By Topic

Journal of Applied Physics

Issue 12 • Date Dec 1994

Filter Results

Displaying Results 1 - 25 of 90
  • Issue Table of Contents

    Page(s): toc1
    Save to Project icon | PDF file iconPDF (375 KB)  
    Freely Available from IEEE
  • The potential generated by current sources located in an insulated rectangular volume conductor

    Page(s): 7671 - 7676
    Save to Project icon | PDF file iconPDF (697 KB)  

    An analytic solution is derived for the potential generated by a current dipole source located in a three‐dimensional insulated volume conductor with rectangular cross section. The solution is first obtained from an eigenfunction expansion for a point current source in the volume with Neumann boundary conditions. A distributed volume current sink is introduced to maintain continuity of charge within the volume. The dipole potential is obtained directly from the point source potential and is shown to be reducible to a doubly infinite summation. An algorithm is presented for modifying the summation to achieve the most rapid convergence dependent on the specific geometry of the problem. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Nonmagnetic contrast in scanning Lorentz electron microscopy of polycrystalline magnetic films

    Page(s): 7677 - 7681
    Save to Project icon | PDF file iconPDF (599 KB)  

    The effect of the probe beam defocus on scanning Lorentz electron microscopy images has been investigated. Wave optical analyses show that, given a finite defocus, the variation of the transmission amplitude across the illuminated area on the specimen causes nonmagnetic contrast in the differential phase contrast (DPC) images. This nonmagnetic contrast depends on the sign and the amount of defocus. Since this contrast is virtually proportional to the gradient of the transmission amplitude, it should appear similar to the differentiated contrast of the bright‐field scanning transmission electron microscope (STEM) image. Therefore, the nonmagnetic noise appearing in the defocused DPC image can be substantially reduced by subtracting a suitably weighted differentiated bright‐field STEM image of the same area. This noise reduction procedure has been demonstrated by applying it to the unintentionally defocused DPC image of a recorded longitudinal magnetic storage medium. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Instability of an expanding charged‐particle beam

    Page(s): 7682 - 7689
    Save to Project icon | PDF file iconPDF (849 KB)  

    It is shown that a uniform collisionless charged‐particle beam expanding under the influence of electrostatic and internal pressure forces is linearly unstable. The instability, which tends to cause spreading and change the radial profile of the beam, implies a limit on the achievable focusing even when the lens system corrects for expansion. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fourier extrapolation of magnetic fields measured by electron‐beam tomography

    Page(s): 7690 - 7694
    Save to Project icon | PDF file iconPDF (504 KB)  

    A method for calculating the magnetic field on a plane near a magnetic recording head is presented. In this method, the magnetic field is measured at a plane farther from the head. The method involves taking the Fourier transform of the measured field, enhancing the large wave‐number region of the spectrum, and effecting the reverse Fourier transform of the modified spectrum. It is estimated that a field 0.1 μm away from the head can be calculated from a field measured 0.3 μm from it, with a maximum error of less than 0.35%. A field 0.5 μm from a thin‐film head calculated from the measured field at 0.7 μm has been compared with actual measurements of the field at 0.5 μm and found to agree within 8%. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Thermal stresses in tightly jacketed double‐coated optical fibers at low temperature

    Page(s): 7695 - 7703
    Save to Project icon | PDF file iconPDF (973 KB)  

    The thermal stresses in tightly jacketed double‐coated optical fibers at low temperature have been analyzed by the thermoelastic approach. The lateral pressure and normal stresses in the glass fiber, primary coating, secondary coating, and jacket have been derived. The thermal stresses in the optical fibers are affected by the temperature difference, material properties of polymeric coatings, and their thicknesses. It is possible to select the suitable polymeric coatings to produce a minimum lateral pressure in the glass fiber. The microbending loss is dominated by the lateral pressure in the glass fiber. To minimize such a microbending loss, the Poisson’s ratio of the primary and secondary coatings, and the Young’s modulus of the jacket should be increased. On the other hand, the Young’s modulus of the primary coating should be decreased. However, the thickness of the primary coating, and the thickness and Young’s modulus of the secondary coating exist the optimum values. The glass transition temperature of the primary coating should be lower than the temperature ranges under consideration. The thermal stresses for plane strain and zero axial force conditions are also discussed. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Extension of the tuning range of a femtosecond Ti:sapphire laser amplifier through cascaded second‐order nonlinear frequency conversion processes

