ALBERT

Notes: This work studies the ponderomotive forces occurring in liquid conductors with macroscopic inclusions. It is shown that if the electric conductivity of inclusions is higher than the electric conductivity of the host medium then the inclusions are attracted to the axis of a conductor. In the opposite case the ponderomotive forces repel the inclusions towards the surface of a conductor. It is shown that if a certain relation between an amplitude and a frequency of an electric current is satisfied, the parametric resonance in the oscillations of the inclusions with electric conductivity higher than that in the bulk of the conductor can occur. The moments of the ponderomotive forces exerted on a particle of a nonspherical shape are determined. It is shown that the direction of the moment of a ponderomotive force alternates depending upon the sign of the difference between the electric conductivity of an inclusion and of the host medium.

Notes: Localized reversal of patterns formed by ferroelectric domains on the naturally grown surface in an iron-doped potassium niobate (Fe: KNbO3) single crystal has been observed. Surface-profile measurement, scanning electronic microscopy, and the metalomicroscopy methods were used to verify the observation. The mechanism of forming this domain structure is discussed.

Notes: Thin-film metal and metal oxide coatings ion sputter deposited onto the cathode electrode of a parallel-plate high-voltage gap were found to suppress electron emission in high vacuum. Electric fields as high as 60 kV/mm have been sustained across a 1 mm gap for pulse durations of 10 μs. Lesser electric fields were sustained for pulse durations exceeding 10 ms. The behavior of 500 nm thick coatings was independent of the type of coating used, whereas the breakdown voltage for thinner coatings depended upon the material and the deposition conditions. The breakdown properties of various coatings, as well as a discussion of the suppression of electron emission is presented.

Notes: An x-ray-diffraction method for determining the thickness of multiple layer thin films grown on a single-crystal substrate is presented. The equations, based on the kinematical theory of x-ray diffraction and the mosaic crystal model, were developed. As an example of the application of the method, thicknesses of a double heterostructure system, namely AlAs/AlGaAs/GaAs, were determined. Good agreement was obtained between the results from x-ray measurement and scanning electron microscopy data, demonstrating the high precision of this technique.

Notes: Diamond films were deposited on ZnO:Al thin-film silica substrates by hot-filament chemical vapor deposition. Ultrasonic irradiation in a diamond suspension enhanced the diamond nucleation density on a ZnO:Al-silica substrate. The nucleation density and the growth rate of diamond film deposited on ZnO:Al thin film is higher than on the silica. The cracks on a ZnO:Al-silica substrate occurred during the diamond deposition process. It is proposed that the cracks were caused by the stress in ZnO:Al film and diamond film, and the peak frequency shift of the Raman line of diamond indicates the presence of compressive stress in the diamond film.

Notes: The structure of the CuInxGa1−xSe2 system can be determined from the viewpoint described in this comment by Tinoco et al. Our argument concerning the x-ray study published by T. Tinoco, C. Rincon, M. Quintero, and G. Sanchez Perez [Phys. Status Solidi A 124, 427 (1991)] is summarized.

Notes: An experiment demonstrating the production of a "radius-tailored beam,'' which has a larger head and smaller tail for more stable propagation through gas, using a fast rise-time focusing coil is described. The results and their analyses for the radius-tailored case, the untailored case, and the "reverse-tailored'' case are presented. A two-dimensional particle simulation of the experiment was performed for each case using the parameters of the experiment for the inputs and it produces results consistent with the experiment.

Notes: The parallel electron velocity distribution function has been measured on an electron beam from a magnetron injection gun (MIG) using an E(parallel)B energy analyzer. The magnetic field, beam voltage, and beam current were scaled down from normal operating parameters. The perpendicular velocity spread, inferred under the assumption that the electrons are monoenergetic, is relatively constant for electron velocity ratios β⊥/β(parallel)(approximately-greater-than)0.7 and increases approximately linearly with the beam current. The current scaling of the perpendicular velocity spread is also consistent with electron-loss currents measured at the control anode of the MIG. Observed perpendicular velocity spreads for the gun design parameters are substantially larger than computational values.

Notes: An endoreversible power cycle can be seen as one in which two heat transfer processes and one work production process in series are sharing (and competing for) the same overall temperature potential. An optimal allocation of the temperature difference available from a heat source and a heat sink between these processes therefore results in a two dimensional search. It is found that, for the Curzon–Ahlborn (C-A) endoreversible power cycle, the maximum power occurs when the available overall temperature difference is shared equally between heat transfer and work production processes, and the specified overall heat conductance equally distributed between the boiler and the condenser heat exchangers. A more flexible optimization approach consists in analyzing the C-A power cycle within a larger family of endoreversible maximum power cycles defined with a specified operating temperature range for the working fluid as a parameter. This general optimization problem is solved as a one degree of freedom search and used to fit observed efficiency and temperatures of actual cycles to separate internal from external irreversibilities.

Notes: Reflection second-harmonic generation from the polished waveguide end face is used to investigate the second-order nonlinear optical properties of as-exchanged and annealed proton-exchanged waveguides in z-cut LiTaO3. It is found that proton exchange results in a strongly reduced nonlinearity which, however, can be restored by thermal annealing. Selective etching is used as a method to further investigate the structural changes following proton exchange and annealing. A correlation is found between the etching properties and the nonlinear optical properties. The results are discussed and related to the hydrogen concentration profile using a linear diffusion model for the anneal process.

Notes: Basic design principles are formulated for minimizing the threshold current density in InGaAsP/InP strained-layer single-quantum-well (SL-SQW) lasers. A quaternary InGaAsP active layer is shown to provide more freedom in design than a ternary InGaAs active layer because the amount of strain (both tension and compression) and quantum-well thickness can be independently determined in the InGaAsP system for a given emission wavelength. Strain-induced changes in the valence-band structures are analyzed within the framework of k⋅p theory by taking into account the interaction with spin-orbit split-off bands as well as heavy-hole and light-hole bands. It is clarified that the quantum-well thickness plays a more significant role than the amount of strain when designing compressive-strained wells, while the situation is just the opposite in tensile-strained wells. It is shown that, although the application of biaxial tension reduces the threshold current density in bulk-like SL-SQW lasers more significantly than biaxial compression, the quantum-confinement effect has a pronounced impact on the reduction in the current density in compressive-strained wells. This makes either type of strain attractive for reducing the threshold current density in InP-based SL-SQW lasers.

Notes: The thermal characteristics of p-type substrate external-cavity surface-emitting laser diodes (SELDs) under pulsed and continuous-wave operation are studied both theoretically and experimentally. The theoretical work establishes models for the current flow, spontaneous emission transfer, heat generation, and temperature rise in the laser. The models incorporate the reduced and anisotropic electrical and thermophysical properties of the Bragg reflectors. The temperature rise at the active region of an external-cavity SELD is determined by measuring the output wavelength as a function of the injection current, modulation frequency, and substrate temperature. Simulation results agree reasonably well with experimental data on the external-cavity SELD. It is shown that the thermal resistance of the studied laser is smaller than previously reported values for the n-type substrate SELDs because of the current spreading in the p-type mirror and its close proximity to the substrate.

