Bibliothek

Notizen: The anisotropy contributions in epitaxial Fe/MnPd bilayers were analyzed in this study. It was found that due to ferromagnetic–antiferromagnetic interfacial exchange coupling, large uniaxial and cubic anisotropy contributions are also induced, in addition to the unidirectional anisotropy. These contributions play an essential role in the magnetization reversal process of the system, in which unusual reversal processes were found upon some fields orientations. © 2000 American Institute of Physics.

Notizen: The magnetic and structural properties of sputtered Co50Fe50(CoFe) films were examined. CoFe films 300 Å thick deposited on Si substrates at room temperature showed large coercive fields of 140 Oe. When similar thickness films were deposited at 100 °C, the coercivity dropped to 90 Oe, and when they were deposited on CoO, the coercivity was reduced to 12 Oe. Cross-sectional imaging with transmission electron microscopy revealed that the CoO underlayer had induced a columnar grain structure in the CoFe, with grain diameters ranging from 50 to 150 Å. CoFe films grown on Si contained larger grains of 200–350 Å in diameter with fewer distinct vertical grain boundaries. Lorentz microscopy showed that domain walls in the hard CoFe film formed complex, fixed patterns in fields less than the coercivity, whereas walls in the CoFe/CoO sample were more conformal and mobile in response to changing fields. Possible structural origins for the wide variation in coercivity obtained with different substrates, deposition temperature, and thickness of CoFe films are discussed. © 2000 American Institute of Physics.

Notizen: Epitaxial La0.35Nd0.35Sr0.3MnO3 (LNSMO) thin films and Pb(Zr0.52Ti0.48)O3(PZT)/LNSMO heterostructures have been grown on LaAlO3 (001) substrates by the pulsed laser deposition method. The oxygen concentration in the LNSMO films is quite sensitive to the deposition oxygen pressure and can be controlled during the fabrication process. It is, however, stable against in situ postdeposition thermal treatments. Consequently, the resistivity and the metal-semiconductor transition temperature of the LNSMO films can be tuned and fixed during film growth. Electrical measurements on the Pt/PZT/LNSMO ferroelectric capacitor show a remnant polarization of ∼35 μC/cm2 and a coercive field of 30–40 kV/cm at low driving voltages. Switching endurance tests suggest no polarization loss up to about 1010 bipolar switching cycles. The advantages of using epitaxial LNSMO films as the semiconducting channel in an all-perovskite ferroelectric field-effect transistor are discussed. © 2000 American Institute of Physics.

Notizen: Co-based nanocrystals encapsulated in carbon have potential applications in ultra-high-density magnetic recording media. In this work, CoxC1−x (x=60, 65, and 70 at. %) nanocomposite films were prepared by pulsed-filtered vacuum arc deposition. Subsequent thermal annealing was performed in vacuum at various temperatures. The films were characterized by non-Rutherford backscattering spectrometry, transmission electron microscopy, Raman spectroscopy, atomic-force microscopy, and magnetic-force microscopy. The as-deposited films were found to be amorphous. After annealing at appropriate temperatures, the films were found to be consisting of hexagonal close-packed nanocrystalline Co grains encapsulated in graphite-like carbon. Clear magnetic-force microscopy images were only observed in those films annealed at sufficiently high temperatures, indicating that there was perpendicular magnetic anisotropy in these films. The magnetic hysteresis loops of the films were measured by a superconducting quantum interference device magnetometer. The optimum annealing temperature for the maximum coercivity was found to depend on the cobalt concentration. For a Co65C35 sample about 20 nm thick after annealing at 350 °C in vacuum for 1 h, the saturation magnetization was 500 emu/cm3, the coercivity was 460 Oe, and the ratio of the remanence to the saturation magnetization was 0.68 at 300 K. Our results are consistent with those of the sputtered Co–C films recently reported in the literature. © 2000 American Institute of Physics.

Notizen: In situ crystalline lead zirconate (PZ) thin films were deposited on platinum metallized silicon substrates, using a 248 nm KrF pulsed excimer laser ablation technique. The antiferroelectricity in PZ thin films was confirmed by means of P vs E and C vs V measurements. The maximum observed saturated polarization (Ps) was 44 μC/cm2 at an applied field of 200 kV/cm. The calculated forward and backward switching fields were 71 and 154 kV/cm, respectively. The dielectric phase transition temperature was at ∼219 °C with zero dc bias and was increased to 290 °C in the presence of 4 V dc bias. Detailed comprehensive study was done on PZ thin films to understand the charge carrier transport with respect to frequency domain of dielectric and ac conductivity measurements. Effect of simultaneous dc bias on the frequency response of the dielectric dispersion and ac conductivity studies were done to understand the nature of charge carrier transport and their activation energies. The calculated activation energies, from ac conductivity and conductance measurements, in the absence of external dc bias were 0.42 and 0.37 eV, respectively. Measurements with the simultaneous application of 4 V dc bias gave two activation energy values 0.38 and 1.76 eV, respectively. The lower and higher activation energies were attributed to the shallow and deep trap energies respectively. © 2000 American Institute of Physics.

Notizen: Schottky barrier photovoltaic detectors have been fabricated on n-AlxGa1−xN(0≤x≤0.35) and p-GaN epitaxial layers grown on sapphire. Their characteristics have been analyzed and modeled, in order to determine the physical mechanisms that limit their performance. The influence of material properties on device parameters is discussed. Our analysis considers front and back illumination and distinguishes between devices fabricated on ideal high-quality material and state-of-the-art heteroepitaxial AlxGa1−xN. In the former case, low doping levels are advisable to achieve high responsivity and a sharp spectral cutoff. The epitaxial layer should be thin (〈0.5 μm) to optimize the ultraviolet/visible contrast. In present devices fabricated on heteroepitaxial AlxGa1−xN, the responsivity is limited by the diffusion length. In this case, thick AlxGa1−xN layers are advisable, because the reduction in the dislocation density results in lower leakage currents, larger diffusion length, and higher responsivity. In order to improve bandwidth and responsivity, and to achieve good ohmic contacts, a moderate n-type doping level (∼1018 cm−3) is recommended. © 2000 American Institute of Physics.

Notizen: The absolute flux, polarization properties, and spatial dependence of the intensity of figure-8-undulator radiation were measured by a gas-scattering method. Absolute-flux measurements were in good agreement with calculated values and the angular divergence measured at 6.6 keV was also close to expected value. Polarization measurements showed oscillations of the degree of linear polarization which varied with photon energy, indicating that both horizontal and vertical polarizations were included. In addition, the polarization plane was found to rotate when the vertical observation angle was changed. These polarization characteristics are not observed in other conventional-undulator radiation and thus unique to figure-8-undulator radiation. © 2000 American Institute of Physics.

Notizen: We study Si–C films (Si content from 0 to 33 at. %) prepared with a pulsed arc discharge method. The structural modifications introduced by annealing up to 1100 °C, irradiation with 53 MeV 127I10+ ions, and deposition onto heated substrates are characterized by Raman spectroscopy. For all the treatments, the structural modifications decrease when the Si concentration increases. Moreover, for high Si content (33 at. %), the energetic iodine ions are found to recover efficiently the structure degraded in preliminary high-temperature annealing. The experiments demonstrate Si-induced stabilization of the three-dimensional Si–C network, which is interpreted as deepening of the corresponding potential energy well. It is highly possible that Si–C films can possess a superior thermodynamic stability for an optimal Si concentration. © 2000 American Institute of Physics.

Notizen: Scanning pyroelectric microscopy was applied to study the local thermal diffusivity in polar perhydrotriphenylene (PHTP) inclusion compounds. A thermal wave technique with special heat flow and pyroelectric detection principles was used to measure the longitudinal (K(parallel)) and transversal (K⊥) thermal diffusivity of needle shaped crystals. As expected, we have found a higher thermal conductivity along the channel direction: K(parallel)〉K⊥. K(parallel) also showed a variation along the needle axis, the K(parallel) values near the capping faces being higher than in the center region. Based on these observations we suggest a relation between the longitudinal thermal conductivity and the local density of orientational defects of the dipolar guest molecules which reaches a maximum in the area around the seed. These findings are in agreement with previous pyroelectric studies, revealing a macrodomain structure and a spatially dependent pyroelectric activity for two different PHTP inclusion compounds. © 2000 American Institute of Physics.

Notizen: This article describes a theoretical approach to carrier-density-fluctuation-induced high-frequency transport noise in n-channel metal–oxide–semiconductor field-effect transistors (MOSFETs) under the condition of nonequilibrium steady-state operation. A partial differential equation describing the charge density fluctuation is formulated assuming the use of the current continuity condition. The model presented herein is applied to the analysis of the high-frequency transport noise in MOSFETs in a higher drain-to-source electric field range than is possible with the state-of-the-art model because the current continuity condition is taken into account. In contrast to the past result, it is shown that the normalized fluctuation power decreases or increases as the drain voltage increases, and it saturates at a certain value. When the drain voltage increases, the ratio of the carrier transit time to the effective relaxation time of fluctuation increases, which promotes the averaging of fluctuation. As a result, the increase in fluctuation is suppressed comparing with past theoretical results. In addition, the dependence of the carrier-density-fluctuation-induced transport noise on gate voltage is considered because both the local channel potential and the local carrier density inside the channel are taken into account in formulating the normalized fluctuation power. It is demonstrated that the normalized fluctuation power increases with gate voltage in the low gate voltage range and saturates at the certain value in the high gate voltage range. When the gate voltage increases, the ratio of the carrier transit time to the effective relaxation time of fluctuation decreases, which suppresses the averaging of fluctuation. © 2000 American Institute of Physics.

Notizen: The deposition of NiCo films with a nominal composition of 50:50 by a single-step, electroless polyol method was investigated. The film thickness increased with increasing deposition time from 5 to 60 min. Measured film compositions showed Ni: Co stoichiometry for films deposited at 5 and 60 min: In the intermediate range of deposition time used, deviation from nominal precursor concentration to Co-rich compositions was observed as a result of change in deposition chemistry. Through-thickness concentration gradient conceivably existed for thicker films as diffusion for compositional homogenization would not be significant at the deposition temperature. The apparent film density increased as a result of coalescence of adjacent particles with increasing film thickness. The particles consisted of nanostructured crystallites. The crystallite size did not significantly change with increasing deposition time, indicating no grain growth had occurred. Nanoscale particles were also observed among the larger particles in the films. All the films showed in-plane magnetization anisotropy, whereas perpendicular magnetic saturation was only achieved for the film deposited at 60 min. The perpendicular coercivity was higher than in-plane coercivity. Possible explanations are suggested. © 2000 American Institute of Physics.

Notizen: Metal matrix composites have many desirable characteristics for precision components, e.g., light weight, dimensional stability, and high thermal conductivity. The understanding of the mechanisms and processes taking place at their production is of crucial importance, since these factors are important for the properties of the composite and its future application. A theoretical model of the processes at diffusion controlled formation of TiB2 in liquid Al under microgravity has been proposed. The main features of the developed model are: mechanisms controlling the formation of the TiB2; degree of distribution of the dispersed TiB2 particles; and mean size and size distribution of the dispersed TiB2 particles. The results obtained with the aid of the proposed model were verified on our recent experiments at the STS95 space shuttle mission. The model predicts correctly the TiB2 particles distribution along the specimens and the characteristic split of the reaction zone observed experimentally. © 2000 American Institute of Physics.

