shown that a deeply buried highly damaged region is created inside the diamond sample by imparting high momenta to lattice atoms aimed towards the same point in the crystal.
Abstract: The nature of damage induced by ion-impact and its annealing are studied both experimentally and theoretically. The experimental methods employed include measurements of the changes in electrical conductivity, in material density and in Raman spectra. These are measured for natural type IIa diamonds containing different amounts of ion-implantation induced damage and subjected to different annealing temperatures (up to 1300 K for 20 min). The simulations performed are based on Molecular Dynamics (MD) computations using the Tersoff potential. A deeply buried highly damaged region is created inside the diamond sample by imparting high momenta to lattice atoms aimed towards the same point in the crystal. The nature of the damage so created is statistically analyzed yielding information on the formation of threefold-coordinated atoms in the damage region. The transformation that the damaged region undergoes as a result of “annealing” (up to 4000 K for 50 ps) is investigated. Both experiment and theory show that diamond which contains a low density of point defects can anneal back to diamond whereas, for damage levels beyond a certain level, it tends to graphitize. The stable defect in damaged diamond seems to be, according to both experiment and theory, the 〈100〉 split interstitial. Electrically, point defects in diamond act as donor centers.

Journal ArticleDOI
TL;DR: In this article, the structural and electronic properties of metal silicide precipitates in silicon and their interrelation were summarized using high-resolution transmission electron microscopy and deep level transient spectroscopy together with numerical simulations.
Abstract: This paper summarizes current understanding of structural and electronic properties of metal silicide precipitates in silicon and their interrelation. Combined studies of high-resolution transmission electron microscopy and deep level transient spectroscopy together with numerical simulations show that the bounding dislocation of nickel silicide platelets is the key to understand their rapid growth and electrical properties. Different misfit relaxation phenomena govern the structural evolution of copper silicide precipitates from their early stages to the well-known colony growth. This evolution involves different types of secondary defects indicating that the deep band-like states observed throughout this process are associated with the silicide precipitates themselves.

Journal ArticleDOI
TL;DR: In this paper, two simple models concerning the static and the dynamic charge transfer are introduced, which result in an electric double-layer and changed conduction electron density in the cluster, while the second reduces, by phase relaxation, the lifetime of the Mie resonance.
Abstract: Nanostructured material is characterized by its structuring interfaces. Cluster–matter, i.e. cluster/matrix systems, proved to be a well suited model system to study their electronic properties. The interactions between cluster and matrix induce structural and electronic effects which cause the interface to evolve into an extended interlayer. If the particles are metallic and develop a well defined optical Mie resonance, this resonance can be used to monitor electronic properties of the interlayer with high accuracy. In Part I (Section 2), after a short description of Mie's theory, two simple models concerning the static and the dynamic charge transfer are introduced. The first one results in an electric double-layer and changed conduction electron density in the cluster, while the second reduces, by phase relaxation, the lifetime of the Mie resonance. Both effects bear information about the electronic adsorbate states in the interlayer. In Part II (Section 3), data from the experimental investigation of a broad field of novel cluster–matter systems based on Ag-clusters are presented and a discussion closes the talk.

Journal ArticleDOI
TL;DR: In this article, the growth of ternary AlGaN and InGaN by molecular beam epitaxy was studied, leading to a quantitative model describing alloy composition and growth rate as a function of group III fluxes, N flux and growth temperature.
Abstract: We report on a study of the growth of ternary AlGaN and InGaN by molecular beam epitaxy, leading to a quantitative model describing the alloy composition and growth rate as a function of group III fluxes, N flux and growth temperature. For low growth temperatures, the composition is exclusively determined by the different bond strengths between the group III elements and N, leading to a complete displacement of the more weakly bound species. The In loss from InGaN observed for typical growth conditions is caused by thermal decomposition of the growing layer with an activation energy between 3.5 and 3.8 eV depending on the In content.

Journal ArticleDOI
TL;DR: In this article, a single crystal GaN thin film was successfully grown on a Si(111) substrate by means of atmospheric-pressure metalorganic chemical vapor deposition, and the results suggest GaN on Si with an AlGaN/AlN intermediate layer provides reliable light emitting devices.
Abstract: A single crystal GaN thin film was successfully grown on a Si(111) substrate by means of atmospheric-pressure metalorganic chemical vapor deposition. An intermediate layer consisting of AlN and AlGaN improved the quality of GaN on Si with a mirror-like surface and reduced the pits and cracks over the surface. The full width at half maximum (FWHM) of the double-crystal X-ray rocking curve for GaN(0004) was 600 arcsec. Photoluminescence measurement at 4.2 K for a nondoped film revealed a sharp band-edge emission with a FWHM of 8.8 meV, which is the narrowest value reported to date. GaInN multi-quantum-well structure was grown on this structure and showed a strong blue emission peaking at 470 nm. The results suggest GaN on Si with an AlGaN/AlN intermediate layer provides reliable light emitting devices on Si substrate.