Notes: Magnetic garnet films of composition (Y,Bi)3(Fe,Al)5O12 are grown by liquid phase epitaxy on [111] oriented substrates of Gd3Ga5O12. They support periodic lattices of parallel stripe domains. A simple strip antenna is used to excite the domain wall resonance and the two branches of the domain resonance in the frequency range up to 5 GHz. The resonance frequencies and the dynamic components of the magnetization are calculated using a hybridization model. Good agreement between calculated and measured resonance frequencies is obtained if the quality factor of the film is larger than 0.6. Optical modes are coupled into the waveguiding film. The excited domain resonances cause dynamic conversion of transverse electric and transverse magnetic modes by the Faraday and the Cotton–Mouton effects. Mode coupling and conversion are calculated by the perturbation theory. The dynamic conversion efficiencies are measured at the fundamental and the first harmonic frequency and at zero diffraction order as a function of the static induction applied in the film plane parallel to the stripes. Conversion efficiencies up to 18% are achieved at a frequency of 2.8 GHz. From the experimental data the precession angles are derived.

Notes: Possible approaches to thermal wave measurement of thermal diffusivity were considered. The calculation of the transverse photodeflection signal in the framework of wave optics proves to be a more accurate approach compared to that of geometrical optics. The influence of the account of wave optics effects on the accuracy of the thermal diffusivity determination was theoretically analyzed under various conditions. It was shown that a difference between the thermal diffusivities calculated using both approaches is most apparent when the thermal diffusivity of a sample is smaller than that of the deflecting medium, at high modulation frequencies and/or at large probe beam radius.

Notes: The numerical modeling of the oxidation of silicon is analyzed from a nonlinear viscoelastic approach. Its mechanical and stress dependent parameters are determined for silicon dioxide and nitride. The study focuses on the rheological behavior of the materials. The two dimensional simulations of silicon cylinders oxidation and local oxidation of silicon processing reveal that at 1000 °C, a nonlinear viscous modeling is equivalent to the nonlinear viscoelastic one. But, for lower temperatures, the discrepancies between these two models, observed in the stress calculation and final oxide shape, demonstrate the necessity for a complete nonlinear viscoelastic formulation. Finally, the calibrated model is used to study the growth of a recessed isolation structure. The investigations quantify the influence of geometrical parameters of the silicon groove on the shape of the final isolation oxide (e.g., parameters such as the silicon overetch under the pad oxide, the depth of silicon etching, the slope of the silicon sidewall and the silicon concave corner rounding).

Notes: Elastically scattered incident radiation (ESIR) from a copper wire target illuminated by a KrF laser pulse at λ=248 nm shows a distinct two-peak structure which is dependent on the incident energy. The time required to reach the critical electron density (nc ≈ 1.8× 1022 electrons/cm3) is estimated at 11 ns based on experimental results. Detailed ESIR characteristics for water have been reported previously by the authors. Initiation of the broadband emission for copper plasma begins at 6.5 ±1.45 ns after the arrival of the laser pulse. However, the broadband emission occurs at 11 ± 0.36 ns for water. For a diatomic substance such as water, the electron energy rapidly dissipates due to dissociation of water molecules, which is absence in a monatomic species such as copper. When the energy falls below the excitation energy of the lowest electron state for water, it becomes a subexcitation electron. Lifetimes of the subexcited electrons to the vibrational states are estimated to be of the order of 10−9 s. In addition, the ionization potential of copper (440–530 nm) is approximately 6 eV, which is about two times smaller than the 13 eV ionization potential reported for water. The higher ionization potential contributes to the longer observed delay time for plasma formation in water. After initiation, a longer time is required for copper plasma to reach its peak value. This time delay in reaching the maximum intensity is attributed to the energy loss during the interband transition in copper.

Notes: Experiments were performed to determine the plasma propagation velocities resulting from KrF laser irradiation of copper wire targets (75 μm diameter) and water droplets (75 μm diameter) at irradiance levels ranging from 25 to 150 GW/cm2. Plasma propagation velocities were measured using a streak camera system oriented orthogonally to the high-energy laser propagation axis. Plasma velocities were studied as a function of position in the focused beam. Results show that both the shape of the plasma formation and material removal from the copper wire are different and depend on whether the targets are focused or slightly defocused (≈0.5 mm movement in the beam axis). Plasma formation and its position relative to the target is an important factor in determining the practical focal point during high-energy laser interaction with materials. At irradiance of 100 GW/cm2, the air plasma has two weak-velocity components which propagate toward and away from the incident laser while a strong-velocity component propagates away from the laser beam as a detonation wave. Comparison of the measured breakdown velocities (in the range of 2.22–2.27 ×105 m/s) for air and the value calculated by the nonlinear breakdown wave theory at irradiance of 100 GW/cm2 showed a quantitative agreement within approximately 50 % while the linear theory and Gaussian pulse theory failed. The detonation wave velocities of plasma generated from water droplets and copper wire targets for different focused cases were measured and analyzed theoretically. The propagation velocities of laser-induced plasma for liquid droplets obtained by previous research are compared with current work.

Notes: Surface preflashover phenomena across alumina ceramics in vacuum are presented. The localized plasmas are identified by the spatially resolved images that appear in a pulse surface preflashover phase using an intensified charge-coupled-device camera, along with coordinated, time-resolved preflashover current and luminosity. It was observed that significant preflashover activity associated with the localized plasmas on the surface appears after a flashover event and disappears after a few pulse voltage applications. Large bursts of current pulses (several hundred milliamperes) with no counterpart in the luminosity signal were also observed during preflashover. These phenomena cannot be explained satisfactorily by electron stimulated gas desorption. The effect of adsorbed gases and surface polishing on preflashover are presented and discussed. The experimental results indicate that preflashover essentially depends on the surface state (physical and chemical) of the solid insulator. The preflashover conduction associated with the localized plasmas is attributed to the detrapping and impact-ionization processes associated with defect centers. The results reported support the surface flashover model based on trapping/detrapping and impact ionization processes.

Notes: Plasma opening switches are believed to develop voltage through opening and magnetically insulating a gap formed in plasma injected between two conducting electrodes. Data will be presented to show that a relatively simple assumption about the behavior of this gap suffices to describe the overall operation of the switch. Specifically, the observed increase, then decrease, of operational voltage as a function of conduction time can be accounted for by the switch gap linearly decreasing with conduction time, while the conducted current increases. In addition to presentation of data to demonstrate this collapse, analysis is performed to couple the gap history to the voltage predicted for the switch, and to optimize this latter quantity.

Notes: A novel mechanism called particle sputtering and deposition is proposed based on the facts that the arced contact morphology, mass transfer, and arc durations in the metallic phase and the gaseous phase undergo transition phenomena with increasing load current. The mechanism is used satisfactorily as an explanation of the material transfer process and the contact resistance degradation of Pd and Ag contacts during the breaking of dc circuits.

Notes: The radiative emission from a soft x-ray laser-produced plasma source is examined in detail for the particular case of a Sn solid target under moderate illumination conditions: 1010–1012 W/cm2. The prominent spectral features which contribute to the relatively large conversion efficiency in the range 124–155 A(ring) are identified using extensive multi-configuration Dirac–Fock calculations. These data are used to generate synthetic spectra with plasma characteristics derived from a two-dimensional simulation of the ablating, radiating plasma. The comparison to the experimental results is generally favorable.