Notizen: The interaction of α-Al2O3 (0001) surfaces with H atoms and N atoms from remote rf plasmas used for the cleaning and nitridation processes, respectively, is investigated at temperatures in the range of 200–600 °C. The chemistry and kinetics of the above processes are monitored in real time by in situ spectroscopic ellipsometry. Also, the chemistry of the nitrided sapphire surfaces is verified by x-ray photoelectron spectroscopy (XPS) analysis. Specifically, H atoms treatments are effective in removing carbon contaminants from the sapphire surface at temperatures of 200–400 °C. Real time ellipsometry is suitable to detect the cleaning end point and to verify the onset of the H-atom diffusion into the sapphire substrate. Remote N2 plasma nitridation at 200 °C is found to yield homogeneous and smooth AlN layers of about 5 Å, after approximately 25 min of nitridation, whereas high nitridation temperatures result in a damaged sapphire surface with AlN protrusions. Both ellipsometric and XPS data show that the sapphire nitridation can be interpreted in the framework of a chemical model, where the formation of NO competes with AlN formation. The chemisorption equilibrium of NO, which strongly depends on surface temperature, is the key factor controlling the nitridation chemistry and kinetics. © 2000 American Institute of Physics.

Notizen: Control of ion energy distribution functions (IEDF) at the substrate during plasma processing is achieved using a specially tailored voltage waveform for substrate bias, consisting of a short voltage spike in combination with a slow ramp. A much narrower IEDF is possible compared to the conventional approach of applying a sinusoidal waveform to the substrate electrode. Measurements in a helicon plasma combined with a time-dependent spherical-shell plasma fluid model demonstrate the benefits of this method in producing a narrow IEDF of precisely controllable energy, independent of ion mass. © 2000 American Institute of Physics.

Notizen: The influence of capillary condensation of water at the atomic force microscope tip and boron carbide film surface on the friction forces and wear of the films is studied by force microscopy. It is shown that friction and wear are strongly dependent on the humidity. At low humidity no wear at the film surface is verified, while at high humidities constant wear rates are measured. The results show that the wear mechanisms of amorphous boron carbide films are highly dependent on the environmental humidity. A wear model based on a film surface interaction with the air moisture induced by the dissipated energy at the microscope tip–surface interface while in relative motion is presented. © 2000 American Institute of Physics.

Notizen: We determine the thermal expansion coefficient of a fluorinated poly(arylene ether) low-k dielectric film using Fourier analysis of x-ray reflectivity data. The approach is similar to that used in Fourier analysis of x-ray absorption fine structure. The analysis compares two similar samples, or the same sample as an external parameter is varied, and determines the change in film thickness. The analysis process is very accurate and depends on no assumed model. We determine a thermal expansion coefficient of 55±9×10−6 K−1 using this approach. © 2000 American Institute of Physics.

Notizen: A series of Pb1−xEuxTe/PbTe multi-quantum well (MQW) samples were grown on (111) cleaved BaF2 substrates by molecular beam epitaxy. The Eu content was maintained at x∼0.05–0.06 and the PbTe well width was varied from 23 to 206 Å. The samples were characterized structurally by high resolution x-ray diffraction in the triple axis configuration. The ω/2aitch-theta scans of the (222) Bragg reflection showed very well resolved satellite peaks up to the tenth-order for all samples indicating that sharp interfaces were obtained. Reciprocal space mapping around the (224) lattice point indicated that the MQW structure tended to the free-standing condition. The (222) ω/2aitch-theta scans were calculated by dynamical theory of x-ray diffraction and compared to the measured ones. Using the in-plane lattice constant as the main fitting parameter, the strain in the PbTe well inside the MQW structure was obtained as a function of its width. It decreased monotonically from an almost fully strained layer to 26% of strain relaxation as the PbTe well increased from 23 to 206 Å. © 2000 American Institute of Physics.

Notizen: In this communication we address two issues essential for low-voltage memory applications of ferroelectric thin films: the size effect on polarization switching, and polarization fatigue. According to the proposed concept, both of these phenomena are controlled by local injection of charge into the interfacial layers of the ferroelectric film. In the experimental part of this work, we show that the entrapped charge relaxation can be enhanced by introducing a thin RuO2 layer into the top interface of the Pt/PZT/Pt ferroelectric capacitor. Capacitors prepared in this way using PZT with a 45/55 Zr/Ti ratio show a substantial improvement of fatigue performance and withstand relatively well the thickness downscaling. As a result, these capacitors exhibited good ferroelectric properties for driving voltage amplitudes as low as 0.6–0.8 V. Our results suggest that control of charge relaxation at the interface is a key issue for development of low-voltage ferroelectric capacitors. © 2000 American Institute of Physics.

Notizen: Oil and gas pipelines often display a bulk magnetic easy magnetization direction, and this may influence results obtained during magnetic nondestructive evaluation. The origin of this magnetic easy axis is not well known. Earlier work [L. Clapham, C. Heald, T. Krause, and D. L. Atherton, J. Appl. Phys. 86, 1574 (1999)] speculated that it was a result of residual stresses introduced during pipeline manufacturing, specifically cold bending. This article reports a study in which a magnetic Barkhausen noise technique is used to follow the development of a magnetic easy axis in a steel plate sample subjected to progressive amounts of cold bending. © 2000 American Institute of Physics.

Notizen: 1.54 μm Er3+ photoluminescent and waveguiding properties of erbium-doped silicon-rich silicon oxide (SRSO) are investigated. Optimum Er3+ luminescence was obtained after an anneal of at least 5 min at 950 °C, and at least 1 at. % excess silicon in SRSO was necessary for the excitation of erbium to be dominated by carriers. The refractive index and the bulk waveguide loss of erbium-doped SRSO film with 0.1 at. % erbium and 1 at. % excess silicon after the optimal anneal treatment was 1.4817 and 4.0 dB/cm, respectively. Fabrication of an erbium-doped SRSO strip waveguide using the standard Si processing techniques and the guiding of internal 1.54 μm Er3+ emission by such a strip waveguide are demonstrated. © 2000 American Institute of Physics.

Notizen: We show that small in-plane magnetization components in magnetic garnet films with perpendicular anisotropy can be imaged using nonlinear magneto-optical proximal probe microscopy, i.e., magnetization-induced second harmonic generation together with a scanning near-field optical microscope. A relationship exists between the in-plane magnetization domains we observed and the typical maze-like out-of-plane magnetization domains that are simultaneously imaged recording the linear Faraday effect. © 2000 American Institute of Physics.

Notizen: Mesoporous silica molecular sieves (MCM-41), which are a mixture of microtubules and hexagonal prisms, were prepared by a hydrothermal self-organizing method under acidic conditions. The nanostructure of silica was characterized by scanning electron microscope and transmission electron microscope. The composition of the sample was analyzed by Infrared spectra, Raman spectra, and energy-dispersed x-ray spectra. The photoluminescence spectrum shows that a broad emission band appears in the wavelength range from 400 to 800 nm, which consists of two broad emission subbands, one at 477 nm and the other at 512 nm. The visible luminescence appearing in calcined MCM-41 silica is ascribed to significant amounts of the twofold coordinated silicon centers (Si20). © 2000 American Institute of Physics.

Notizen: Wei et al. [J. Appl. Phys. 85, 7418 (1999)] performed the room temperature spectroscopic ellipsometry to determine the dielectric function of CdTe1−xSx films. They have fit the obtained dielectric function using the Holden's model dielectric function [Phys. Rev. B 56, 4037 (1997)], and derived conclusions about the line shape at the band gap E0. However, their description of the fitting procedure is ambiguous, and some model parameters in Table I [J. Appl. Phys. 85, 7418 (1999)] are missing which makes it impossible to reproduce their calculations. Furthermore, the results of Wei et al. [J. Appl. Phys. 85, 7418 (1999)] do not represent conclusive proof of the advantages of their approach over other models available in the literature. © 2000 American Institute of Physics.

Notizen: This communication is a response to the Comment on "Optical properties of CdTeS: Experiment and modeling" by A. B. Djurišic and E. H. Li. The shortcomings of a number of schemes to model the optical constants of diamond- and zincblende-type semiconductors, which do not include continuum exciton effects at either the E0, E0+Δ0 or E1, E1+Δ1 critical points (CPs), are discussed. The Holden model, which has been used to fit the CdTeS data, does include these effects, making it possible to extract important parameters, such as the binding energy of the exciton associated with the E1, E1+Δ1 CPs, not deduced by the other approaches. © 2000 American Institute of Physics.

Notizen: In a recent article, Nagayama and Mori [J. Appl. Phys. 84, 6592 (1998)] arrived at several conclusions concerning the thermodynamic nature of shock loaded plexiglass in the low stress range. Their departure point is the kink in the shock velocity-particle velocity curve for this material from which they obtain kinks in the temperature versus shock stress and Gruneisen parameter versus compression curves. These results are in sharp contrast with our conclusions concerning these entities which were published in the early 80's. Our experimental work relied on measuring the temperature of the shock loaded plexiglass using very thin copper thermistors embedded in the specimens. This technique and our main results are reviewed here. © 2000 American Institute of Physics.

Notizen: © American Institute of Physics

Notizen: In the previous article, we showed that a very large Grüneisen parameter for poly(methylmethacrylate) is estimated for a very large slope in shock-particle velocity Hugoniot below the kink point at 150 m/s. This large gamma corresponds to the intermolecular vibrational mode, which is supposed to be excited almost exclusively at the shock front. Rosenberg claimed that his shock temperature measurement by gauge method showed no deflection, which could be observed across the Hugoniot kink, if the above condition is fulfilled. Thermal energy transfer between intermolecular to intramolecular vibrational modes does not occur in the time scale realized in experiments. Then, thermalization between the nonequilibrium polymer region and the gauge material is expected to have similar or longer relaxation time than stress relaxation. Rosenberg's temperature records seem to support this scenario. © 2000 American Institute of Physics.

Notizen: We have investigated the temperature distribution in land/groove phase-change optical disks. The incident beam is linearly polarized either parallel to track (E(parallel) polarization) or tangential to track (E⊥ polarization). Calculations have shown that temperature profiles in the medium are dependent on the wavelength of light, the state of polarization, the geometry of the grooved structure, and the multilayer stack. The temperature profiles are quite different between the land track and the groove track. Thermal cross-track cross talk from a land track to its neighboring groove tracks is higher than that from a groove track to its neighboring land tracks. The interaction between the E⊥ electric field and the grooved structure is mainly responsible for thermal cross-track cross talk. © 2000 American Institute of Physics.

Notizen: Upconversion emission from Tm3+ and Ho3+ ions in a yttrium–aluminum–garnet crystal was observed on resonant excitation of the 3H4 level of Tm3+ and the 5I4 level of Ho3+ with a Ti:sapphire laser at room temperature and 10 K. The upconversion and the energy transfer mechanisms between Tm3+ and Ho3+ ions are discussed. A rate equation model is developed which supported the energy transfer mechanism. © 2000 American Institute of Physics.