Journal ArticleDOI
TL;DR: In this paper, Nitrogen was incorporated by adding N 2 to the C 2 H 2 plasma used for the deposition of ta-C : H films, resulting in a polymeric-like material.
Abstract: Structural changes induced by the incorporation of nitrogen into ta-C : H films have been studied by Electron Energy Loss Spectroscopy, X-Ray Photoelectron Spectroscopy, Fourier Transformed Infrared Spectroscopy and Ultraviolet-Visible Spectroscopy. ta-C:H films have been synthesised using a low pressure Electron Cyclotron Wave Resonance (ECWR) source which provides a plasma beam with a high degree of ionisation and dissociation. Nitrogen was incorporated by adding N 2 to the C 2 H 2 plasma used for the deposition of ta-C : H films. The N/C atomic ratio in the films rises rapidly until the N 2 /C 2 H 2 gas ratio reaches three, and then increases more gradually, while the deposition rate decreases steeply. Chemical sputtering of the forming films and the formation of molecular nitrogen within the films limit the maximum nitrogen content to about N/C = 0.6. For low nitrogen content the films retain their diamond-like properties, however as N/C atomic ratio increases, a polymeric-like material is formed, with > C=N- structures and terminating C=N and NH groups that decrease the connectivity of the network.

Journal ArticleDOI
TL;DR: Optical second-harmonic generation, although a sophisticated technique, is making a significant contribution to the characterization of surfaces and interfaces as discussed by the authors, with all pressure ranges being accessible; insulators being studied without the problem of charging effects; solid-solid and solid-liquid interfaces being characterized by utilizing the large penetration depth of optical radiation.
Abstract: Optical second-harmonic generation, although a sophisticated technique, is making a significant contribution to the characterization of surfaces and interfaces. Advantages include: all pressure ranges being accessible; insulators being studied without the problem of charging effects; solid–solid and solid–liquid interfaces being characterized by utilizing the large penetration depth of optical radiation. Single wavelength studies yield information on the crystallographic and magnetic structure of surfaces and interfaces, and are particularly sensitive to symmetry changes. Spectroscopic and time-resolved studies, which have recently become simpler due to the availability of high pulse power, tunable, broad band laser sources, additionally provide electronic structure and dynamics information, although interpretation may not be simple. Recent examples from well-characterized semiconductor interfaces are discussed, and it is shown that the technique is now well understood at the phenomenological level, although theoretical calculations are proving to be difficult.

Journal ArticleDOI
TL;DR: In this article, Co-Ni-Cu/Cu multilayers with a maximum giant magnetoresistive (GMR) ratio of about 10% were produced, and various nucleation and growth mechanisms were obseived.
Abstract: Cu, Co and Ni thin films and Co-Ni-Cu/Cu multilayers have been electrodeposited directly on n-type silicon substrates. This removes the need of using a seed-layer deposited by some other methods as a part of the growth process and integrates an efficient, inexpensive and convenient method for fabricating thin films with silicon technology. The deposits were prepared under potentiostatic conditions from different aqueous solutions, containing basically: (i) sulphates of the metallic ions plus sodium sulphate and boric acid for thin films and (ii) Ni sulphamate, Co sulphate, Cu sulphate, boric acid and sulphamic acid for multilayers. Aspects related to the deposition process and deposited layers were investigated by cyclic voltammetry, current transients, scanning electron microscopy. Rutherford backscattering and magnetoresistance measurements. Typically, thin compact metallic layers with regular granularity were obtained. Depending on the additives used, different nucleation and growth mechanisms were obseived. Magnetic multilayers with a maximum giant magnetoresistive (GMR) ratio of about 10% were produced.

Journal ArticleDOI
TL;DR: In this article, the type and density of threading dislocations in GaN epitaxial layers grown on c-plane sapphire have been analyzed by using nondestructive high-resolution X-ray diffraction.
Abstract: The type and density of threading dislocations in GaN epitaxial layers grown on c-plane sapphire have been analyzed by using nondestructive high resolution X-ray diffraction. The highly distorted GaN layers were described as mosaic crystals characterized by a mean tilt and twist angle between the mosaic blocks which are correlated with the densities of screw and edge type threading dislocations, respectively. Triple axis rocking curves of (00l) reflections for varying l-indices were used to determine the tilt angle, while the twist was extrapolated from ω-scans for (hkl) Bragg reflections with h or k nonzero, measured in skew symmetric diffraction geometry. This defect analysis was applied to selected GaN layers grown by molecular beam epitaxy (MBE) and metalorganic chemical vapour deposition (MOCVD) revealing clear differences between both sample types.