Notes: Thin C60 films have been deposited on mica(001) substrates by thermal evaporation at substrate temperatures between room temperature and 200 °C and at a constant deposition rate. The influence of the substrate temperature on the growth of C60-thin films has been systematically investigated by x-ray diffraction. θ–2θ measurements of the (111) peaks show a decrease of the full width at half-maximum (FWHM) with increasing substrate temperature, leading to a minimum FWHM of 0.15° for a substrate temperature of 200 °C. Oriented films with an out-of-plane mosaic spread of Δω=0.2° could be grown at a substrate temperature of 150±25 °C. It can be shown that the in-plane epitaxial arrangement C60(111)(parallel)mica(001) is determined by the seeding conditions and is independent of the substrate temperature. An increasing substrate temperature enhances the epitaxial alignment of the C60 crystals oriented with a {111} face parallel to surface and also the azimuthal alignment of the twins which are rotated by 60° about the surface normal.

Notes: The behavior of nitrogen and the formation of nitrogen-oxygen donors in nitrogen-doped Czochralski-silicon crystals (Cz-Si) were studied by the electrical and infrared-absorption measurements in samples annealed in different conditions. Experiments showed that nitrogen-oxygen shallow donors are formed during the nitrogen-doped Cz-Si crystal growth. Nitrogen-oxygen thermal donors are generated in temperature range of 300–550 °C, and the behavior of these thermal donors resembles that of the thermal donors in Cz-Si crystals [P. Wagner, C. Holm, and E. Stirtl, Festkoperprobleme 24, 191 (1984)], but they are monovalent donors with the same levels as those of neutral thermal donors; no new electrically active center is generated in the temperature range of 600–900 °C.

Notes: In the Ni-Mo system, amorphous alloys and three metastable crystalline (MX) phases, i.e., an hcp (hcp-I), an enlarged hcp (hcp-II), and a fcc phase, were synthesized by solid-state reaction in the Ni-rich multilayers. An interesting point is that with increasing the annealing temperature, the hcp-I phase was formed first in the Ni67Mo33 film and then turned amorphous, while in the Ni75Mo25 film, the amorphous phase was formed first and then changed into the hcp-I phase. The compositions of three MX phases were determined experimentally. A free-energy diagram of the Ni-Mo system was established by calculating the free energies of the amorphous phase and the three MX phases, as well as those of the solid solutions and the related compounds on the basis of the model of Niessen et al. [CALPHAD 7, 51 (1981)] and Alonso's method [Solid State Commun. 46, 765 (1983)]. The calculated free-energy diagram can give relevant interpretation to the observed phase formation behaviors.

Notes: Sintered disks of TiO2 (rutile) have been exposed to the beam of a 355 nm laser. X-ray photoelectron spectroscopy studies show that the strong energy deposition at the surface is responsible for the formation of Ti3+ ions down to a depth of 0.4 μm. The presence of these ions produces a large decrease in the electrical resistivity from 109 Ω cm in nonirradiated TiO2 to 10 Ω cm after laser treatment. Temperature dependent resistance measurements indicate that the creation of Ti3+ ions leads to the formation of energy levels in the gap and to a hopping-driven conduction at room temperature.

Notes: High energy (MeV) Cu ions were implanted into n-type Si samples at angles of 7°, 30°, 45°, and 60°. The doses were 5×1014 and 2×1014 ions/cm2. The damage profiles in Si(100) were investigated by a Rutherford backscattering/channeling technique with 2.1 MeV He ions. The longitudinal damage straggling and lateral damage spread are estimated for 1.0 MeV Cu+ implanted in Si(100). The values obtained are compared with the trim (transport of ions in matter) code. The results show that the longitudinal damage straggling is found to be in good agreement with the calculated one within 13% by use of the trim code, but the experimental value of the lateral damage spread is higher than the calculated one by about 28% using the trim code. The effect of dose rate, energy, and dose on damage distribution is investigated also.

Notes: A quantitative measure of elastic wave nonlinearity in crystals is provided by the acoustic nonlinearity parameters. The nonlinearity parameters are defined for arbitrary propagation modes for solids of arbitrary crystalline symmetry and are determined along the pure mode propagation directions for 33 crystals of cubic symmetry from data reported in the literature. The magnitudes of the nonlinearity parameters are found to exhibit a strong dependence on the crystalline structure and symmetries associated with the modal direction in the solid. Calculations based on the Born–Mayer potential for crystals having a dominant repulsive contribution to the elastic constants from the interatomic pair potential suggest that the origin of the structure dependence is associated with the shape rather than the strength of the potential. Considerations based on variations in crystal symmetry during loading along pure mode propagation directions of face-centered-cubic solids provide a qualitative explanation for the dependence of the acoustic nonlinearity parameters on modal direction.

Notes: The thickness and composition of the InGaAs layer in GaAs/AlGaAs/InGaAs/AlGaAs/GaAs high-electron-mobility transistor devices were determined to within ±5 A(ring) and ±0.003, respectively, using high-resolution x-ray diffraction. The combined thickness of the capping AlGaAs and GaAs layers were also determined to within ±5 A(ring). Although the interference effects near the substrate peak in the diffraction pattern may be identical for structures with different InGaAs thicknesses, the peak from the buried InGaAs layer will be different. In other words, if the diffraction from the buried layer is measured, one can readily distinguish between structures whose interference peaks are otherwise the same. It is also shown that the use of different reflections removes the ambiguity associated with interference peaks.

Notes: Ion implantation has frequently been shown to modify the shape of quantum wells following thermal annealing by enhancing the interdiffusion. We have shown that, independent of chemical effects on the interdiffusion, there is a contribution from the vacancies created by the implant. The effects of the diffusion of these vacancies can be modeled using a very simple expression that does not rely on any knowledge of the diffusion coefficients for the vacancies in the material. By comparing the problem to the classic problem of the gambler's ruin, we have shown that implantation into a surface should produce even intermixing of layers at different depths below it, ignoring the effects of vacancy trapping and other depth dependent diffusion processes.

Notes: High-coercivity magnets with a nanometer-sized microstructure were fabricated by the shock-compaction technique using an acicular iron-alloy powder as a starting material. The magnetic properties of these magnets depended on the initial magnetic property of the acicular powder even after shock compaction and were considered to be attributed to the shape anisotropy of acicular single-domain fine particles. The preferred orientation of the acicular particles by which the powder was precompacted in the external magnetic field affected final magnetic properties of the shock-compacted magnets. The best value of the maximum energy product (BHmax) obtained in this study was 23 kJ/m3 (2.9 MG Oe), which was comparable to that of the elongated single-domain magnet.

Notes: An improved hydrodynamic model has been developed for analysis and interpretation of shock-induced chemical reactions in inorganic powder mixtures. Attention is focused on the interaction of hydrodynamic flow with the thermodynamic changes associated with chemical reactions such as specific volume, specific internal energy, and material properties. Predictive capabilities of the model have been illustrated using experimental shock wave profile data involving Ni-Al and Al-Fe2O3 mixtures.

Notes: The mechanical properties of nanocomposite Ag-Al2O3 and Fe-SiO2 granular metal films were investigated by low-load indentation and laser ultrasonic methods. A discontinuity in the rate of change of hardness as a function of metal volume fraction was observed near the percolation threshold for the metal. However, the sign of the rate of change was different for the two systems. It is suggested that this discontinuity was the result of a change in deformation mechanism as the percolation threshold for the softer phase was crossed. The elastic moduli of the Ag-Al2O3 films as measured by a laser ultrasonic technique displayed a rule-of-mixtures-type behavior, in contrast to nanoindentation measurements that gave a peak in the apparent compliance. Indentation crack testing indicated that the toughness increased in the Ag-Al2O3 system when the metallic phase was introduced into the ceramic, but that the Fe-SiO2 system displayed unexpected and novel fracture behavior.