Notizen: Optical nonlinearity and dynamic response of a novel inorganic metal cluster (μ3-MoSe4)Ag3(PPh3)3Cl were studied. Large optical limiting effect was observed with the inorganic metal cluster (μ3-MoSe4)Ag3(PPh3)3Cl dissolved in CH2Cl2. We also proved that nonlinear scattering contributes much to its optical limiting performance, which counts for nearly 75% of the total nonlinear energy loss used. The dynamic response of the cluster when pumped at high fluence was tested by a standard time-resolved pump-probe technique. The effective third-order susceptibility χ(3) for this sample was determined in a standard backward degenerate four wave mixing experiment with pump fluence lower than limiting threshold Fth. The conjugate signal intensity tends to saturate when the incident fluence rolls over this value (Fth) in spite of the fact that large optical limiting effect was observed in this fluence region, which implies that the nonlinearity mechanism is different for these two regions. © 2000 American Institute of Physics.

Notizen: Recent studies on two aspects of silica glass as a photonic material will be described. Part A of this review will be focused on structural disorder and structural relaxations in silica glass. With regard to the structural disorder, investigations have been made to improve transparency and to shift the optical absorption edge in the ultraviolet towards shorter wavelengths. Remarkable advances have been achieved in the understanding of both light scattering, which is a dominant factor in the optical losses in silica fibers, and the absorption edge. Freezing of the structural disorder was observed, and structural relaxations are found to be important for improving the transparency, whereas for the absorption edge thermal vibration effects seem to be more predominant than the structural disorder. From the results, the present authors have tried to control the structural relaxation for developing silica glass with an ultimate optical transparency, finding that a very tiny amount of the proper impurity species gives rise to structural subrelaxations, which are effective in reducing the Rayleigh scattering. The scattering was reduced by 13% by addition of only 10 wt ppm Na2O, for example. In part B of this review the second-order optical nonlinearity induced in Ge-doped silica glass will be described based on recent experiments carried out by the group of present authors. A large second-order optical nonlinearity has been successfully induced in the glass by simultaneous applications of a high dc electric field and ultraviolet (UV) irradiation, so-called UV poling. The nonlinearity induced by UV poling in bulk and film samples has achieved a magnitude of χ(2), comparable to or even larger than those of LiNbO3 and other crystals. Surprisingly enough, the nonlinearity induced by this method then decays after the UV poling as an exact single-exponential function of time, very much unlike the usual decay processes observed in glasses. Evidence is presented associating the nonlinearity with GeE′ defect centers created from oxygen deficient vacancies through photochemical reactions. The decay or degradation can be made much slower with the addition of proper impurities which work as electron scavengers. In addition, we have found that crystallites are generated in the glass by the UV poling, which leads to an increase in the third-order nonlinearity, χ(3), approximately 15 times larger than before the treatment. As a whole, the evidence strongly suggests that a major origin of the second-order nonlinearity induced in the glass is a combined effect of a large third-order nonlinearity associated with the crystallites and an internal space-charge field, where the charges to build up the field are produced during the formation of GeE′ centers. © 2000 American Institute of Physics.

Notizen: The effects of As doping, at concentrations CAs≤4.8×1018 cm−3, on the growth kinetics of Si(001):As layers deposited at temperatures Ts=575–900 °C by gas-source molecular-beam epitaxy from Si2H6 and AsH3 have been investigated. With constant AsH3 and Si2H6 fluxes, film deposition rates RSi increase while CAs decreases with increasing Ts. All incorporated As resides at substitutional electrically active sites for CAs up to 3.8×1018 cm−3 (Ts=800 °C), the highest value yet reported for Si(001):As growth from hydride source gases. Immediately following film growth or partial-monolayer As adsorption on clean Si(001), the samples were quenched to 300 °C and exposed to atomic deuterium (D) until saturation coverage. In situ D2 temperature-programmed desorption (TPD) spectra from both as-deposited Si(001):As and As-adsorbed Si(001) layers are composed of β1 and β2 peaks, due to D2 desorption from Si monodeuteride and dideuteride surface phases, together with a new peak β3 which we attribute to desorption from Si–As mixed dimers. Analyses of the TPD spectra show that, because of the lone-pair electrons associated with each As surface atom, the total dangling-bond coverage, and hence RSi, decreases with increasing incoming flux JAsH3 at constant Ts. From measurements of the steady-state As surface coverage θAs vs CAs and Ts, we obtain an As surface segregation enthalpy ΔHs=−0.92 eV. Dissociative AsH3 adsorption on Si(001) was found to follow second-order kinetics with a relatively Ts-independent reactive sticking probability of 0.3. Associative As2 desorption is also second order with a rate constant kd,As2=1×1013 exp(−3.0 eV/kTs). From the combined set of results, we develop a predictive model with no fitting parameters for CAs vs JAsH3, JSi2H6, and Ts. © 2000 American Institute of Physics.

Notizen: By adopting the superposition principle and Fourier integral methodology, this work studies the energy of a dislocation array in a strained epitaxial layer deposited on a finite substrate with the same elastic constants. The total energy comprises the self-energy of the dislocations, the strain energy arising from the lattice mismatch, and the interaction energy between the dislocations and the mismatch. The sum of the self-energy and the interaction energy constitute the dislocation formation energy. Zero formation energy is used as the criterion to determine the epilayer critical thickness hc. No dislocation can appear when the epilayer thickness is below hc. When the epilayer thickness equals the critical thickness and the dislocation density is extremely low, the total energy is independent of the dislocation spacing p. If the critical thickness is less than the substrate thickness and the epilayer thickness is greater than the critical thickness, the total energy has a local minimum at dislocation spacing p=pmin; in addition, the corresponding dislocation density is the critical dislocation density. When p〉pmin, the total energy decreases by decreasing the dislocation spacing, i.e., increasing the dislocation density. The total energy curve near p=pmin changes to a steep valley when the epilayer's thickness approaches that of the substrate thickness. This corresponds to the experimental observation that a fast relaxation of misfit strain occurs when the epilayer thickness grows to a sufficient thickness. If p〈pmin, the total energy markedly increases by decreasing the dislocation spacing. This phenomenon implies that work hardening is inevitable due to the dislocation–dislocation core interaction. © 2000 American Institute of Physics.

Notizen: Empirical molecular dynamics simulations based on the Tersoff potential are carried out for SiC(0001) surfaces and bonded interfaces. It is demonstrated that such a classical interatomic potential is able to correctly describe SiC-4H (0001)3×3 and 3×3R30° surface reconstructions. The surprising accuracy of the empirical simulations compared to results of density functional methods as well as experiments is demonstrated not only by obtaining reasonable structural parameters, but also by the correct prediction of such intricate effects like buckling in the topmost carbon layer of the 3×3 surface and polymerization in the silicon wetting layer of the 3×3 reconstruction. Because of the established good applicability of the Tersoff potential the simulations are used to predict the formation of SiC interfaces to be generated by wafer bonding and so far experimentally unobserved. It is shown that the bond energy crucially depends on the local atomic structure at the interface. The resulting bond energies range from 0.56 up to 3.16 J/m2 depending on the initially prepared reconstruction and alignment of the surfaces. © 2000 American Institute of Physics.

Notizen: Erbium-doped ZnO (ZnO:Er) thin films were fabricated by the KrF excimer laser ablation technique, which is a useful and simple technique to dope Er atoms on the order of 1020 cm−3 into a host material. As-prepared ZnO:Er films showing strong c-axis orientation with a hexagonal crystalline structure indicate a low electrical resistivity of 6.4×10−3 Ω cm. The sharp and intense photoluminescence (PL) at 1.54 μm originating from the intra-4f transition in the Er3+ ions as well as PL in UV region from the ZnO host were observed even at room temperature. Significant distinction arising from the different Er emission centers responsible for the 1.54 μm emission cannot be found in the temperature dependence between the ZnO:Er and Si:Er film as a reference, except for the PL spectrum feature and main PL peak position. This result suggests the existence of Er emission centers in ZnO:Er and Si:Er films that are different from each other. The details of Er-related 1.54 μm emission dynamics of ZnO:Er films have been investigated for the different excitation conditions, where the Er3+ ions have been excited either through a carrier-mediated process in the ZnO host, or through direct pumping into the 4f energy level of the Er3+ ions. There is no change in the 1.54 μm PL spectrum feature in spite of the different excitation conditions, whereas a sensible change can be seen in the rise time of the 1.54 μm emission. The shorter rise time of the 1.54 μm emission observed for indirect excitation implies an excitation efficiency superior to the direct excitation of Er3+ ions. This result indicates that the ZnO:Er thin films are expected to be a promising infrared optoelectronic materials candidate for carrier injection devices because of the high electrical conductivity and high excitation efficiency of the Er3+ ions of an electron–hole-mediated process resulting from ZnO host excitation. © 2000 American Institute of Physics.

Notizen: The energy band gap and lattice constant of the quaternary alloy GaxIn1−xAsySb1−y were determined over the entire composition space (x, y) using a correlated function expansion (CFE) technique to interpolate from observed ternary compound data. Considerable anomalous band gap behavior (i.e., deep bowing) was found, which produced band gap minima as a function of composition. This deep bowing effect was experimentally observed in a quasibinary alloy (GaSb)1−z(InAs)z at low values of z. In addition, the CFE lattice matching relations of the alloy grown on GaSb and InAs were obtained as a function of composition, and the corresponding band gaps were estimated. The CFE estimates were in good agreement with existing experimental data. © 2000 American Institute of Physics.

Notizen: In this article, a study of the proton generation in Si/separation by implantation of oxygen (SIMOX) SiO2/Si structures is reported. The interest was focused on the ability to create mobile charges in SIMOX oxides following a forming gas anneal composed of 92% of N2 and 8% of H2 at 700 °C. The kinetics of the charge generation at 700 °C are presented for two SIMOX oxide thicknesses. A study of the influence of the forming gas pressure is also reported. From both types of experiments, it was concluded that the charge generation mechanism is an interface controlled reaction and that molecular hydrogen H2 or atomic hydrogen H are not the limiting diffusing species in the charge generation. By varying the pressure of the forming gas atmosphere, it was shown that the charge generation is dependent on the square root of the forming gas pressure. From these results and other authors' work, a model is proposed in which the proton generation is a two-step process. © 2000 American Institute of Physics.

Notizen: Epitaxial wide-band gap II–VI beryllium chalcogenide semiconductor films BeTe, BeSe, and BeSeTe were grown on arsenic-terminated silicon substrates. Electrical characteristics of Schottky contacts to the films were measured. Current-temperature measurements at elevated temperatures showed conduction band offsets of 1.2 eV for the BeSe0.41Te0.59/As/Si and 1.3 eV for the BeSe/As/Si heterostructures. At room temperature, the current density through BeSe/Si and BeSe0.41Te0.59/Si films was mid-10−9 A/cm2 at 0.1 MV/cm, similar to previously reported values for ZnS/Si, while BeTe/Si films had orders of magnitude higher current density possibly due to interfacial recombination. © 2000 American Institute of Physics.