Journal ArticleDOI
TL;DR: In this paper, the initial stages of electrolytic metal deposition are described and exemplified by the classical model system, Cu on Au(hkl), and some of the parameters governing nucleation and growth are pointed out.
Abstract: This paper comprises, in the form of an overview, various aspects of the initial stages of electrolytic metal deposition, demonstrated and exemplified by the classical model system, Cu on Au(hkl). Describing underpotential as well as overpotential deposition, some of the parameters governing nucleation and growth are pointed out. The profound effect of substrate orientation is shown and various examples of the influence of organic additives on the deposition process are given. Most recently, metal deposition onto electrodes covered with self-assembled monolayers has received increased attention as a possible model system for metal deposition onto a non-conducting stirface and therefore, preliminary results for Cu deposition on alkanethiol covered Au(111) are presented.


Journal ArticleDOI
TL;DR: In this article, the authors indicate the market size, industrial requirements including technical and economical considerations, the state of the art and the remaining open issues for GaN-based optoelectronic applications.
Abstract: More than a promising material for optoelectronics and electronics, GaN already represents an economic reality. Domains of applications include blue and green LEDs, blue laser, high power–high frequency electronics, high temperature electronics, ultraviolet detectors… For each of these applications, I will indicate the market size, the industrial requirements including technical and economical considerations, the state of the art and the remaining open issues.

Journal ArticleDOI
TL;DR: In this paper, the structural quality of GaN stripes grown selectively via lateral epitaxial overgrowth (LEO) and pendeo-epitaxy (PE) or maskless lateral epitaxy was investigated.
Abstract: Recent studies of the structural quality of GaN stripes grown selectively via lateral epitaxial overgrowth (LEO) and pendeo-epitaxy (PE) or maskless lateral epitaxy, reveal that the regions of lateral growth exhibit four-to-five orders of magnitude lower density of dislocations compared to the regions of vertical growth. In both cases the crystallographic templates for the lateral growth are the {112-0}, {11-00}, or the {11-01} side facets of the GaN. The examination of the morphology of the top surfaces, side facets, and interfaces of the LEO- and PE-GaN stripes with the underlying and adjacent interfaces, reveal their striking similarity with the thermally generated stress/strain gradient profiles as calculated via finite element analysis. A comparison between the stress distribution as a result of the mismatches in the coefficients of thermal expansion among the films in the structures, grouped in four types with different geometries will be presented: (I) conventional LEO-GaN, (II) conventional LEO-GaN case without SiO2 layer, (III) PE-GaN mode A, and (IV) PE-GaN mode B.

Journal ArticleDOI
L.F. Eastman1
TL;DR: In this paper, the frequency response and drain-source breakdown voltage have been determined experimentally and the average electron transit velocity, determined from cut-off frequency, is 1.2 × 107 cm/s and the breakdown electric field is estimated to be 3 MV/cm.
Abstract: The fabrication of AlGaN/GaN high electron mobility transistors (HEMTs) employs strong spontaneous and piezoelectric polarization. The material was grown by OMVPE at 1050 °C and 75 Torr. The Hall measurements yielded ≈1 × 1013/cm2 electron sheet density, with 1200 to 1600 cm2/Vs electron mobility for 300 A Al0.3Ga0.7N/GaN on sapphire at 300 K. The process steps for mesa isolation, Ohmic and Schottky contacts are presented. The frequency response and drain–source breakdown voltage have been determined experimentally. The average electron transit velocity, determined from cut-off frequency, is 1.2 × 107 cm/s, and the breakdown electric field is estimated to be 3 MV/cm. On sapphire substrates, large periphery devices yielded 1.8 W/mm at 78% power-added efficiency and small periphery devices yielded 3.3 W/mm at 36% power-added efficiency at 4 GHz. Performance limitations due to electron trapping during HEMT operation are included. Predictions of >20 W/mm microwave output power are presented for these devices on SiC substrates.

Journal ArticleDOI
TL;DR: In this paper, the authors performed compression creep tests with changes in stress to determine the stress exponent n of the creep rate in the steady state of deformation of pure Al at 923 K, i.e. close to the melting point.
Abstract: Compression creep tests with changes in stress were performed to determine the stress exponent n of the creep rate in the steady state of deformation of pure Al at 923 K, i.e. close to the melting point. The tests confirm that n increases with decreasing σ and e. The decrease in n to 1 reported in the literature and associated with an autonomous dislocation mechanism of viscous deformation called Harper-Dorn creep could not be reproduced.