Notes: Open-volume defects introduced in Si(100) crystals during fluorine implantation were investigated by variable-energy positron beam depth profiling. The behavior of the implantation-induced lattice defects upon high temperature annealing and their role in the surface-oriented diffusion of F impurities were examined. The defects become mobile and undergo recovery at temperatures below 550 °C, i.e., well before the onset of fluorine diffusion as seen by secondary ion mass spectroscopy (SIMS) profiling. This behavior suggests that after irradiation and annealing the fluorine occupies substitutional sites to which positrons are insensitive. The anomalous F diffusion seen in SIMS has been explained through a two-step diffusion mechanism, in which the diffusion kinetics is determined by dissociation of the substitutional F into an interstitial F and a vacancy, followed by a rapid diffusion of the interstitial F and the vacancy through the crystal to the surface.

Notes: A defect chemistry approach is used to analyze the observed concentration dependence of the lithium in-diffusion rate into lithium niobate. An ambipolar diffusion model is used where the lithium and niobium diffusion rates are coupled and therefore cause an internal electric field to develop during the diffusion process. Both lithium and niobium diffusion are found to proceed via simple vacancy mechanisms. The internal electric field forces the faster lithium diffusion to be reduced to a level consistent with the slower niobium diffusion process. This coupling gives a concentration dependence that matches the observed linear increase in interdiffusion rate as the total cation vacancy concentration decreases.

Notes: Interconnects are susceptible to solid diffusion under residual stress, electric current, and elevated temperature. As atoms diffuse, voids nucleate, drift, and enlarge. At some point, the voids of rounded shape can collapse to narrow slits and sever the interconnects. The fatal slits are often found to be transgranular, i.e., each slit cuts across a single grain. They have raised many concerns, but the underlying mechanism has remained unclear. It is proposed that a void changes shape due to surface diffusion under the combined action of surface energy, elastic energy, and electric current. The void will be rounded if surface energy prevails, but will collapse to a slit if the elastic energy or the electric current prevails. A cylindrical void in an infinite crystal under biaxial stresses, but under no electric current, is analyzed. Four things are done, as follows: (1) A suitable thermodynamic potential is minimized and maximized to select, among a family of ellipses, equilibrium void shapes. The bifurcation diagram consists of a subcritical pitchfork and two Griffith cracks. (2) A void under biased stresses is analyzed to illustrate the effect of imperfections. (3) Exact initial bifurcation modes are determined. The critical loads for the successive modes are closely separated, indicating that the shape evolution will be sensitive to initial imperfections. (4) A variational principle for shape evolution under stress, current and surface energy is identified. Stress-induced evolution time is estimated by using this variational principle.

Notes: Atomic force microscopy has been used to study the surface irregularities of hydrogenated and unhydrogenated carbon films grown by rf-powered glow discharge and dc-magnetron sputtering, respectively. In general films produced with the latter technique have rougher surfaces. In glow discharge produced samples, roughness can be reduced by increasing rf power during deposition. In sputtered films surface roughness is reduced by either lowering substrate temperature or reducing the power density during deposition. The relation between the microstructure and the crystalline state was done using Raman scattering.

Notes: We report the growth of high-quality As-based ternary and quaternary alloys lattice matched to InP using a valved arsenic source that can post-heat the As beam after evaporation. We find that the optimum group-V-to-group-III beam-equivalent pressure ratio for growth of (In,Ga)As alloys using this source is considerably lower than values reported previously for growth using conventional As4 sources. Consequently, high-quality (In,Ga)As, (In,Al)As, and (In,Al,Ga)As alloys (and quantum wells made from these alloys) can be grown under the same growth conditions, i.e., substrate temperatures between about 525 °C and 540 °C and V/III pressure ratios between 10:1 and 15:1. Thick-film alloys and multiple-quantum-well structures grown under these conditions show superior structural and optical quality. Strong excitonic features are observed in the room-temperature absorption spectra of a number of multiple-quantum-well structures with well widths ranging from 30 A(ring) to 170 A(ring) . Calculations of the exciton transition energies using a simple empirical two-band model are in excellent agreement with experiment, even for a structure containing quantum wells in tensile strain in which the ordering of ground-state light- and heavy-hole excitons is reversed. The optical absorption spectrum of a 50-A(ring) -period (In,Ga)As/(In,Al)As superlattice shows room-temperature excitons involving electronic states at both the bottom and top of the minibands. Exciton line widths for these quantum-well structures, measured using low-temperature photoluminescence, are consistent with the limits imposed by random alloy fluctuations. We tentatively explain the lower optimum V/III pressure ratio for growth of (In,Ga)As in terms of the increase in kinetic energy of As4 molecules (compared with the kinetic energy of molecules from a conventional As4 source) and the consequent enhancement in the efficiency of dissociation of As4 molecules into As2 molecules at the growing surface.

Notes: Molecular slip is expected to occur at the interface between the surface of a bulk-acoustic-wave quartz chemical sensor and a liquid with which it is in contact. A new property of the interface, called the interfacial slip parameter, accounts for molecular slip. It is introduced in the theory of the sensor-liquid system by generalizing the no-slip boundary condition. The interfacial slip parameter is a complex-valued quantity which is defined as the displacement of a particle of liquid in contact with the sensor surface divided by the displacement of a particle on the surface of the sensor. The theoretical expression for the impedance of the sensor is derived in terms of the interfacial slip parameter. The impedance of the sensor is measured by the network analysis method for hydrophilic and hydrophobic surfaces in contact with water-glycerol solutions. The experimental values of impedance are fitted to the theory by nonlinear regression analysis to find the interfacial slip parameter. A mechanical model of molecular slip is devised which explains the variation of the interfacial slip parameter with the viscosity of the liquid. The results indicate that there is some slip in low-viscosity liquids, near the viscosity of water, and very little slip in liquids of high viscosity, near that of glycerol.

Notes: Efficient collection of atoms into a vapor-cell laser trap requires a special wall material for the cell that minimizes the interactions between the vapor and the wall. Tests of several different wall coatings and materials are reported, and measurements of adsorption energies, outgassing, and chemical reaction rates between the alkali vapor and the walls are described. It is demonstrated that each of these parameters affects the collection efficiency.

Notes: Diamond and carbonaceous films grown in a microwave assisted plasma reactor have been characterized by x-ray diffraction, scanning electron microscopy, and Raman spectroscopy. This study is mainly focused on the identification of the different carbonaceous compounds which can coexist with the diamond depending on the synthesis parameters. Selective etching reactions and the excitation wavelength dependence of Raman cross sections of the different carbon species reveal that the broad and poorly structured Raman spectra in the 1100–1700 cm−1 region contain six components that arise from three different carbonaceous species: (i) one species is graphitic and is identified from the components at about 1350 and 1590 cm−1 which are resonantly enhanced as compared to the other components when the excitation energy is lowered from 3 to 2 eV. (ii) The second species is associated with the two broad bands at 1350 and 1550 cm−1 and is attributed to amorphous diamond-like carbon (a-C:H); both bands are resonantly enhanced by excitation energies in the range from 2.41 to 3.53 eV. These bands do not shift with exciting energy. (iii) The third species is associated with a band centered at about 1470 cm−1 and a weak band at 1150 cm−1. This 1470 cm−1 band overlaps with the 1550 cm−1 band of the diamond-like component, resulting in an apparent frequency down-shift when the incident photon energy decreases below the resonant enhancement range of the 1550 cm−1 band. Possible models for the structure of the third species are discussed.