Notizen: Proceeding from recently published crystal field parameters, I demonstrate that the first-order spin reorientation transition point in DyFe12−xTix depends strongly on the value of titanium content x. This leads, due to inevitable small spatial variation of composition, to significant broadening of the first-order transition, giving it an appearance of a continuous process. Yet it can be unmistakably identified as a first-order phase transition, thanks to the characteristic shape of the temperature dependence of the spin orientation angle θ. What makes first- and second-order spin reorientation transitions clearly distinct is the shape, rather than width, of the "step" in the θ(T) dependence. Small differences of composition are also found to account for the outstanding discrepancies between the experimental studies carried out on different single crystals of DyFe11Ti, thus bringing the decade-long controversy to an end. © 2000 American Institute of Physics.

Notizen: The effects of postannealing on the structural and leakage current characteristics of Ta2O5 thin films deposited on a TiN/Ti bottom electrode by chemical-vapor deposition at 350 °C were investigated. The Ta2O5 film of 50 nm thickness shows an amorphous phase and smooth surface morphology after annealing in N2O plasma at 350 °C, but the film treated with rapid thermal annealing in nitrogen (RTN) at 750 °C prior to N2O plasma annealing at 350 °C has an orthorhombic structure and rugged surface morphology. In addition, the TiSi phase also appears in the specimen of RTN-treated Ta2O5 film. Moreover, the residual carbon can be effectively eliminated and the oxygen deficiency can be well compensated by the N2O plasma annealing for both amorphous and crystalline Ta2O5 films. The I–V relation of the amorphous Ta2O5 films is asymmetric with respect to the biasing polarity, which indicates that the leakage current is primarily dominated by the Schottky emission. On the other hand, the crystalline Ta2O5 film has a symmetrical I–V relation, suggesting that the non-Ohmic leakage in the crystalline Ta2O5 film is related to the Poole–Frenkel effect. The barrier heights cursive-phiB for the non-Ohmic leakage in the amorphous and crystalline Ta2O5 films, as evaluated from the temperature dependence of the I–V relation, are 0.52 and 0.82 eV, respectively. © 2000 American Institute of Physics.

Notizen: Fatigue rates in the Pb(Zn1/3Nb2/3)O3–PbTiO3 perovskite solid solution were investigated for different compositions in the rhombohedral and tetragonal phases in a variety of directions, at various field levels. It was found that the fatigue rates depend on both the ferroelectric phase and crystallographic orientation. In the rhombohedral phase when the field is applied along the [001]C direction (where the direction is given in terms of the cubic prototype), excellent fatigue resistance was obtained. © 2000 American Institute of Physics.

Notizen: We present experimental results of a Kretschmann-prism coupler method used in Raman scattering experiments with ultrathin (few nanometers) TiO2 sol–gel films deposited on a metallic silver layer characterized by surface plasmon resonance (SPR) experiments. In this article, we describe a new optical device as an application of SPR properties and coupled with a micro-Raman scattering setup. The enhancement factor (150) of the Raman intensity in comparison to the classical way allows us to determine the critical thin films structure. Optical results are then discussed and compared to transmission electron microscopy. © 2000 American Institute of Physics.

Notizen: Waveguides fabrication in Z- and X-cut LiNbO3 crystals by vapor zinc diffusion has been studied by secondary ion mass spectrometry and optical measurements. Compositional analysis of diffused substrates shows the formation of a Zn rich superficial layer, followed by a deeper region with decreasing Zn concentration. A subsequent thermal annealing process has been performed to let zinc ions further diffuse into the substrates. The composition profiles have been correlated with the refractive index profiles, reconstructed by optical reflectivity measurements. The results and the nature of the detected layers are discussed and correlated with previous x-ray studies obtained in the same samples. © 2000 American Institute of Physics.

Notizen: A plasma discharge due to a sustaining pulse of an alternating current plasma display panel was analyzed using a two-dimensional particle-in-cell code, and basic characteristics of the plasma discharge were calculated. The characteristics of the plasma discharge due to a sustaining pulse are as follows. (i) A large amount of space charge remained after drawing the discharge current. This excess space charge did not contribute to wall charge formation or ultraviolet radiation. (ii) The electron energy distribution function was evaluated and could be well fitted by the Druyvesteyn distribution in the high-energy region. The Druyvesteyn distribution was a consequence of the cross section for electron-Ne elastic scattering. (iii) The calculated ultraviolet radiation efficiency η of the plasma discharge due to a sustaining pulse was between 5.51% and 30.7%. Examination of the sensitivity of the efficiency to the electron temperature showed that reducing the electron temperature was a key to improving the efficiency. (iv) A detailed understanding of the conditions for a stable plasma discharge, memory margin, firing voltage, and electrode voltage of the sustaining pulse were obtained from the voltage transfer curve. The voltage transfer curve could be used to optimize the discharge cell design. © 2000 American Institute of Physics.

Notizen: The resistance and tunneling spectra of samples formed by depositing silver electrodes at the two ends of aligned, template-grown, carbon nanotube arrays were measured in the temperature range 0.67–440 K. Two types of samples were fabricated, one with small oxide tunnel junctions separating the carbon nanotubes from the metal electrodes, the other with a significant Al2O3 tunnel barrier. The measurements indicate the presence of three regimes for dI/dV(V). For T〉220 K, dI/dV(V) and the zero-bias conductivity show a broad minimum and an activation temperature dependence suggesting semiconductor behavior. In the temperature range 10〈T〈140 K, the zero-bias conductivity shows a square-root temperature dependence. For T〈2 K, a very steep rise in the zero-bias tunneling resistance is observed with a strong simultaneous suppression of the tunneling conductivity near the Fermi energy. Coulomb blockade is suggested as a plausible explanation of the observed behavior. © 2000 American Institute of Physics.

Notizen: Spin-glass states above room temperature have been found in (Mg, Fe){Mg,Fe,Ti}2O4 spinel ferrite thin films formed on Al2O3 (0001) and SrTiO3 (111) substrates. The films show a long-time relaxation of the magnetization in zero-field-cooled operation and a frequency dependence of the cusp temperature in ac susceptibility measurement, both of which are typical characteristics of a spin glass. This high-temperature spin-glass state has been achieved by the fine tuning of spin states through the control of composition, random oxygen deficiencies, and the stress induced by the film/substrate lattice mismatch, all of which give rise to effects that enhance the exchange interaction of spins in the ferrite films. We have demonstrated the change of magnetic state by means of light irradiation from spin glass to a ferrimagnet over a wide range of temperatures below 290 K. The direct photoexcitation of spins with photon energy in the visible-light region (1.7–3.2 eV), which corresponds to the spin excitation energy of crystal field for Fe ions located in both tetrahedral and octahedral sites, is effective for realizing photoinduced magnetization. © 2000 American Institute of Physics.

Notizen: A detailed formula for the equivalent pyroelectric coefficient of a bimorph structure acting as a pyroelectric detector was deduced. It is shown that electric fields occur in the structure when it is subjected to a temperature variation, due to the unequal charges generated by the two components. The field dependence of the ferroelectric polarization has to be considered in this case and the (∂P/∂E) derivative appears explicitly in the deduced formula. The effect is that the equivalent pyroelectric coefficient has different values for heating and cooling. This could lead to a redressing effect on the pyroelectric signal if the temperature difference varies periodically. The equivalent pyroelectric coefficient is, also, thickness dependent. This fact offers the possibility of designing structures with high values of pyroelectric coefficient compared with the values of the component phases. © 2000 American Institute of Physics.

Notizen: Natural quartz single crystals (α-SiO2) have been exposed to pulsed heavy ion beams (12C, 19F, 32S) with energies of 1 MeV amu−1 in the electronic slowing down regime. The simultaneous recording of the ion fluence and emitted photons with time-resolved spectroscopy experiments allowed the measurement of the "blue luminescence" time decay at 85 K as a function of the fluence at the various electronic stopping power, Se=(−dE/dx)e, of the ions. For all ions, regardless of fluence, the spectra are similar and have two broad bands centered at 1.60 and 2.75 eV with full widths at half maximum around 0.30 and 0.75 eV, respectively. Single-exponential time decay curves are found regardless of Se increasing from 1.4 keV nm−1 (12 MeV 12C) to 5.2 keV nm−1 (32 MeV 32S) across the amorphous track-formation threshold at 2.5±0.5 keV nm−1. At low damaged fractions (≤22%), the decay-time constant ranges between 1.0 and 1.6 ms. The luminescence intensities at zero delay time approximately decrease in an exponential fashion versus fluence with a decay cross section increasing by around one order-of-magnitude at the track-formation threshold, as found in the previous experiments with continuous beams. We analyze to which extent the luminescence decay versus fluence could be due to the quenching of the self-trapped exciton (STE) radiative recombinations by interactions with the ion-induced defects. For this, a STE diffusion model is devised where the STEs recombine nonradiatively at the neighboring cylindrical tracks. The model gives luminescence decay curves versus fluence in good agreement with the experimental data by varying the STE diffusion constant and the amorphous track-core radius in a reasonable range of values. © 2000 American Institute of Physics.

Notizen: A general equation for the susceptibility of disordered systems is proposed. It is based on the experimental observation of power laws at frequencies far from the peak frequency of the imaginary part of the frequency dependent relaxation function, the susceptibility. The obtained general expression contains the equations of other proposed relaxation functions as special cases and, thus, it might be considered as a generalization of these. From this general equation we derive an equation specially adapted for the α relaxation in glass-forming materials. This equation contains only three fit parameters and it is thus very suitable for fitting real experimental data. It is shown that this equation is a good frequency domain representation of the time domain Kohlrausch–Williams–Watts stretched exponential. From the general equation we also derive a four-parameter "universal" equation that describes most types of responses and even inverted response data, i.e., response peaks more stretched on the low frequency side than on the (as is normal) high frequency side. The physical significance of the different parameters is qualitatively discussed and the proposed functions are shown to satisfactorily describe typical experimental data. © 2000 American Institute of Physics.

Notizen: The interdiffusion of Si/Si0.85Ge0.15/Si single quantum well (SQW) structures subjected to inert- and oxidizing-ambient annealing was investigated as a function of temperature (900–1200 °C) and time. Point defect injection allowed modification of the vacancy and interstitial mediated components of interdiffusion, DV and DI. Diffusion profiles of samples processed in inert and oxidizing ambients were similar, which indicates a vacancy-dominated mechanism. Activation energies of diffusion in inert and oxidizing ambients were found to be 5.8 and 5.0 eV, respectively. A fractional interstitial component fI of ∼0.10 was estimated for the lower temperatures, while a significantly smaller fI of ∼0.02 was estimated for the higher temperatures. Experiments using SQWs with buried boron marker layers showed that dislocations in the Si1-xGex trap point defects and affect interdiffusion behavior. © 2000 American Institute of Physics.