    Page(s): 7704 - 7712
    Save to Project icon | PDF file iconPDF (1197 KB)  

    The article presents single step and cascade methods for converting the femtosecond radiation from a 1 kHz repetition rate Ti:sapphire regenerative amplifier to the near‐infrared (1.2–1.5 μm and 1.6–2.2 μm), visible (554–593 nm), and for the first time to our knowledge to the vacuum ultraviolet spectral region well below 190 nm. Using only solid state materials (nonlinear optical crystals) femtosecond pulses are generated through phase‐matched mixing processes down to 172.7 nm. The developed scheme for femtosecond parametric conversion provides nearly bandwidth limited femtosecond pulses, whose duration is comparable or even shorter than that of the pump pulses. The temporal gain narrowing in the optical parametric amplifier is identified as a pulse compression mechanism on the femtosecond time scale leading to generation of infrared light pulses as short as ≊50 fs. As a whole our all‐solid state laser complex provides independently tunable and synchronized sub‐200 fs light pulses simultaneously at several wavelengths and practically without relative jitter. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Optical pulse mixing measurement of carrier lifetime and absorption recovery time in reverse‐biased GaAs/AlGaAs single quantum well laser structures

    Page(s): 7713 - 7719
    Save to Project icon | PDF file iconPDF (739 KB)  

    We describe an optical pulse mixing experiment with a two‐segment GaAs/AlGaAs single quantum well, graded‐index separate confinement heterostructure laser to determine the effective lifetime of photogenerated carriers in a short reverse‐biased segment with subpicosecond resolution. This lifetime is of importance if such a segment is used as a fast photodetector or as a saturable absorber in a monolithic mode‐locked laser structure. We found that the lifetime depends not only on the applied bias but also on the excitation pulse energy. Lifetimes shorter than 5 ps have been observed. The strong dependence on excitation pulse energy is attributed to screening effects of the escaped carriers. A simple model based on an exact solution of the one‐dimensional Schrödinger equation for a particle in a quantum well in an electric field together with an electrical equivalent circuit verifies this assumption. © 1994 American Institute of Physics.   View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Imaging InGaAsP quantum-well lasers using near-field scanning optical microscopy

    Page(s): 7720 - 7725
    Save to Project icon | PDF file iconPDF (822 KB)  

    The application of near-field scanning optical microscopy (NSOM) to the characterization of InGaAsP multiquantum-well lasers is reported. Collection mode images are collected at varying drive currents from well below to well above the threshold current. The high resolution of NSOM (∼λ/20) provides a detailed mapping of the laser output from the active region as well as additional output from the surrounding mesa. Spectral analysis of the image shows that the extra emission is due to InP electroluminescence. In addition to the emission characteristics of the laser it is also possible to detect local heating of the laser facet via thermal expansion. Topographical images are achieved simultaneously with NSOM images by digitizing the feedback signal which maintains a constant tip-surface gap. It is shown that these data have direct implications on device performance and problems associated with carrier leakage and nonradiative defects. © 1994 American Institute of Physics.   View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Ultrasonic phase velocity and elastic modulus in isotopically enhanced manufactured diamonds

    Page(s): 7726 - 7730
    Save to Project icon | PDF file iconPDF (635 KB)  