Notes: Al films were grown by chemical vapor deposition at 400, 550, and 700 °C on GaAs(100) substrates using the molecular precursor dimethylethylamine alane. The film morphology and composition were studied in situ by reflection high-energy electron diffraction and Auger electron spectroscopy, and ex situ by atomic force microscopy and scanning electron microscopy. Chlorine (at 400 °C) and C and N (at 550 and 700 °C) at or below the percent level were found to be the major contaminants of the deposited films. Systematic studies for deposition at 400 °C established that the film microstructure evolves via the growth and coalescence of three-dimensional faceted islands with (100)Al(parallel)(100)GaAs or (110)Al(parallel)(100)GaAs preferential orientation. Coalescence of such crystallites was observed only for equivalent coverages of Al above 150 nm. Comparison with the microstructure of Al films obtained by evaporation suggests that in the temperature range examined the evolution of film morphology during chemical vapor deposition from dimethylethylamine alane was mainly determined by surface diffusion of isolated adsorbed Al atoms.

Notes: An atomistic model consistent with a variety of experimental observations is developed for GaN growth by molecular-beam epitaxy. The model is used in Monte Carlo simulation to study the impact of substrate temperature, Ga flux, and V/III (group-V element to group-III element) ratio on growth rate and growth front quality. The growth rate increases with the V/III ratio reaching a saturation value which is determined by the Ga flux. The quality of the growth front improves by using a smaller Ga flux for a fixed temperature and V/III ratio or by reducing the V/III ratio at a given temperature. A consideration of the growth kinetics suggests that GaN grown surfaces are likely to be Ga stabilized. These theoretically estimated trends are evidenced by two-dimensional and three-dimensional growth front contours evaluated under various growth conditions.

Notes: Double-crystal and triple-axis x-ray diffractometry was used to characterize in detail the strain and composition of short period Si6Ge4, Si8Ge8, Si9Ge6, and Si17Ge2 strained-layer superlattices (SLSs), grown by molecular-beam epitaxy. Nominally strain-symmetrized superlattices, intended to be free standing from underlying buffer layers and the substrate, grown on rather thin (20 nm thick) SiGe alloy buffers with constant Ge content (Si6Ge4 and Si8Ge8) are compared to those grown on 1.3-μm-thick step-graded SiGe alloy buffers (Si6Ge4 and Si9Ge6). Due to the much higher instrumental resolution offered by triple-axis diffractometry (Δ2aitch-theta=12 arcsec) buffer and SLS peaks are clearly separated from each other, which overlap in corresponding double-crystal-diffractometry measurements (Δ2aitch-theta in the range of 180 arcsec to 2°). The lattice constants parallel and perpendicular to the [001] growth direction are determined independently from each other and thus precise strain data of the buffers and the SLS constituting layers were extracted from two-dimensional reciprocal space maps around (004) and (224) reciprocal lattice points (RELPs). The only fitting parameter in a dynamical simulation of conventional rocking curves is then the relative thickness ratio of the Si and Ge layers in the superlattice. The strain relaxation process and the principle of tailoring the strain status in the electronically active layers are shown to be different in structures with single-step and step-graded buffers. For these superlattice samples the RELPs originating from the zeroth-order SLS reflection show significant mosaic broadening (full width at half-maximum of 1100–1300 arcsec perpendicular to the ω/2aitch-theta scan direction in reciprocal space). In contrast, the corresponding RELPs from a pseudomorphic SLS with a much higher average Si content (Si17Ge2), but consisting just of 10 periods, which was grown directly on a Si buffer layer deposited on (001)-oriented Si substrate, are not broadened (full width at half-maximum of 14 arcsec). Besides the determination of strain and composition, examples for the interpretation of diffuse x-ray scattering are given, conveniently measured by reciprocal space mapping.

Notes: The results of a computer simulation for islands in semiconductor strained heterojunctions are presented. The atomic positions are individually calculated by means of a minimization procedure of the total elastic energy, evaluated within Keating's model. Results obtained for Ge/Si and InAs/GaAs systems differ substantially from the ones obtained via the macroscopic theory, which largely overestimates the elastic relaxation energy. The results are applied to evaluate the critical thickness corresponding to undislocated island nucleation in epitaxial strained heterostructures.

Notes: Plates of β-NbD0.7 were irradiated with single pulses of a KrF excimer laser having a wavelength of 248.4 nm. The power level was varied from roughly 0.1 to 1 GW/cm2. The pulse length was about 15 ns. The irradiated areas at low and medium power displayed individual hydrogen (deuterium) blisters which started overlapping at higher power levels. Calculation of the thermal diffusivity showed that a surface layer roughly 0.6 μm thick was heated during the pulse. Applying Bechtel's formula [J. Appl. Phys. 46, 1585 (1975)] for the temperature increase in the heated layer demonstrated that melting of the heated layer must have occurred during the pulse. The formation of hydrogen (deuterium) blisters is understandable in terms of the very high equilibrium vapor pressure of the hydrides near the melting temperature. A model for the formation of the blisters is presented.

Notes: The point defects in β-iron disilicide (β-FeSi2) have been investigated by electron paramagnetic resonance (EPR) along with the thermoelectric properties. The samples used are FeSix (1.9≤x≤2.8) ceramics sintered with FeSix micrograins processed in SiH4-plasma and nonprocessed. EPR detects (1) several S=1/2 signals and (2) a multiplet signal. An S=1/2 signal with orthorhombic g factors (g1=2.061, g2=2.047, g3=2.024) is detected in all n-type specimens, and ascribed to unpaired electrons of donors associated with iron vacancies. The other S=1/2 signals are detected in both n-type and p-type specimens. The centers responsible for these signals are considered to exist in an oxidized intergrain region of the ceramics. The EPR intensity of the multiplet signal due to the closed-pair of a high spin Fe3+ ion (S=5/2) and an S=1/2 center is reduced by SiH4-plasma treatments. SiH4-plasma processing of FeSix micrograins prior to sintering changes the Seebeck coefficient sign and magnitude, depending upon the condition of plasma processing. This Seebeck change is explained in terms of the introduction of silicon- and/or iron-vacancies. SiH4-plasma processing increases the electrical conductivity of nearly stoichiometric specimens and reduces the EPR intensity of the multiplet signal. A possible model for the increased conductivity is discussed using an energy band model for the ceramic grains and grain boundaries. We consider that the change of the thermoelectric properties by SiH4-plasma processing is mainly due to the modification of defects in and near the ceramic intergrain region.

Notes: The effects of transgranular microcrack nucleated from a grain-boundary ledge on a dislocation-free zone model of fracture are investigated by a continuous dislocation modeling method. The dislocation distribution functions to simulate the crack and plastic zone, the number of dislocations in the crack and plastic zone, the stress field, and stress intensity factor at the crack tip are obtained. If the Burgers vector of dislocations composing the grain-boundary ledge has the same sign as that of the plastic zone dislocations, all the dislocation distributions that simulate the crack and plastic zone, the number of dislocations in the crack and plastic zone, the stress field in the dislocation-free zone, the stress intensity factor at the crack tip, the dislocation-free zone size, and the plastic zone size increase with an increasing number of grain-boundary ledge dislocations, but the applied stress σ to accumulate the plastic zone dislocations decreases with an increasing number of grain-boundary ledge dislocations. The effects of transgranular microcrack nucleated from a grain-boundary ledge on a fracture without a dislocation-free zone are also considered. When there is no dislocation-free zone in front of the crack tip, the stress intensity factor is zero, regardless of whether the grain-boundary ledge dislocations exist.