Notizen: Thin tungsten nitride (WNx) films were produced by reactive dc magnetron sputtering of tungsten in an Ar–N2 gas mixture. The effects of the variation of nitrogen partial pressure on the composition, residual stress, and structural properties of these films as well as the influence of postdeposition annealing have been studied. The films were analyzed in situ by a cantilever beam technique, and ex situ by x-ray photoelectron spectroscopy, electron energy-loss spectroscopy, x-ray diffraction, and transmission electron microscopy (TEM). It was found that at N concentrations below 8 at. %, the films (typical 150 nm in thickness) were essentially bcc α-W. An amorphous phase was observed in the range of about 12–28 at. % N. When N concentrations reached ∼32 at. % or above, a single-phase structure of W2N was formed. Annealing of the as-deposited films resulted in crystallization of the amorphous or an improved crystallinity of the W2N structure, which was related to the N concentration. Stresses of all W and WNx films were compressive. As the N concentration was increased, the stress decreased and reached its lowest value for amorphous samples near 20 at. % N. Past this point, the compression of films rose again. These results can be ascribed to structural changes induced by the pressure-dependent variation in the average energy of particles bombarding the film during deposition. Cross-sectional TEM studies showed that all crystalline WNx films had columnar microstructures. The average column width near stoichiometry of W2N was ∼20±5 nm near the film surface. © 2000 American Institute of Physics.

Notizen: Studies of pulsed laser annealing (PLA) of CdTe/CdMnTe quantum well structures are made in order to examine depth dependent effects in laser irradiated semiconductors. Since diffusion coefficients are strongly dependent on the temperature, depth resolution is achieved because the diffusion of Mn from the barriers into the quantum wells is depth dependent. Multiple quantum well (MQW) structures of CdTe/CdMnTe were annealed with single pulses from an XeCl laser at 308 nm. At a threshold of 90 mJ cm−2 two new emission bands are observed that are attributed to the diffusion of Mn from barrier layers to QWs. The diffusion associated with these bands, measured as the integrated product of the diffusion constant and time, is found to be 300 and 30 Å2. Calculations of the temperature, reached within the surface following PLA, using an analytical solution of the heat diffusion equation coupled with known high temperature diffusion coefficients predict the diffusion to decrease by one order of magnitude within one period at the top of the MQW stack. It is suggested that at the threshold surface melting occurs and that these emission bands arise from the QWs immediately beneath the melt front. The diffusion of Mn ions into the QWs is confirmed by magneto-optical data. A further emission band occurs at this same threshold with a Mn concentration above that of the concentration in the barrier layers of the MQW stack. This emission is attributed tentatively to the segregation of the Mn ion within the molten region following recrystallization. © 2000 American Institute of Physics.

Notizen: Germanium islands were grown on a sample surface by accumulating atoms from the surrounding area through directional surface diffusion initiated by the electric field of a scanning tunneling microscope (STM). The Ge islands grew with a constant rate determined by the tip–sample bias voltage. The parameters of tip–sample interaction were estimated from the kinetic data for island growth by using a scaling relationship among the growth rate, the dipole moment of atoms on surfaces, and the tip–sample bias voltage. The results show that continuous atom transfer with a STM occurs with a rate significantly higher for Ge than for Si. © 2000 American Institute of Physics.

Notizen: The high strengths of gold thin films on silicon substrates have been studied with particular reference to the possible effect of strain gradients. Wafer curvature/thermal cycling measurements have been used to study the strengths of unpassivated, oxide-free gold films ranging in thickness from 0.1 to 2.5 μm. Films thinner than about 1 μm in thickness appear to be weakened by diffusional relaxation effects near the free surface and are not good candidates for the study of strain gradient plasticity. Our search for plastically induced strain gradients was thus limited to thicker films with correspondingly larger grain sizes. Three related x-ray diffraction techniques have been used to investigate the elastic strains in these films. The standard dhkl vs sin2 Ψ technique has been used to find the average strain through the thickness of the films. The results are consistent with wafer curvature measurements. We have also measured a number of dhkl's as a function of penetration depth to construct depth-dependent dhkl vs sin2 Ψ plots. These data show that the residual elastic strain is essentially independent of depth in the film. Finally, a new technique for sample rotation has been used to measure the dhkl's for a fixed set of grains in the film as a function of penetration depth. Again, no detectable gradient in strain has been observed. These results show that the high strengths of unpassivated gold films relative to the strength of bulk gold cannot be rationalized on the basis of strain gradients through the film thickness. However, a sharp gradient in strain close to the film substrate interface cannot be ruled out. © 2000 American Institute of Physics.

Notizen: The sublimation epitaxy growth process has been studied. The structural quality of the grown layers improves compared with the substrate mainly due to a diminished domain structure misorientation. Optical microscopy shows that the as-grown surfaces are free of typical defects appearing in silicon carbice (SiC) epitaxy, whereas atomic force microcopy measurements show macrosteps. As a possible technique to produce high-quality 4H–SiC, sublimation epitaxy was performed on substrates containing a layer grown by liquid phase epitaxy which is a growth process for closing micropipes in the initial substrate. In spite of the initial surface roughness of the liquid phase epitaxy layer, the surface morphology of the sublimation grown epilayers remained smooth and the structural quality improvement was maintained. This does not occur if the initial surfaces are too rough. A suggestion for roughness reduction is presented. The growth conditions (growth rate ramp up, growth temperature, temperature gradient, source to substrate distance, and substrate surface orientation) leading to the results are presented. A model for the mechanism for structural improvement is outlined and supporting experimental observations are given. © 2000 American Institute of Physics.

Notizen: The effect of intermixing on the apparent interface stress is studied in 〈111〉-textured dc-magnetron sputtered Au/Ni multilayers by use of two methods commonly used for determining interface stress. The method using profilometry and in-plane x-ray diffraction does not take intermixing into account and yields an apparent interface stress of −8.46±0.99 J m−2. However, observed discrepancies between model calculations and measured high-angle x-ray diffractograms indicate intermixing, and by use of the profilometry and sin2 ψ method the real interface stress value of −2.69±0.43 J m−2 is found. This method also reveals a significant and systematic change of the stress-free lattice parameter of both constituents as a function of modulation period which is shown to account for the difference between the two findings. The method using in-plane diffraction is thus shown to be inapplicable to interface stress determinations in systems exhibiting a modulation period-dependent stress-free lattice parameter. Finally, a deviation of the interface stress in the Au/Ni sample with the smallest modulation period as compared to specimens with larger bilayer lengths is observed to be concurrent with a significant decrease in the interface roughness measured by x-ray reflectivity, which suggests that the deviation is of geometrical origin. © 2000 American Institute of Physics.

Notizen: The oxidation of Si catalyzed by 170-nm-thick Cu3Si at elevated temperatures has been investigated by transmission electron microscopy, glancing angle x-ray diffraction, and Auger electron spectroscopy. For wet oxidation at 140–180 °C, the thickness of the oxide was found to increase parabolically with time with an activation energy of 0.4±0.2 eV. The activation energy is close to that of diffusivity of Cu in Si. At 180–200 °C, the growth rate became slower with increasing temperature. The growth of oxide tended to be discontinuous at the surface as the oxidation temperature was increased to a temperature at or higher than 300 °C. The anomalously fast growth of oxide at low temperatures is attributed to the presence of filamentary structures of Cu clusters in the oxide to expedite the diffusion of the oxidants through oxide. At 200–250 °C, more Cu atoms diffuse to the Cu3Si/Si interface and less Cu atoms stay in the oxide, which slows down the oxide growth. The lack of filamentary structures of Cu as diffusion paths retards the growth of SiO2. At 300 °C or higher temperatures, the lack of filamentary structures of Cu clusters stopped the growth of continuous oxide layer altogether. © 2000 American Institute of Physics.

Notizen: Silicides are often used in Si technology for both their ohmic and rectifying properties. In this work, we have compared Zr and Ti germanosilicides as possible metallic contacts on SiGeC alloys in terms of phase formation and stability of the unreacted SiGeC alloy. The germanosilicides are obtained after rapid thermal annealings of Zr or Ti with strained SiGeC layers. The interactions of the metal films with these alloys have been investigated by sheet resistance measurements, x-ray diffraction (XRD), cross-sectional transmission electron microscopy (TEM), and energy dispersive spectroscopy in situ in the TEM. Four crystal x-ray diffraction was performed to measure the residual strain of the unreacted SiGeC epilayer after reaction. The analyses indicate that the final compounds are the C49–Zr(SiGe)2 and C54–Ti(SiGe)2 phases, respectively: In both cases, the compound is formed by monocrystalline grains with various orientations. Nevertheless, neither XRD, nor sheet resistance measurements give any clear information about the C incorporation in the phase, when the reaction occurs with a SiGeC layer. We have observed that the use of Zr completely avoids Ge segregation with an uniform layer formed, while in the case of the reaction with Ti, the grains do not form a continuous layer and Ge-segregation is evidenced: A Ge-rich Si1−z−yGez(Cy) alloy is detected in between the metallic grains. In addition, an early strain relaxation of the unreacted SiGe layer is observed after reaction, and it is much more important after reaction with Ti. During the reaction with nearly compensated SiGeC layers, Zr totally prevents the initial state of strain, while Ti strongly affects the unreacted SiGeC alloy and destroys its initial state. All these results indicate that Zr may be an interesting candidate for realizing germanosilicide contacts on IV–IV alloys, due to its good thermal stability. © 2000 American Institute of Physics.

Notizen: Periodically poled titanium in-diffused lithium niobate (Ti:LiNbO3) seems to be an attractive candidate for applications in the field of integrated optics for wavelength division multiplexed networks it could allow building of polarization-independent devices. In this study, we present the fabrication of reversed ferroelectric domain gratings by direct electron beam writing on both clear LiNbO3 and Ti:LiNbO3. First, a preliminary study was carried out on clear lithium niobate in order to determine the irradiation parameters (voltage, probe current, scanning speed) needed to create inverted ferroelectric domains, and then the same irradiation parameters were used for Ti:LiNbO3. The irradiated patterns were observed after chemical etching using a scanning electron microscope and optical microscopy. HeNe laser diffraction through the samples shows that the irradiation induces a variation of the linear optical refractive index in Ti:LiNbO3 but not in LiNbO3. Moreover, no topographical modification was observed before chemical etching by atomic force microscopy on either sample. © 2000 American Institute of Physics.

Notizen: We applied photoacoustic (PA), photoluminescence (PL), photoluminescence excitation (PLE), and atomic force microscopy (AFM) techniques on porous silicon (PS) layers to study the influence of chemical etching by low-concentration hydrofluoric acid. The chemical etching reveals the formation of PS layers of small dimensions by AFM observations, indicating the possibility of a strong quantum confinement effect. PA spectroscopy is useful to obtain the optical absorption characteristic for strongly scattering media such as PS and it helps to confirm the above speculation by indicating the blueshift of the fundamental absorption edge for the PS layer with chemical etching. PL spectroscopy also confirms the possibility of a quantum confinement effect by revealing the strong intensity and blueshift for the PS layer with chemical etching. PLE measurements suggest that the site for the radiative processes is different from that for the recombination of carriers and the PL of PS layers were dominated only by small crystallites in various size distributions. © 2000 American Institute of Physics.