    We have performed ultrasonic pulse‐echo experiments to measure the longitudinal and transverse acoustic velocities in nine single‐crystal manufactured diamonds. The 13C concentration in the samples ranged from nearly 0% to 99%. Small (4%–5%) but distinct decreases in both the longitudinal and transverse 〈100〉 phase velocities with increasing 13C content were observed. Using these velocities and density data, values for the cubic elastic moduli c11 and c44 were determined. Within experimental uncertainty (typically ≪0.5%), c11 remained constant with 13C content. However, c44 was observed to decrease by 2% over the range 0%–99% 13C. Phase velocities in 〈111〉‐oriented samples remained nearly constant with isotope content, implying a 2%–6% increase in 13C diamond’s effective elastic constants 1/3(c11+2c12+4c44) and 1/3(c11-c12+c44) over those of 12C diamond. Together with experimentally determined densities and crystalline orientations and the above results for c11 and c44, the 〈111〉 velocities were used to numerically invert the Christoffel equation for the elastic modulus c12. Surprisingly, c12 nearly doubled in the range 0% to 99% 13C, implying a 17% increase in the bulk modulus K=1/3(c11+2c12). © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Variational approach to the calculation of the radii in the stagnant core model of shaped charge jet formation

    Page(s): 7731 - 7740
    Save to Project icon | PDF file iconPDF (1167 KB)  

    The formation of a jet and slug from a collapsing shaped charge liner can be viewed, by appropriate choice of reference frame, as the result of two fluid streams impinging upon each other. In this article we consider this formation process and develop further the concept of a stagnant core model. In this model a core region of material is supposed to be stationary at the junction where the liner material turns to form the jet and slug. In our two‐dimensional treatment the boundaries of the core region and the free streamlines are assumed to be arcs of circles and the main problem is to determine the radii of these boundaries. However, unlike in previous work, a nonuniform flow field is assumed to exist in the circular flow region from the outset. The nonuniform flow field we derive needs to be matched with the (assumed) uniform flow in the impinging stream. To accomplish this a transition region in the impinging stream is postulated. Consideration of the mass and momentum balances in this region leads to further model equations. The first of these balances gives a relation between the radii of the free streamline and the stagnant core boundary. It is shown that there are no physically acceptable exact solutions to the model equations when the energy is minimized. However a very accurate approximate solution is shown to exist. This solution leads to an expression for the liner speed on the core boundary which is identical to the critical speed used in a recent study on the formation of incoherent jets. Physically sensible values of the free streamline radius are also shown to result from this approximate solution. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Plasma‐filled applied B ion diode experiments using a plasma opening switch

    Page(s): 7741 - 7749
    Save to Project icon | PDF file iconPDF (1346 KB)  

    In order for a plasma opening switch (POS) to open quickly and transfer power efficiently from an inductively charged vacuum transmission line to an applied B ion diode, the load impedance of the ion diode may be required to have an initial low impedance phase. A plasma‐filled diode has such an impedance history. To test the effect of a plasma‐filled diode on POS‐diode coupling, a drifting plasma was introduced from the cathode side of an applied B ion diode operated on the LION accelerator (1.5 MV, 4 Ω, 40 ns) at Cornell University. This plasma readily crossed the 2.1 T magnetic insulation field of the diode, and resulted in both increased diode electrical power, and an increased ability of the ion beam to remove material from a target. The plasma did not appear to have a noticeable effect on local beam steering angle. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Spectroscopic measurements of laser induced plasma during welding with CO2 laser

    Page(s): 7750 - 7756
    Save to Project icon | PDF file iconPDF (888 KB)  

    Results of spectroscopic measurements of laser‐induced plasma under welding conditions are presented. Welding was performed with the use of a continuous‐wave CO2 laser operating at a power of 2 kW. Argon or helium was used as a shielding gas. The welding metal was stainless steel or titanium. Emission spectra from plasma were measured with a spectrograph and 1254 Silicon Intensified Target detector connected to the optical multichannel analyzer. The plasma electron temperatures were determined from the relative intensities of spectral lines and the electron densities were determined either from Stark broadening of atomic line or the absolute intensity of ionic line. The distributions of the electron temperature and density over the metal surface are presented and the influence of a shielding gas on plasma parameters is demonstrated. The plasma parameters obtained were used to calculate the absorption of a laser beam in the plasma over the metal surface. No significant absorption was found in our experimental conditions. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Mathematical modeling of high‐power transferred arcs