Notes: The nature of the interfaces of chemically vapor deposited diamond films on Fe substrates with and without a protective TiN coating is investigated. For unprotected Fe substrates a thick graphitic soot containing 6.5% Fe grows upon the Fe in the first few minutes of exposure to the plasma and, once this soot completely covers the substrate, diamond can nucleate and grow upon it into an average quality unfaceted continuous diamond film. However, adhesion is poor, the weak link being the lack of structural integrity of the soot layer itself. A TiN coating is found to prevent soot formation, C diffusion into the Fe bulk, and Fe diffusion into the diamond films. In the initial stages of growth the TiN is covered with a thin layer of amorphous carbon (a-C), and it is on this layer that diamond nucleates and grows. Here, again, adhesion is not strong, with delamination occurring at the TiN/a-C interface.

Notes: We have studied the initial stages of oxidation of the hydrogen-terminated Si(111) and (100) surfaces stored in air, using infrared spectroscopy in the multiple internal reflection geometry. We investigate the effect of surface roughness and humidity of air on the oxidation of the hydrogen-terminated Si surfaces. We suggest that surface roughness on a microscopic scale does not significantly affect the oxidation of the hydrogen-terminated Si surface and the oxidation occurs on the entire surface. It is demonstrated that water is predominantly involved in the oxidation of the surface Si—H bond, and that the surface Si—H bond is quite inert to the oxygen molecule.

Notes: Amorphous GeO2 optical thin films were grown in an oxygen ambient on heated Si substrates using the technique of pulsed laser deposition. The application of a partially ionized oxygen plasma generated by passing the plume through a ring electrode facilitated stoichiometric film growth in low O2 partial pressures. Emission spectroscopy of the plume revealed an enhancement in the ionic and neutral excited Ge species. The concentration of excited neutral and ionic oxygen atoms also significantly increased when the ring electrode was activated at P(O2)(approximately-greater-than)10 mTorr. Coupling the results of the film property measurements with the emission studies suggested that the presence of O atoms near the substrate surface during film growth was more critical in promoting oxidation than the gas phase process in the plume. The low-pressure conditions that were utilized to deposit stoichiometric film growth identified the appropriate conditions to produce uniform films over a large area that may be suitable for waveguide fabrication.

Notes: Pulsed laser deposition was used to grow epitaxial β-FeSi2 films on Si(111) (1×1) and Si(111) (7×7) with the following epitaxial orientations: β-FeSi2(001)//Si(111) with β-FeSi2[010]//Si〈110〉 and three rotational variants. Silicide growth was influenced by substrate temperature and deposition rate, but not by the structure of the starting surface. Films containing both β-FeSi2 and FeSi were formed at low substrate temperatures and high deposition rates, while films containing only β-FeSi2 were formed at higher substrate temperatures and lower deposition rates. FeSi grains had the following epitaxial relationship to the Si substrate, FeSi(111)//Si(111) with FeSi(1¯10)//Si(112¯). The microstructure of the silicide films varied with film thickness, as did the roughness at the silicide/Si interface. These results suggest that an Fe-rich environment was created during the growth of the silicide films.

Notes: The temperature dependence of the conductivity of composite polymeric electrolytes based on a poly (ethylene oxide) matrix is described by a model based on effective medium theory. The influence of grain size distribution of the powder used as well as grain concentration on the conductivity of the electrolytes studied is analyzed. It will be shown that increases in the concentration of inorganic fillers influence the conductivity of the interface layer covering the filler. This layer is assumed to be responsible for an increase in the conductivity of composite polymeric electrolytes in comparison to pristine poly (ethylene oxide)-based systems. Utilization of effective medium models to study the temperature dependence of the conductivity of composite electrolytes enables us to predict the applicability of the compensation law for studies of the entropy effect in the electrolytes studied. The characteristic order-disorder temperature found from calculations based on effective medium theory is equal to that experimentally found and corresponds to the melting temperature of the crystalline polymer phase.

Notes: A numerical algorithm for the solution of the two-dimensional effective-mass Schrödinger equation for current-carrying states, the quantum transmitting boundary method, is extended to magnetotransport problems where a magnetic field is applied. Boundary conditions appropriate for such states are developed and a solution algorithm based on the finite-element method is constructed. The algorithm is valid for general device shapes, general potential profiles, and multiple leads of general orientations. The technique is applied to a quantum channel with a single scatterer, an antidot, in the channel. Magnetic quasibound states (MQBS) are formed around the scatterer and MQBS-induced resonant reflection is observed.

Notes: Steady-state photoconductivity, sub-band-gap absorption and electron spin resonance (ESR) measurements were carried out on annealed and light soaked intrinsic hydrogenated amorphous silicon (a-Si:H) films. The experimental results were modeled using detailed numerical analysis. The defect densities derived from the sub-band-gap absorption in the light soaked films were correlated with the ESR spin densities. Self-consistent fitting of the data was obtained using a gap state distribution which consists of positively charged defect states above, negatively charged defect states below and neutral defect states at about the midgap. Both the annealed and the light degraded states are modeled using the same gap states which increase upon light soaking and have a slight increase in the ratio of the neutral to charged defect densities. These results on intrinsic a-Si:H are consistent with proposed charged defect models and clearly indicate the importance of charged defect states in determining sub-band-gap absorption as well as its correlation with neutral dangling bonds.

Notes: Reoxidized nitrided oxide gate dielectrics are characterized following electrical stress, in order to study the mechanism by which they are able to inhibit interface-state generation, as compared to silicon dioxide. It is found that the energy spectrum of the few interface states which are generated in reoxidized nitrided oxides differs from that of states generated in lightly nitrided oxides or in silicon dioxide. We have also measured interface-state generation as a function of stress time and stress polarity for different dielectrics, and observed two characteristic types of fluence dependencies. Interface states which are generated linearly with fluence are modeled using a simple first-order rate equation, which depends on the density of interface-state precursor sites. A sublinear fluence dependence is successfully modeled by including the effect of interfacial strain relaxation. Based on these results, we argue that the first type of interface-state generation, which follows a linear fluence dependence, is suppressed in electrically-stressed reoxidized nitrided oxides because the interface has fewer defect precursor sites.

Notes: Waveguide losses in thin film polyimides using waveguide loss spectroscopy and photothermal deflection spectroscopy as a function of cure cycle and structure were studied. Fluorinated sidegroups on the polyimide backbone lead to decreases in birefringence and absorption. The primary waveguide loss mechanism is absorption, not scattering. Waveguide losses as low as 0.4 dB/cm at 800 nm have been measured. Losses as low as 0.3 dB/cm at 1300 nm can be inferred from the photothermal deflection spectroscopy.

Notes: An electro-optical device consisting of a stack of undoped alternating layers of narrow- and wide-gap materials (e.g., GaAs/AlGaAs) together with a series resistor under constant voltage bias is analyzed theoretically. The GaAs heterolayers with a width in the order of a carrier mean free-path are a photo-active region of this vertical device. The device operates only in the presence of a low-intensity light beam due to the intrinsic photoconductivity of the active region. The Franz–Keldysh effect, the strong accumulation of the photocarriers in the photo-active layers due to the charge separation in the presence of the static electric field, and the ballistic component of the total current are responsible for the unusually large electro-optical nonlinearity of the device. Kirchhoff's law for the electrical circuit of the device provides a sensitive feedback between the voltage drop over the layer and the photocurrent. In a Fabry–Perot cavity a room temperature electro-optical bistabitity is obtained at light intensities less than 10 mW/cm2 with a switching time of about 100 ns.