Notizen: We present the results of a first principles study on the ordered Ge0.50C0.50 and Sn0.50C0.50 cubic alloys. A linear combination of atomic orbitals approach in the framework of density functional theory is employed for total energy calculations in the zincblende phase. A fitting of the energy surface to the equation of state yields the lattice constant of 4.61 and 5.17 Å and the bulk modulus of 181 and 119 GPa for GeC and SnC, respectively. Analysis of band structure suggests a crossover of the nature of the band gap from indirect to direct in going from SiC to GeC to SnC. Although both alloys predicted to be unstable with respect to their elemental components at zero pressure and temperature, GeC appears to become stable at higher pressure. It appears that both the lattice constant and bulk modulus of the ordered alloys do not follow Végard's linear rule, though the calculated dielectric constant of the cubic alloys is approximately the average of the dielectric constant of their elemental components. © 2000 American Institute of Physics.

Notizen: δ(Si)-doped GaAs samples grown by metalorganic vapor phase epitaxy are studied by capacitance–voltage and deep level transient spectroscopy (DLTS) techniques. A detailed analysis of the DLTS signal (including spatial profiles) is performed. DLTS spectra exhibit a clear development depending on the sheet dopant concentration ranging from 5×1014 to 2×1016 m−2. Two observed peaks do not change its activation energy with the doping level while their amplitude increases rapidly when the doping rises. We assign them to defects generated by high silicon concentration, probably related to gallium vacancy. Another peak in the most densely doped sample seems to correspond to the DX level which is occupied near the δ layer. Peculiar features of the EL2 level are observed in δ-doped GaAs and explained by the band bending due to the dopant sheet. No indication of the emission from the quantum confinement states is found in DLTS spectra taken at temperatures 80–400 K. © 2000 American Institute of Physics.

Notizen: Pulsed glow discharge plasma from a 1 Pa gas mixture of acetylene (C2H2) and hexafluoroethane (C2F6) was used to produce fluorinated diamond-like carbon (F-DLC) films on glass and polymethyl-methacrylate (PMMA) substrate. The composition of the F-DLC coatings was measured by using Rutherford backscattering spectroscopy and elastic recoil detection analysis techniques. The transmittance, absorption coefficient, and optical band gap of 100 nm thick F-DLC coatings was measured by using an ultraviolet/visible spectrometer. The friction and wear properties were measured with a conventional pin-on-disk device. In addition, contact angle measurements were taken in order to determine the nonwetting properties of the coatings. The results showed an increase in nonwetting properties, transmittance, and optical band gap with increasing fluorine content in the coatings. The increased fluorine contents suppressed the incorporation of hydrogen and increased the optical band gap energy, which is quite different from the general DLC films whose optical properties are highly improved with increasing amount of hydrogen incorporated in the films. Furthermore, the F-DLC coatings on PMMA and glass substrates proved to have low friction and wear and similar nonwetting properties such as Teflon®.

Notizen: We have measured the temperature dependence of the charge noise power spectral density Sq in two-junction Al–Al2O3–Al single-electron transistors at temperatures from 85 mK to 4 K. Although individual Lorentzians are often visible, the noise spectra are dominated by excess low-frequency noise with a power-law dependence on frequency f where Sq∝1/fβ and β(similar, equals)1. Below about 0.5 K, Sq is weakly dependent on the temperature T. Above 1 K, the charge noise Sq increases with T, and at 4 K Sq(approximate)10−4 e2/Hz at 1 Hz, about a factor of 100 greater than at 85 mK. © 2000 American Institute of Physics.

Notizen: The electrical characteristics of highly dense SnO2 ceramic varistors are believed to be caused by the existence of potential barriers at the grain boundary. A complex plane analysis technique (to eliminate the influence of trapping activity associated with the conductance term observed via depression angle of a semicircular relaxation in the complex capacitance plane), allied with an approached Mott–Schottky model, are used to demonstrate that the potential barriers at the grain boundary are Schottky-type barriers in SnO2 varistors such as those observed in the traditional ZnO varistor. © 2000 American Institute of Physics.

Notizen: We have verified the accuracy of free-charge determinations from Raman scattering in doped semiconductors by comparing the results obtained from phonon–plasmon coupled-mode line-shape fits with the charge-density values extracted from the analysis of the Shubnikov–de Haas oscillations. The experiments were carried out on n-InP layers, and conduction band nonparabolicity was included both in the Lindhard–Mermin model used to fit the Raman spectra and in the Shubnikov–de Haas analysis. We find a very good agreement between Raman and magnetotransport results, which confirms the reliability of the charge-density determination from Raman-scattering measurements when the line-shape analysis is carried out using the Lindhard–Mermin model. © 2000 American Institute of Physics.

Notizen: The electroabsorption (EA) of CdS0.1Se0.9 nanocrystals was measured under pulse-like modulated fields and recovered both with similar frequency (1f ) and with double frequency (2 f ) of the applied field. The EA properties and the effects of applied field form and recovered frequency on the EA properties were studied. Electric field induced strong change of absorption (10−3) was observed in resonant and nonresonant regions. The EA signal in the resonant region was due to the quantum confined Stark effects and the signal in the nonresonant region was due to the change of local field that resulted in the change of refractive index, the change of absorption. The quadratic dependence of EA signal intensity on the electric fields indicated that the nanocrystals had third nonlinear optical susceptibility. The EA signal intensity measured with 1f mode was about ten times of that measured with 2 f mode. Perhaps the lower response measured with 2 f mode was due to the measured mode and the different response coefficient in different frequency regions. © 2000 American Institute of Physics.

Notizen: Photoreflectance (PR) spectroscopy, x-ray diffraction (XRD) method, and secondary ion mass spectroscopy (SIMS) were used to observe the radial distribution of the band-edge transition and modification of the band structure due to the variation of indium in 3 in. indium-alloyed semi-insulating GaAs (InxGa1−xAs) grown by the liquid-encapsulated Czochralski method. The data from room temperature PR measurements showed the variation of the transition energy with positions indicating the radial distribution of the indium content across the wafers; indium content being higher around the edge region than the central area. The splitting of the degenerate valence band around the edge region of the wafers was also shown in PR data due to the different indium content in adjacent regions where indium content varies rapidly. The XRD measurements showed the drastic change in the distribution of lattice constant on where the splitting of the heavy and light holes happened and the SIMS analysis was adopted to confirm the distribution of indium content across the wafer. The possible model was proposed from the experimental data. © 2000 American Institute of Physics.

Notizen: This work attempts to determine how percolation at an equilibrium state is correlated to percolation under experimental conditions. The dynamic process of forming conductive networks in carbon-black (CB)-filled poly(methyl methacrylate) composites was investigated by real-time tracing the time dependence of electrical resistivity during isothermal treatments. It was observed that the dynamic percolation curves maintains the same shape and shift to a shorter percolation time with increasing annealing temperature and filler concentration. An Arrhenius plot of the shift factor against the annealing temperature shows a linear relationship, irrespective of the filler concentration, and the activation energy of the percolation time is close to the activation energy of the zero-shear-rate viscosity of the polymer matrix. Furthermore, an increase in the thermodynamic interactions between CB and the polymer matrix causes a large reduction in polymer mobility, resulting in an increase in the percolation time. These results lead to the conclusion that percolation is delayed by the bulk mobility of polymer layers surrounding CB particles. An experimental approach for determination of the retardation time is proposed based on theoretical analysis of the dynamic movement of the carbon particles. It is suggested that the difference in the kinetic history with respect to percolation among different composite systems can be eliminated by normalizing the experimental conditions to the same value of retardation time. © 2000 American Institute of Physics.

Notizen: The conventional method of semiconductor charge carrier transport investigations using full band ensemble Monte Carlo simulations is extended to allow for tunneling between bands during accelerated drift of the carriers. The essentially classical picture of transport, as simulated, is preserved by implementing a stochastic selection of the band index of the initial state of each scattering process associated with phonons, with impurities, or with impact ionization. Relative probabilities for the band assignment are calculated from the overlap integrals of the cell-periodic parts of Bloch wave functions belonging to different bands, for k-vectors along the carrier k-space trajectory between successive scattering events. As an example, the method is applied to Monte Carlo transport simulations for holes in 4H SiC in a homogeneous applied electric field. Tunneling between valence bands during the drift phases is shown to have a significant impact on the carrier energy distributions when large electric fields are applied, and on physical parameters that directly depend on the carrier energy, such as the hole initiated impact ionization coefficient. © 2000 American Institute of Physics.

Notizen: Gate oxide degradation induced by electron-beam irradiation has been studied. A large increase in the low-field excess leakage current was observed on irradiated oxides and this was very similar to electrical stress-induced leakage currents. Unlike conventional electrical stress-induced leakage currents, however, electron-beam induced leakage currents exhibit a power law relationship with fluency without any signs of saturation. It has also been found that the electron-beam neither accelerates nor initiates quasibreakdown of the ultrathin gate oxide. Therefore, the traps generated by electron-beam irradiation do not contribute to quasibreakdown, only to the leakage current. © 2000 American Institute of Physics.

Notizen: We present the measurements and the theoretical model of the frequency-dependent noise temperature of a superconductor lattice-cooled hot-electron bolometer mixer in the terahertz frequency range. The increase of the noise temperature with frequency is a cumulative effect of the nonuniform distribution of the high-frequency current in the bolometer and the charge imbalance, which occurs at the edges of the normal domain and at the contacts with normal metal. We show that under optimal operation the fluctuation sensitivity of the mixer is determined by thermodynamic fluctuations of the noise power, whereas at small biases there appears additional noise, which is probably due to the flux flow. We propose the prescription of how to minimize the influence of the current distribution on the mixer performance. © 2000 American Institute of Physics.

Notizen: Photoemission from GaAs cathodes with negative electron affinity (NEA) is applied for producing electron beams with very low longitudinal and transverse velocity spread. GaAs transmission-mode cathodes were activated with Cs and either O2 or NF3 in an extremely high vacuum setup (base pressure below 10−12 mbar). Quantum efficiencies of 20%–25% (at 670 nm) and long dark lifetimes (about 1000 h) could be achieved for both types of activation in a reproducible way. Using a method based on the adiabatic transverse expansion of an electron beam in a spatially decreasing magnetic field, the mean transverse energy (MTE) of the photoemitted electrons was measured systematically, recording longitudinal energy distribution curves. Both the MTE and the longitudinal energy spread strongly depend on the value of NEA and the position of the extracted distribution relative to the bulk conduction band minimum. Electrons with energies above the conduction band minimum are thermalized with the lattice temperature of the cathode, while electrons with energies below this level show a non-Maxwellian distribution with enhanced transverse energies. Thus, when extracting all electrons in a current limited emission mode, the MTE increases with the absolute value of NEA and reaches values up to (approximate)100 meV. By cutting off the low energy electrons with an external potential barrier, the longitudinal as well as transverse energy spread of the extracted electron ensemble are reduced. The MTE could be reduced down to about 28 meV at room temperature and to about 14 meV at liquid nitrogen temperature. The behavior of the MTE was found equivalent for (Cs, O) and for (Cs, F) activation layers on the same cathode. Conclusions about energy loss and scattering in the emission of photoelectrons from NEA GaAs cathodes are discussed. © 2000 American Institute of Physics.