    Page(s): 7757 - 7767
    Save to Project icon | PDF file iconPDF (1077 KB)  

    The behavior of an argon arc in a transferred arc furnace with a cold cathode and anode has been investigated using an elliptic model with emphasis on a high‐current, long arc (L=100–200 mm, I=500 A). Conservation equations for the laminar arc column are written assuming local thermodynamic equilibrium. The governing equations are solved by employing an efficient finite difference method. The anode is included in the calculation domain. Numerical calculations performed for a transferred argon arc show good agreement with temperature measurements obtained by emission spectroscopy. The results also indicate that, except in regions close to the anode, the arc behavior is parabolic. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Local crystal orientation in III‐V semiconductors: Polarization selective Raman microprobe measurements

    Page(s): 7768 - 7773
    Save to Project icon | PDF file iconPDF (629 KB)  

    The determination of local crystal orientation is of interest in understanding the dynamics of crystal growth processes. Examples are the investigation of grain and twin boundaries or microstructures in semiconductors. Raman microprobe measurements can be performed contactless, without destruction and with a local resolution of about 1 μm. The information contained in the scattering intensity of the LO and TO phonons in dependence on the polarization direction of the exciting laser beam is used to determine the complete orientation of zinc‐blende‐type III‐V semiconductor crystals. Sample rotation is not necessary. The reliability of the Raman method is demonstrated investigating a twin lamella in HB‐grown GaAs. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Surface modification during scanning tunneling microscope measurements on YBa2Cu3O7 thin films

    Page(s): 7774 - 7777
    Save to Project icon | PDF file iconPDF (573 KB)  

    During scanning tunneling microscopy measurements on YBa2Cu3O7 thin films with positive sample bias voltages, different kinds of surface modifications appear. While in the topography condition no surface modification occurs, the cleaning and etching conditions irreversibly change the surface topography. In the deformation condition a strong elastic deformation of the tip and/or the surface is observed. It is shown that field evaporation of material from the surface to the tip is responsible for the observed surface modification. In consequence, the tip is contaminated with different oxides. Localized states in the oxides make resonant tunneling of electrons through the oxide layer possible. The configuration of these localized states determines the kind of surface modification by the scanning tunneling microscope. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Linear electro‐optic effect based on flexoelectricity in a cholesteric with sign change of dielectric anisotropy

    Page(s): 7778 - 7783
    Save to Project icon | PDF file iconPDF (667 KB)  

    The linear electro‐optic effect in short‐pitch cholesterics is based on the linear coupling of the medium with an applied electric field. It has a number of remarkable properties. The electric field causes the optic axis to tilt in a plane parallel to the surfaces of the cell glass plates, giving the same symmetry as the electro‐optic effects in the smectic C* phase (surface stabilized ferroelectric liquid crystals and deformed helix mode) and the smectic A* phase (soft mode/electroclinic effect). For shutters and displays this guarantees a very wide viewing angle. The induced tilt is a linear function of the applied field, at least for small fields, which gives a well‐controlled continuous grey scale. Furthermore, it is practically independent of temperature. Response times of the order of 100 μs are easily achievable. The most interesting development in this effect would be to extend the linear regime to much larger tilt angles, in particular to 22.5°, where light could be modulated from 100% to zero transmission. In order to do this the perturbation from the quadratic dielectric coupling has to be ruled out or minimized, which requires materials with essentially zero dielectric anisotropy. This has been done, and it has been found that the bare flexoelectric‐induced tilt has a surprising range of linearity: The linear response in tilt could be followed up to about 30° after which the high electric field caused breakdown. The response time is typically about 100 μs and below. © 1994 American Institute of Physics.   View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Damping capacity of layered materials

    Page(s): 7784 - 7789
    Save to Project icon | PDF file iconPDF (687 KB)  