Notes: Low-temperature photoluminescence (PL) and Raman measurements were performed on AlInAs grown lattice matched to InP by molecular beam epitaxy at reduced growth temperature (Ts). The PL of layers grown at Ts above 500 °C is dominated by excitonic emission, whereas for lower Ts donor-acceptor related transitions prevail. Below a critical Ts of 450 °C a marked shift towards lower emission energies with a maximum shift near 400 °C is observed that is attributed to a modified band edge due to clustering. Comparable trends are detected by Raman spectroscopy. The observed reduction of the separation of the InAs- and AlAs-like longitudinal optical phonon modes (LOInAs and LOAlAs) demonstrates local internal strain to be present as a result of clustering. This effect reaches a maximum for Ts at 400 °C. A shift of the LOInAs solely accounts for this behavior. In addition strong asymmetric broadening of the LOAlAs-phonon line observed on low Ts material indicates an increasing reduction of the correlation length and suggests the structural disorder to be correlated with the AlAs sublattice. Taking into account the pressure dependence of the AlInAs energy gap and the frequency shift of the LOInAs phonon, the local internal strain equivalent pressure was calculated from the PL and Raman results, respectively, giving similar values of up to 5 kbar for material grown at 400 °C.

Notes: Recent experimental results show a linear energy response in high quality Nb-Al-AlOx-Nb superconducting tunnel junction detectors for photon energies between 1.5 and 6.4 keV. The experimental data are based on both direct x-ray illumination and on the escape and re-absorption of fluorescent photons created in the junction electrodes and in the silicon substrate. The observed linearity of the energy response raises questions on the validity of some theoretical models which describe the relaxation process occurring in a superconducting thin film after x-ray photoabsorption. Such models generally predict nonlinear effects due to large quasiparticle number densities and short recombination times.

Notes: The current through porous silicon p-n junctions is measured as a function of bias voltage and temperature. The ideality factor m=4 value observed under forward bias indicates that channels are present at the unpassivated porous silicon junction surfaces. The generation-recombination current stemming from the space-charge regions between the surface channels and the bulk silicon regions dominates the measured diode current. From the thermal activation energy of the diode saturation current a value of 2.2 eV is calculated for the band gap of porous silicon, which is in good agreement with the value of 2.0 eV found from the photo- and electroluminescence spectra emitted by these diodes.

Notes: Efficient generation of Stokes lines has been obtained in H2, CH4, N2, and O2 by 10 ps, 300 Hz, 193 nm laser pulses. A 60% energy conversion efficiency was recorded for the first Stokes of H2 and 40% in CH4. For the pressure range 1–30 atm the conversion efficiency did not change with pump pulse repetition rate up to 300 Hz. The higher order Stokes observed were generated mainly by means of four wave mixing.

Notes: A new two-color n-type GaAs/AlAs/AlGaAs double barrier quantum well (DBQW) and bound-to-miniband GaAs/AlGaAs quantum well infrared photodetector (QWIP) with photovoltaic (PV) and photoconductive (PC) dual-mode operation in the 3–5 and 8–14 μm atmospheric spectral windows has been demonstrated in this work. It consists of a stack of the midwavelength infrared (MWIR) QWIP and the long-wavelength infrared (LWIR) QWIP. The PV detection scheme uses transition from the ground bound state to the first quasi-bound excited state for the MWIR-QWIP. The PC detection scheme has two different transitions, one identical to the PV mode detection scheme while the other uses transition from the ground bound state to the miniband state of the superlattice barrier LWIR-QWIP. The peak responsivity for the PV mode was found to be 17 mA/W at λp=4.1 μm and T=50 K with a bandwidth Δλ/λp=15%. The peak responsivities for the PC mode were found to be 25 mA/W at λp=4.1 μm, Vb=1.0 V, and 0.12 A/W at λp=11.6 μm, Vb=3.2 V, and at T=50 K, with a bandwidth Δλ/λp=18%. The PV responsivity was found to be 68% of the PC responsivity at λp=4.1 μm and T=50 K, demonstrating the ability for efficient PV mode operation at 3–5 μm by using the DBQW structure. This is the highest ratio reported for the spectral region of 3–5 μm wavelength with a DBQW structure.

Notes: Electroluminescence (EL) in pure poly(N-vinylcarbazole) (PVK) has been observed in a simple single active layer EL device with the structure of ITO/PVK/Al (or Ca). The emitted light is violet with a maximum intensity at 426 nm; a part of the spectrum lies within the UV region. Blends of PVK with a conjugated-nonconjugated multiblock copolymer (CNMBC) previously described are also electroluminescent. In these blends a new emission peak in the EL spectra appeared and can be attributed to the emission from an exciplex formed by the two polymers. As the composition of the aforementioned blend changes from a PVK-poor to a PVK-rich ratio, the emitted light changes from green to blue. Further, blends containing 97 wt. % PVK and 3 wt. % CNMBC yielded an EL spectrum with a single emission peak different in location from either that of pure PVK or the pure CNMBC; the latter is blue with a maximum intensity at 476 nm. The EL output of the blends can be clearly seen in a brightly lit room at ambient temperature. An electron transport layer consisting of butyl-2-(4-biphenyl)-5-(4-tert-butylphenyl-1,3,4-oxadiazole) (PBD) in PMMA on the cathodic side of the EL device greatly increased the brightness of the emitted light to 200 cd/m2 in the case of a PVK-rich blend, but it had less effect on a PVK-poor blend or in pure PVK and had little effect on the pure form of the CNMBC.

Notes: Powdered samples of the type Ce1−xRExO2−y, where RE=La, Pr, Nd, Eu, Gd, and Tb, are synthesized over the range 0≤x≤0.5 starting from nitrate solutions of the rare earths. X-ray diffraction and Raman scattering are used to analyze the samples. These compounds, at least in the low doping regime and for strictly trivalent dopants, form solid solutions that maintain the fluorite structure of CeO2 with a change in lattice constant that is approximately proportional to the dopant ionic radius. The single allowed Raman mode, which occurs at 465 cm−1 in pure CeO2, is observed to shift to lower frequency with increasing doping level for all the rare earths. However, after correcting for the Grüneisen shift from the lattice expansion, the frequency shift is actually positive for all the strictly trivalent ions. In addition, the Raman line broadens and becomes asymmetric with a low frequency tail, and a new broad feature appears in the spectrum at ∼570 cm−1. These changes in the Raman spectrum are attributed to O vacancies, which are introduced into the lattice whenever a trivalent RE is substituted for Ce4+. This conclusion is supported by a simple model calculation of the effects of O vacancies on the Raman spectrum. The model uses a Green's function technique with the vacancies treated as point defects with zero mass.