Notizen: We calculate the second-order streaming force in a fluid in the vicinity of the solid/fluid interface for two systems of importance in the technology of megasonic cleaning of silicon wafers. The first system consists of a single planar interface between a solid elastic medium representing silicon and a viscous fluid, namely water. The second system accounts for the finite thickness of silicon wafers. It is composed of one silicon slab (wafer) immersed in water. The components of the streaming force parallel and normal to the silicon/water interface are determined as functions of frequency and wave vector of the incident acoustic wave. The normal component of the streaming force is used to calculate the removal force defined as the net force perpendicular to the solid/fluid interface acting on a spherical contaminant particle adhering to the silicon surface. The removal force is too small to remove submicron particles. In contrast the streaming force parallel to the solid/fluid interface may remove particles by pushing or rolling them. The streaming force is shown to be very sensitive to the angle the incident acoustic wave makes with the silicon/water interface. © 2000 American Institute of Physics.

Notizen: Nickel–silica nanocomposites were prepared by the sol-gel route. By subjecting these to an oxidation treatment in the temperature range 723–1023 K, nickel oxide films of estimated thicknesses in the range 0.5–1.5 nm were grown on the nickel nanoparticles. dc electrical resistivity was measured in the temperature range 300–570 K. The data indicate that an amorphous phase in the interfacial region of the oxide-coated nickel nanoparticles determines the electrical conduction. A small polaron hopping conduction is found to be operative. The humidity sensing properties of these nanocomposites were measured at 300 K in the range of relative humidity (RH) from 3% to 87%. Electrical conductance in all specimens showed an increase of about three orders of magnitude as the RH was raised. Specimens subjected to higher oxidation treatment showed a higher conductance for a specific value of RH. This is ascribed to the presence of a larger number of Ni3+ ions in such specimens. © 2000 American Institute of Physics.

Notizen: The surface roughening of a silicon wafer during annealing in argon at an extremely low oxygen partial pressure, which restricts the conditions for the suppression of oxidation-induced stacking faults, was investigated. It was found that surface roughening occurs when the oxygen partial pressure is lower than 3×10−5 atm. The roughening is brought about by the generation of square pits and the anomalous growth of film. Fiber-like material is, sometimes, formed in the periphery of the wafer. It was found that grown film and the fiber-like material are SiO2 and that the pits have the shape of an inverted pyramid. The mechanism of this roughening is discussed with considering a boundary layer. As a result, the formation of the inverted-pyramidal pits is attributed to the occurrence of SiO evaporation in the boundary layer, where the gas flow is stagnant and the oxygen partial pressure is extremely low. The four exposed surfaces at the bottom of a pit have a {111} surface orientation. The anomalous growth of the oxide and the formation of the fiber-like material were interpreted to be due to the reoxidation of the evaporating SiO at the edge of the boundary layer and its redeposition in the form of SiO2 on the wafer. It was concluded that the inclusion of oxygen at a partial pressure higher than 4×10−5 atm is important to prevent surface roughening during annealing even in an inert gas like argon. © 2000 American Institute of Physics.

Notizen: Multilayered thin films consisting of high electrical conductivity copper layers sandwiched between pairs of low coercivity ferromagnetic alloys can exhibit giant magnetoresistance. The magnitude of the magnetoresistance increases with the structural and chemical perfection of the interfaces. Recent atomistic modeling and experimental observations have shown that nickel and cobalt atoms in the ferromagnetic layer readily exchange with underlying copper atoms during the deposition of the ferromagnetic layer upon the copper spacer. This results in mixing at the ferromagnetic metal on copper interface. Low energy (1–20 eV) inert gas ions can be used during deposition to flatten the surface of layers, in some cases without causing interlayer mixing. Here we use the molecular dynamics simulation method to investigate the effects of the assisting ion fluence upon the surface roughness and interlayer mixing of a model Ni/Cu/Ni multilayer system. The results reveal that the surface roughness initially drops rapidly with ion fluence and then approaches a limiting roughness that is dependent upon the surface type, the ion energy, and the ion mass. For a Cu on Ni surface irradiated by 2.0 eV Xe+ ions, the flattening transition occurs at a fluence of about 0.2 ions/Å2 (corresponding to an ion to metal deposition flux ratio of about 5). The same transition was seen at a similar fluence for a Ni on Cu surface, but at a higher Xe+ ion energy of 14.0 eV. Threshold energies for flattening and mixing were identified for various surfaces. The probabilities of both flattening and mixing were found to increase with ion fluence and ion energy. Because the threshold energy for mixing was lower than that for smoothing, significant interfacial mixing was only seen during ion assisted flattening of the Ni on Cu interface. Simple models have been developed to establish the functional dependence of interfacial structural parameters upon the assisting ion fluence. © 2000 American Institute of Physics.

Notizen: Resonant photoluminescence from Si1−xGex alloy nanocrystals as small as 4 nm in diameter embedded in SiO2 thin film matrices was studied. In pure Si nanocrystals, phonon structures were clearly observed, indicating that optical transitions are assisted by momentum-conserving phonons. These structures are quenched by adding a small amount of Ge atoms in Si nanocrystals, i.e., due to the formation of Si1−xGex alloy. Furthermore, the lifetime of the radiative recombination became much faster in doped nanocrystals. These results suggest that Si1−xGex alloy formation leads to a disorder in the translation invariance of the crystalline lattice, thus resulting in the breakdown of the k-conservation rule. © 2000 American Institute of Physics.

Notizen: High-energy hole scattering rates for a full-band Monte Carlo simulation in Si are verified using a quantum-yield experiment. We compare two models that yield the correct drift velocity and ionization coefficient but quite different energy distributions. It is demonstrated that the quantum-yield experiment provides a means for monitoring hole scattering rates in Si; the model based on the ab initio impact ionization rate shows good agreement with the experiments, while the random-k approximation proposed by Kane overestimates the ionization rate of holes near the threshold energy. © 2000 American Institute of Physics.

Notizen: Highly resistive molecular beam epitaxial GaN layers are characterized by temperature dependent conductivity and Hall effect measurements. Seven n-type GaN samples with room temperature layer resistivity ranging between 8 and 4.2×106 Ω cm are used in this study. The experimental data are analyzed by considering various transport models such as band and hopping conduction, scattering on charged dislocations and grain boundaries controlled transport. The same defect level of 0.23 eV, attributed to nitrogen vacancy, is found for layers with ρ300≤3.7×103 Ω cm. The Hall mobility for two lower resistivity layers is influenced mainly by phonon scattering (μH∼Tx, x=−1.4). However, higher resistivity layers show positive mobility power, x=0.5–0.9, which can be explained by dominating scattering on charged dislocations. Properties of layers with the highest resistivity (1×105 and 4.2×106 Ω cm) and extremely low Hall mobility (6 and 〈0.1 cm2 V−1 s−1) are consistent with grain boundary controlled transport. The barrier height between grains of 0.11 eV and an average grain size of 200 nm are found. Neither nearest-neighbor or variable range single phonon hopping nor multiphonon hopping can be clearly attributed to the conduction of the layers investigated. © 2000 American Institute of Physics.

Notizen: The modification of the elastic properties of giant magnetostriction alloy films due to an applied magnetic field (the ΔE effect), has been studied. Two different types of films were deposited on Si substrates: (i) single layers of TbDyFeCo alloys typically 1000 nm thick and (ii) nanocomposite multilayer films of FeCo/TbFeCo each having a typical thickness of 6 nm. Both types of films were rendered magnetically anisotropic with a well defined in-plane easy axis. Rectangular samples were cut out of these bimorphs and firmly glued at one end to a heavy base to form a simple cantilever structure. The variations of film elastic moduli were deduced from the shifts of the cantilever resonance frequencies as a function of bias field for two basic configurations: (i) field applied along the easy axis and (ii) field applied along the hard axis. In contrast with previous work, both flexural and torsion resonance modes were excited and studied. As a result the field induced variations of both planar traction modulus and the shear modulus were obtained and new interesting features were discovered. In particular strongly negative values of the shear modulus were observed (at least in the nanocomposite films) in the vicinity of the divergence in the transverse magnetic susceptibility at saturation field along the hard axis. A simple but complete theoretical analysis shows that the uniaxial anisotropy model together with the assumption of isotropic magnetoelastic coupling gives a good semiquantitative understanding of all the experimental results. © 2000 American Institute of Physics.

Notizen: The energy distribution of electrons that were transported through a thin intrinsic AlN film was directly measured as a function of the applied field. The measurements were realized by extracting the electrons into vacuum through a semitransparent Au contact and measuring their energies using an electron spectrometer. At moderate applied fields (100 kV/cm), the energy distribution was found to follow a Maxwellian model corresponding to a temperature of 2700 K and a drift component below the spectrometer resolution. At higher fields, intervalley scattering was evidenced by the presence of a second peak at 0.7 eV. This coincides well with the energy position of the LM valleys in AlN. © 2000 American Institute of Physics.

Notizen: An enhancement of the giant magnetoresistance effect is investigated in spin valves where oxide layers, which are formed from magnetic layers, are incorporated in the structure. Information about Co–Fe based nanooxide layer (NOL) is obtained via x-ray photoelectron spectroscopy and Rutherford backscattering spectrometry. Cross-section transmission electron microscopy is also used to explore the effect of an NOL on the polycrystalline structure of the spin valve. © 2000 American Institute of Physics.

Notizen: A two-dimensional ordered structure can be formed in a thin film of homogeneous magnetic fluids by applying a magnetic field perpendicular to the plane of the film. The characteristic spacing of the ordered structure can be controlled by several parameters, such as the field strength, the sweep rate, the thickness of the film, and the volume concentration of homogeneous magnetic fluids. The length of the characteristic spacing ranges from several micrometers to submicrometers. Depending on the size of the spacing, magnetic fluid films are capable of diffracting an incoming white light. The diffracted light then interferes resulting in a chromatic effect. This effect is well illustrated by the chromatic rings produced by passing a white light through the magnetic fluid films possessing an ordered structure. Because of the changeability of the characteristic spacing, the magnetochromatic effect of magnetic fluid films can be controlled by altering these control parameters. © 2000 American Institute of Physics.

Notizen: Two sharp bands at ∼1350 and ∼1600 cm−1 were observed in the Raman spectra of carbon-implanted GaN after postimplantation annealing treatments. The intensities of these two bands increased while their full widths at half maximum decreased with increasing annealing temperature. The observation of these two bands indicates the formation of microcrystalline graphite in C-implanted GaN. Hall measurements demonstrated that some dispersed C in GaN acted as acceptors and played a role in reducing electron concentration and Hall mobility. The facts that in 1100 °C furnace annealing the intensities of these two Raman peaks decreased rapidly to zero and the resistivity increased by 3 orders of magnitude indicate the dissolution of microcrystalline graphite at this temperature. © 2000 American Institute of Physics.