    Extensive measurements of the internal friction in a variety of layered materials such as graphite, metal dichalcogenides (MoS2, TiS2, NbSe2, TaS2, TaSe2, and TaTe2), and montmorillonite clays are reported. The internal friction of these layered materials exhibits features similar to those observed in graphite. In particular, a high temperature internal friction plateau develops at a temperature that is characteristic of the material. The physical mechanism for this plateau is discussed. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Plastically deformed region around indentations on Si single crystal

    Page(s): 7790 - 7796
    Save to Project icon | PDF file iconPDF (891 KB)  

    Expansion of a hemispherical shell by inner pressure has been widely applied for the model of the deformation by an indentation on a flat surface; however, the deformed region is not necessarily spherically symmetric, especially in anisotropic materials such as single crystals. Therefore, whether the spherical model is applicable in an indentation process for objective materials must always be kept in mind. Indentations have been made on the (111) surface of silicon single crystal at various temperatures. The three‐dimensional shape of the plastically deformed region was experimentally measured by means of an etching technique and its difference from the hemisphere was observed. It was never spherical but much more complicated, similar to a bottle gourd. The slip mechanism, which resulted in the observed shape of the plastic region, is discussed further. The plastic region was analytically obtained also on the assumption that the stress distribution was spherically symmetrical. The result is approximately in accordance with the observed shape. It is therefore concluded that the stress distribution is nearly spherical although the plastic region is far from it. The yield strength of silicon crystals and their temperature dependence were obtained based on the spherical model. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Influence of In on Si local vibrational modes in InxGa1-xAs (0≤x≤0.12)

    Page(s): 7797 - 7804
    Save to Project icon | PDF file iconPDF (1076 KB)  

    Local vibrational modes (LVM) of Si in substitutional sites have been observed by resonant Raman spectroscopy in highly doped (≥8×1018 cm-3) InxGa1-xAs layers, either relaxed or under strain, on [100] GaAs substrates. The peak frequency ωLVM of the Si on Ga site (SiGa) LVM in unstrained samples shifts to lower values with increasing In content. For x≤0.10 this shift is clearly higher than expected from a linear interpolation between the measured values in the binaries. The comparison between the SiGa peak frequency measured in both a full strained layer and a relaxed layer with similar composition provides a rough determination of the deformation potentials for the SiGaLVM in these layers: q/ω2LVM=-2.7±1 and p/ω2LVM=-2.5±1. As the In content becomes higher the width of the SiGa peak increases much more than that of the GaAs‐like longitudinal optical‐phonon peak, revealing the splitting due to the loss of local symmetry introduced by the In. New calibration factors for the Si‐defect concentrations have been deduced, which allow estimation of the solubility limit for the Si incorporation in substitutional positions, which ranges from 2.3×1019 to 2.6×1019 cm-3 for the layers at the growth conditions used. The analysis of the integrated intensity of the LVM Raman peaks indicates that the degree of electrical compensation is clearly reduced for increasing In up to x≤0.05, due to both an increase of the solubility limit for Si in these layers and a saturation or slight reduction of the SiAs‐related defect concentrations. This conclusion is also supported by Hall and plasmon measurements. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Surface energies and morphologies of chemical‐vapor‐deposited diamond films

    Page(s): 7805 - 7808
    Save to Project icon | PDF file iconPDF (480 KB)  

    Using a model where a fraction of carbon dangling bonds on the growth surface are saturated by hydrogen, the surface energies of chemical‐vapor‐deposited diamond crystals during the preparation process have been calculated. The results show that the surface energies increase with the increasing substrate temperature, and the surface energy‐temperature curves of the three most energetically favored crystal planes, (100), (110), and (111), intersect at a critical substrate temperature. Then, the changes of predominant facets in the surface morphology from {100} to {111} with the increasing substrate temperature can be interpreted. Furthermore, it is found that the surface energies of diamond are lower than that of graphite in the chemical‐vapor‐deposition process of diamond. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Real‐structure effects in the dynamical theory of grazing incidence x‐ray diffraction