Journal ArticleDOI
TL;DR: In this paper, the deformation of GaAs, InP, and InSb under hydrostatic confining pressure has been performed in the temperature range 77 to 500 K. Electron microscopy reveals long screw dislocations left in the deformed specimens.
Abstract: Plastic deformation of GaAs, InP, and InSb under hydrostatic confining pressure has been performed in the temperature range 77 to 500 K. Slip lines indicate cross-slip of screw dislocations. Electron microscopy reveals long screw dislocations left in the deformed specimens. The temperature dependence of the critical shear stress τc below 300 K is weak, producing a hump in the τc — T curve. The kink pair formation of non-dissociated screw dislocations in the shuffle set planes is responsible for the plastic deformation at low temperatures.

Journal ArticleDOI
TL;DR: A compact electron cyclotron wave resonance (ECWR) source has been developed for the high rate deposition of hydrogenated tetrahedral amorphous carbon (Ta-C:H) as discussed by the authors.
Abstract: A compact electron cyclotron wave resonance (ECWR) source has been developed for the high rate deposition of hydrogenated tetrahedral amorphous carbon (ta-C:H). The ECWR provides growth rates of up to 900 A/min over a 4 diameter and an independent control of the deposition rate and ion energy. The ta-C : H was deposited using acetylene as the source gas and was characterized in terms of its sp 3 content, mass density, intrinsic stress, hydrogen content, C-H bonding, Raman spectra, optical gap, surface roughness and friction coefficient. The results obtained indicated that the film properties were maximized at an ion energy of approximately 167 eV, corresponding to an energy per daughter carbon ion of 76 eV. The relationship between the incident ion energy and film densification was also explained in terms of the subsurface implantation of carbon ions into the growing film.

Journal ArticleDOI
TL;DR: In this paper, a coherency defect technique is developed for domain pattern energetics for (111) oriented epitaxial tetragonal ferroelectric (FT) films, where six possible domain variants may form, half of which are related by the inversion of the polarization vector.
Abstract: The possible domain patterns are developed for (111) oriented epitaxial tetragonal ferroelectric (FT) films. The domain patterns in the film form as the result of phase transition from the paraelectric to ferroelectric state to minimize the elastic energy of the system at the expense of creating of domain boundaries and developing non-uniform elastic fields near the film/substrate interface. Six possible domain variants may form, half of which are related by the inversion of the polarization vector. The possible domain walls arising between pairs of variants can be derived from the conditions of the mechanical and charge compatibility. These walls are {101} boundaries (pseudocubic indexing) and can either be inclined or normal to film/substrate interface. The domain patterns with inclined boundaries have a flat free film surface and possess non-zero net polarization in the direction normal to the film surface, i.e., they correspond to the poled film state. The domain patterns with normal boundaries lead to ‘puckering’ of the film surface, simultaneously they are related to the unpoled state of ferroelectric films. The coherency defect technique is developed for domain pattern energetics for (111) oriented FT films. The coherency defects include (i) a cross-grid of edge dislocations with unbalanced densities (which lead to in-plane biaxial strain field), (ii) Somigliana screw dislocations (which produce alternating sense of shear in neighboring domains), and (iii) wedge disclinations (which are related to out-of-plane rotations in neighboring domains). Analytical calculations of the pattern energy are performed for single embedded domain and multidomain patterns. These calculations are based on the use of screened configurations for representative coherency elements: disclination and Somigliana screw dislocation dipoles and quadrupoles. It is predicted that there is no critical thickness for domain pattern formation in (111) oriented epitaxial tetragonal ferroelectric films. Tiling of different domain patterns in complex mesoscopic structures is also discussed and supported by experimental observations.

Journal ArticleDOI
TL;DR: In this article, it was shown that the non-self-limiting adsorption process is more effective for uncharged polymers than the electrostatic attraction mechanism, which was attributed to H-bonding.
Abstract: The layer-by-layer or self-assembly (SA) technique, based on the electrostatic interaction between oppositely charged layers, provides a simple and elegant way of producing multilayer structures from a variety of materials. Multilayers are normally employed because one layer is usually too thin for any device application. We show that considerably thicker layers may be obtained if the adsorption process is non-self-limiting, unlike the case of mechanisms entirely based on electrostatic attraction. This was obtained in spontaneously adsorbed layers of polyaniline (PANi), poly(o-methoxyaniline) (POMA) and tetrahydrothiophenium polyphenylenevinylidene, a PPV precursor. The non-self-limiting adsorption is attributed to H-bonding which is more effective for uncharged polymers. It is also shown that the adsorption process is only non-self-limiting under appropriate experimental conditions, since adsorption must be performed under a non-equilibrium condition. Producing thicker layers via the non-self-limiting adsorption procedure may have important implications in device applications, but for that one needs to overcome the drawback caused by aggregation that affects film quality. For PANi and POMA, we have also demonstrated that multilayers can still be produced from uncharged polymers, simply by protonating them in an acid solution, and then alternating one of them with a polyanion.