Notizen: A study of the electron paramagnetic resonance of Fe-based nanoparticles embedded in polyethylene matrix was performed as a function of temperature ranging from 3.5 to 500 K. Nanoparticles with a narrow size distribution were prepared by the high-velocity thermodestruction of iron-containing compounds. A temperature-driven transition from superparamagnetic to ferromagnetic resonance was observed for samples with different Fe content. The unusual behavior of the spectra at about 25 K is considered evidence of a spin-glass state in iron oxide nanoparticles. © 2000 American Institute of Physics.

Notizen: The curves depicting the temperature dependence of the global depolarization curves of poly(2-[naphthylacetyl)oxy]ethyl acrylate), poly(2-[naphthylacetyl)oxy]ethyl methacrylate), and the model compound of the side groups of these polymers, 2-(acetyloxy)-ethyl-(2-naphthylacetate), present ostensible glass-rubber relaxations located at −45, 34, and 53 °C, respectively. These values are very close to the respective calorimetric glass transition temperatures of these materials. As these substances exhibit a weak dielectric activity in the glassy state, it was possible to determine the components of the complex dielectric permittivity at very low frequencies from partial depolarization curves obtained on these materials in this state. Possible mechanisms responsible for the subglass relaxations are suggested from the comparison of the relaxation behavior of the model compound and the polymers. © 2000 American Institute of Physics.

Notizen: Thin film electroluminescent (TFEL) structures composed of ZnS:Mn (phosphor) and Y2O3 (insulator) films were annealed using pulsed (∼20 ns) KrF laser irradiations under an environment of 150 psi (∼10.34 bars) of argon. The crystallinity of both layers was studied via x-ray diffraction analysis. Their degree of lattice misfit and the integrated diffraction line intensities were examined to assess the thermal effects of pulsed laser annealing. In agreement with a thermal model of the laser–matter interaction and previous results, we suggest that the pulsed laser annealing technique improves the luminescent efficiencies of thin film electroluminescent phosphors by generating an in-depth solid-state phase transition. We report that the laser processed TFEL structure exhibits better display performance than equivalent devices thermally annealed at a temperature of 450 °C as demonstrated by sharper turn-on response and also an increase in brightness by a factor greater than four times. We attribute these performance improvements to a smaller lattice misfit value, which has been previously correlated to interface states density. © 2000 American Institute of Physics.

Notizen: In this article, we discuss the utilities of corrugated quantum well infrared photodetectors (C-QWIPs) in detector material characterization. By measuring the detector responsivity as a function of corrugation period, several important detector parameters, such as the absorption coefficient α of parallel propagating light and the energy resolved photoconductive gain g, can be directly deduced. For the QWIP material presented, α at the peak was found to be 0.21 μm−1 under the usual operating condition. This value of α corresponds to an absorption length of 4.8 μm. Instead of being a constant, the value of g also varies significantly across the excitation spectrum, and the peak value is larger than the noise gain at large bias. Our results show that the present characterization technique is capable of providing accurate and detailed information on the intrinsic properties of QWIP materials under actual operating conditions. It is extremely useful in detector optimization. In addition, we also show the characteristics of C-QWIPs with an additional vertical trench at the center of each corrugation to gain more insights into the distribution of light intensity in a C-QWIP structure. © 2000 American Institute of Physics.

Notizen: The complex photonic band gap (PBG) structure of frequency-dependent metallic–dielectric systems, using actual measured values for index of refraction, has been computed using nonlinear numerical techniques. Transmissivity and reflectivity were also computed and compared with the PBG structure. For application to PBG systems consisting of alternate layers of Ag and MgF2 over a range of frequencies from the near ultraviolet to the near infrared, we find that features of the PBG structure, transmissivity, and reflectivity are distinctly different from those computed using the Drude model. These features include plasma resonant structure not included in the Drude model. © 2000 American Institute of Physics.

Notizen: A type of luminescence was observed from fluorescent dye-doped polymer films by corona discharge excitation. The polymer film was deposited on a transparent electrode, and direct current corona discharge was caused in argon atmosphere using this electrode as the anode. Emission of three elemental colors, i.e., red, green, and blue, were obtained from polyvinilcarbazole films doped with 4-(dicyanomethylene)-2-methyl-6-(p-dimethylaminostyryl)-4H-pyran, 3-(2′-Benzothiazolyl)7-diethylaminocoumarin, and 1,1,4,4-tetraphenyl-1,3-butadiene, respectively. The drive current required for the luminance of 10 cd/m2 was as low as 100 μA/cm2 and the maximum quantum efficiency was 9.95 cd/A, which is nearly four orders of magnitude higher than that of the common injection type electroluminescent cells with the same material. The origin of this high efficiency can be attributed to the efficient hole blocking by the gas phase and also by concurrent effects of different luminescence processes including electroluminescence by charge recombination, photoluminescence by ultraviolet excitation, and cathodoluminescence by electron impact excitation. © 2000 American Institute of Physics.

Notizen: The convenient basis representation for the valence band Hamiltonian in wurtzite semiconductors is proposed and the polarization selection rules in GaN-based quantum wells are found. The polarization dependent optical matrix elements in quantum wells are calculated for a finite in-plane electron momentum, that allows to calculate peak optical gain and laser threshold in the GaN-based quantum well structures. This article also addresses the role of alloy inhomogeneities in the optical gain in quantum well lasers. The peak gain is found as a function of statistical broadening caused by the In-content fluctuations. It is shown that the composition fluctuations in the active region influence the gain differently depending on a light polarization. The absolute value of the calculated gain is found to be in agreement with the experiment. © 2000 American Institute of Physics.

Notizen: Deep level transient spectroscopy (DLTS) measurements have been made on GaAs n+p diodes containing GaSb self-assembled quantum dots and control junctions without dots. The self-assembled dots were formed by molecular beam epitaxy using the Stranski–Krastanov growth mode. The dots are located in the depletion region on the p side of the junction where they act as a potential well that may capture and emit holes. Spectra recorded for temperatures between 77 and 440 K reveal several peaks in diodes containing dots. A control sample with a GaSb wetting layer was found to contain a single broad high temperature peak that is similar to a line found in the GaSb quantum dot samples. No lines were found in the spectra of a control sample prepared without GaSb. DLTS profiling procedures indicate that one of the peaks is due to a quantum-confined energy level associated with the GaSb dots while the others are due to defects in the GaAs around the dots. The peak identified as a quantum-confined energy level shifts to higher temperatures and its intensity decreases on increasing the reverse bias. The activation energy for the quantum-confined level increases from 400 meV when measured at a low reverse bias to 550 meV for a large reverse bias. Lines with activation energies of 400, 640, and 840 meV are associated with defects in the GaAs based on the bias dependence of their peak positions and amplitudes.

Notizen: A simple complementary metal–oxide–semiconductor compatible process for the preparation of very thin (1–5 nm thick) poly-Si layers embedded in SiO2 is presented. The process consists of repeated cycles of poly-Si deposition, oxidation, and wet etching steps. Periodic structures with up to three Si/SiO2 layers were grown using this process. Transmission electron microscopy analyses show that the layered structure can be conserved down to a Si layer thickness of 2 nm. For thinner layers the resulting structure is more granular like. Samples with a Si-layer thickness lower than 3 nm show room temperature photoluminescence at about 1.55 eV that shifts to higher energies when the thickness is further reduced. The maximum shift obtained with respect to the c-Si band gap is 0.55 eV. Intensity of the photoluminescence as a function of temperature shows a behavior similar to the one observed for 0 and one-dimensional Si structures. On the basis of the thickness dependence, the temperature dependence and the saturation studies, this emission is attributed to recombination of electron–hole pairs in quantum confined Si. © 2000 American Institute of Physics.

Notizen: Fex(SiO)1−x thin films, with volume fraction x ranging from 0.04 to 0.6, have been prepared by coevaporation. Magnetization, resistivity, and magnetoresistance measurements show that the behavior of this composite is metallic like for x≥0.35. For 0.2≤x≤0.27, the confrontation of all measurements prove the existence of a tunneling giant magnetoresistance between ferromagnetic grains at T〈300 K, where the mean radius of the grains is estimated to be about 1 nm. © 2000 American Institute of Physics.

Notizen: Intense green luminescence associated with the 5S2–5I8 transition of Ho3+ in holmium doped LiTaO3 crystals has been excited at room temperature and at 15 K by a 647.1 nm line of a krypton ion laser. Based upon the analysis of excited states relaxation dynamics and of the rise of upconverted luminescence it is concluded that the mechanism involved is an excited state absorption from the long lived 5I7 level. Efficient single wavelength excitation of upconverted luminescence in LiTaO3:Ho is due to the advantageous coincidence of transition energies of the ground and excited state absorptions. © 2000 American Institute of Physics.

Notizen: The nucleation and growth of Hg1−xCdxTe on CdTe(211)B/Si(211) substrates by molecular beam epitaxy are comprehensively studied by in situ reflection high energy electron diffraction and transmission electron microscopy. Microtwins are observed to be formed at the interface, but are overgrown later as the growth proceeds. It is shown that the three-dimensional growth at the nucleation stage of HgCdTe on CdTe(211)B/Si and CdZnTe(211)B is more likely due to the higher surface energy of Hg1−xCdxTe than that of CdTe(211)B and CdZnTe(211)B. © 2000 American Institute of Physics.

Notizen: In this reply, the authors show that the argument by Scott regarding the band gap of bulk SrBi2Ta2O9 (SBT) is not based on concrete evidence. The authors will also show additional data from a Raman study of a powdered SBT sample to prove that the surface of the specimen was not covered by Bi2O3. © 2000 American Institute of Physics.

Notizen: Optical characterization methods were applied to a series of microcrystalline silicon thin films and solar cells deposited by the very high frequency glow discharge technique. Bulk and surface light scattering effects were analyzed. A detailed theory for evaluation of the optical absorption coefficient α from transmittance, reflectance and absorptance (with the help of constant photocurrent method) measurements in a broad spectral region is presented for the case of surface and bulk light scattering. The spectral dependence of α is interpreted in terms of defect density, disorder, crystalline/amorphous fraction and material morphology. The enhanced light absorption in microcrystalline silicon films and solar cells is mainly due to a longer optical path as the result of an efficient diffuse light scattering at the textured film surface. This light scattering effect is a key characteristic of efficient thin-film-silicon solar cells. © 2000 American Institute of Physics.

Notizen: We present an investigation of intersubband emission at far-infrared wavelengths from semiconductor superlattices with parabolically graded quantum wells. Light emission is produced by sequential resonant tunneling injection of electrons into higher energy levels of the quantum wells and subsequent radiative decay. The current versus voltage curves of these devices exhibit negative differential resistances characteristic of the sequential resonant tunneling injection. A single, narrow emission peak is observed from the superlattices with parabolic quantum wells demonstrating radiative decay through multiple evenly spaced energy levels. When a chirped superlattice acting as an electron energy filter replaces the barrier in each period of the structure, clearer resonances are observed in the current–voltage characteristics and more efficient injection is achieved. © 2000 American Institute of Physics.