    Page(s): 7809 - 7815
    Save to Project icon | PDF file iconPDF (744 KB)  

    A study is presented on the grazing incidence x‐ray diffraction in multilayers, with theoretical considerations on the effects of large‐scale and small‐scale surface and interface roughness and on the effects of interface transition layers. Based on the dynamical theory of grazing incidence x‐ray diffraction, the study provides a matrix form of solution for an arbitrary sequence of multilayers. The computations obtained with a developed computer program show the differences that can be observed by the effects of large‐scale and small‐scale roughness. The effects of small‐scale roughness and those of transition layers are similar and thus cannot be separated from each other. But the thickness of surface and interface transition layers or the rms‐values of roughness heights are determinable from the shape of diffraction curves by using two different methods. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Triple axis x‐ray investigations of semiconductor surface corrugations

    Page(s): 7816 - 7823
    Save to Project icon | PDF file iconPDF (1110 KB)  

    X‐ray reciprocal space mapping around the symmetrical (004) Bragg reflection and a kinematical x‐ray diffraction model were employed in order to determine the geometry and the structural perfection of surface corrugations or quantum wires. This method was used for the analysis of (001) Cd1-xZnxTe surface corrugations fabricated by holographic lithography and subsequently reactive ion etched with typical periods of 500 nm. Comparison of the measurement and simulation provides conclusive information on etching depth, wire period, wire width, and the inclination of the side walls. Furthermore, the analysis yields a parameter that contains information on side wall roughness, shape fluctuations and, in principle, the crystallographic damage caused by the reactive ion etching process. Due to the high resolution of triple axis diffractometry small strain gradients are observable in the damaged region. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Strain effects in epitaxial GaN grown on AlN‐buffered Si(111)

    Page(s): 7824 - 7828
    Save to Project icon | PDF file iconPDF (561 KB)  

    Growth of GaN thin films on AlN‐buffered Si(111) by ultrahigh‐vacuum rf glow discharge reactive magnetron sputtering is reported. Epitaxy of GaN is established by x‐ray and electron diffraction. Raman scattering from the epitaxial films consistent with that of wurtzitic GaN is observed. The ion energies involved in the growth process are quantified by measuring the plasma potentials of the Ar/N2 glow discharge by an emissive Langmuir probe technique. As a function of increasing input power, a systematic increase in ion energies and a systematic straining of the GaN lattice are observed. Measured GaN phonon energy scales with lattice strain. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Misfit dislocation generation mechanisms in InGaAs/GaAs heterostructures

    Page(s): 7829 - 7832
    Save to Project icon | PDF file iconPDF (893 KB)  

    An experimental investigation of misfit dislocation generation mechanisms at an InGaAs/GaAs heterointerface is reported. InGaAs epitaxial layers were grown by low‐pressure organometallic vapor‐phase epitaxy on patterned and unpatterned GaAs substrates having etch‐pit densities (EPD) of 200, 1400, and 10 000 cm-2. After epitaxial growth, the samples were annealed at temperatures between 650 and 750 °C, and analyzed by optical and transmission electron microscopy. For the range of substrate EPD studied, it was found that the substrate EPD controls the onset of misfit dislocation generation for low‐temperature epitaxy (≪600 °C) on unpatterned substrates. When epilayers were annealed at 750 °C, the density of misfit dislocations was independent of the substrate EPD. These studies also show that the dominant misfit dislocation generation mechanism for films grown on patterned substrates is nucleation at the growth‐mesa edge. The density of preexisting threading dislocations has little influence on misfit dislocation generation for films selectively deposited within 100×100 μm2 growth windows. For selective heteroepitaxy, misfit dislocation generation strongly depends on the crystallographic orientation of the growth‐mesa edge. © 1994 American Institute of Physics. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.

Aims & Scope

Journal of Applied Physics is the American Institute of Physics' (AIP) archival journal for significant new results in applied physics

Full Aims & Scope

Meet Our Editors

Editor
P. James Viccaro
Argonne National Laboratory