Notes: The pseudospark cathode has the remarkable property of macroscopically homogeneous electron emission at very high current density ((approximately-greater-than)1 kA/cm2) over a large area (some cm2). The model of electron emission presented here is based on the assumption that the pseudospark microscopically utilizes explosive arc processes, as distinct from earlier models of "anomalous emission in superdense glow discharges.'' Explosive emission similar to vacuum are cathode spots occurs rapidly when the field strength is sufficiently high. The plasma remains macroscopically homogeneous since the virtual plasma anode adapts to the cathode morphology so that the current is carried by a large number of homogeneously distributed cathode spots which are similar to "type 1'' and "type 2'' spots of vacuum arc discharges. The net cathode erosion is greatly reduced relative to "spark gap-type'' emission. At very high current levels, a transition to highly erosive spot types occurs, and this "arcing'' leads to a significant reduction in device lifetime. Assuming vacuum-arc-like cathode spots, the observed current density and time constants can be easily explained. The observed cathode erosion rate and pattern, recent fast-camera data, laser-induced fluorescence, and spectroscopic measurements support this approach. A new hypothesis is presented explaining current quenching at relatively low currents. From the point of view of electron emission, the "superdense glow'' or "superemissive phase'' of pseudosparks represents an arc and not a glow discharge even if no filamentation or "arcing'' is observed.

Notes: The effects of dilution gases on hydrogenated amorphous silicon nitride (a-SiNx:H) films were investigated. Silane and ammonia were used as the reactive species, while nitrogen, helium, hydrogen, and argon were used as the dilution gases in a plasma-enhanced chemical vapor-deposition system at a substrate temperature of 300 °C. The electrical, physical, and chemical properties of the a-SiNx:H films were found to be highly sensitive to the various kinds and flow rates of the carrier gases in the deposition. Additionally, the physical properties of growth rate, refractive index, and etching rate were also investigated. The hydrogen bonding configuration was explored by infrared spectroscopy. The total hydrogen concentrations for all a-SiNx:H films were observed to be smaller than 3.0×1022 cm−3. The electrical properties were characterized by I-V and C-V measurements in metal-insulator-semiconductor structures. The breakdown strength was determined at the current density of 3 mA/cm2; in addition, the dominant mode of electronic conduction would appear to be the Poole–Frenkel emission. The interface trap state density Dit which ranged from 3.4×1011 to 1.3×1012 cm−2 eV−1 was evaluated by the C-V characteristics. Finally, the influences of the gas dilution in the a-SiNx:H films, as applied to the devices, were investigated by using the hydrogenated amorphous silicon thin-film transistors (a-Si:H TFTs). Analyses of the transfer characteristics of the TFT devices revealed that the density of deep gap states is 4×1012 cm−2 eV−1 and the field-effect mobility μFE is changing from 0.37 to 1.45 cm2/V s.

Notes: A careful examination of the dissolution/diffusion through liquid/precipitation steps of the well-accepted vapor-liquid-solid whisker growth mechanism has been conducted. Various diffusional processes involved in the whisker growth, such as solid surface diffusion, liquid surface diffusion, and bulk liquid diffusion, have been evaluated. It is noted that the diffusion along liquid alloy droplet surfaces, driven by both a concentration gradient and surface energy decrease, should not be ignored. Mass transport rate calculations using a simplified one-dimensional model show that the surface diffusion becomes more important with decreasing whisker diameter, decreasing solubility of growth species in the liquid phase, and increasing ratio of surface diffusion coefficient to bulk liquid diffusion coefficient. The diffusion along the liquid alloy droplet surface may play a major role in the growth of compound whiskers. Inclusion of the liquid droplet surface diffusion in the vapor-liquid-solid mechanism is proposed.

Notes: We studied interfacial reactions in Co/amorphous Si(a-Si) multilayers by transmission electron microscopy. We found that an intermixed layer of amorphous cobalt silicide formed in the as-deposited state. To explain the solid-state amorphization reaction, two parameters were used. They were the thermodynamic driving force (heat of formation) and the interfacial energy. The initial amorphization reaction in Co/a-Si multilayers was thermodynamically and kinetically favored. However, the formed amorphous interlayer remained about 1 nm thick and did not grow thicker with increasing modulation period and annealing temperature. The reason for this phenomenon was that the amorphous interlayer acted as a diffusion barrier to impede the amorphization reaction in Co/a-Si multilayers. Co2Si phase was always the preferred phase in the crystallization process for different average compositions of the multilayers. The mechanism that controlled the phase selection in Co/a-Si interfacial reaction was interpreted by using the model of modified heat of formation.

Notes: Most device structures based on strained epitaxial layers are capped by a second, unstrained layer to increase the mechanical stability of the structure. In order to calculate the energies of these structures it is necessary to synthesize the total energy from the energies of the line defects they contain (interfacial dislocations and dislocation dipoles). The self energies and interaction energies of dislocations and dipoles are calculated and their behavior examined as a function of their spacing and the thicknesses of the strained and capping layers. The results confirm the observations that capped strained layers are more stable than uncapped ones (of the same strained layer thickness) and that capping layers do not need to be thicker than approximately three times the strained layer thickness. An expression is deduced for the total energy of finite, nonuniform arrays of dipoles in capped layers and, by analogy with a similar earlier expression for dislocations in uncapped layers, it is concluded that the effect of a nonuniformity in the dipole spacing will be to increase the energy of the system compared with that of a uniform array having the same average spacing. The results in this paper can be used to assess the stability of devices and their rate of degradation by strain relaxation.

Notes: We characterize the development of nanometer scale topography (roughness) on SiO2 surfaces as a result of low energy, off-normal ion bombardment, using in situ energy dispersive x-ray reflectivity and atomic force microscopy. Surfaces roughen during sputtering by heavy ions (Xe), with roughness increasing approximately linearly with ion fluence up to 1017 cm−2. A highly coherent ripple structure with wavelength of 30 nm and oriented with the wave vector parallel to the direction of incidence is observed after Xe sputtering at 1 keV. Lower frequency, random texture is also observed. Subsequent light ion (H, He) bombardment smoothens preroughened surfaces. The smoothing kinetics are first order with ion fluence and strongly dependent on ion energy in the range 0.2–1 eV. We present a linear model to account for the experimental observations which includes roughening both by random stochastic processes and by development of a periodic surface instability due to sputter yield variations with surface curvature which leads to ripple development. Smoothing occurs via ion bombardment induced viscous flow and surface diffusion. From the smoothing kinetics with H and He irradiation we measure the radiation enhanced viscosity of SiO2 and find values on the order of 1–20×1012 N s m−2. The viscous relaxation per ion scales as the square root of the ion induced displacements in the film over the range of the ion penetration, suggesting short-lived defects with a bimolecular annihilation mechanism. The surface instability mechanism accounts for the ripple formation, while inclusion of stochastic roughening produces the random texture and reproduces the observed linear roughening kinetics and the magnitude of the overall roughness.

Notes: A procedure for fitting of Monte Carlo calculated impact ionization coefficient to experimental data has been proposed. This procedure has been applied for optimization and sensitivity analysis of fitting parameters of impact ionization model for electrons in Si. Strong correlation between threshold energy and preexponential factor of the impact ionization model and redundancy of power exponent have been found. A wide range of data on impact ionization coefficient can be fitted by adjusting the parameters of the microscopic impact ionization model.

Notes: The generation of delocalized E'δ defects through hole injection in buried SiO2 (BOX) layers of standard separation-by-implantation-of-oxygen structures has been studied by electron spin resonance. Selective photoinjection of holes was carried out under positive biasing of deposited semitransparent electrodes. This leads to the production of numerous delocalized E'δ centers (density≈1.8×1013 cm−2) in addition to a large density (∼9×1013 cm−2) of the common E'γ defect (O3≡Si⋅+Si≡O3)—a result qualitatively similar to previous glow discharge damage data. Comparison of damaging agents applied to BOX indicates that E'γ and E'δ defects are generally produced in an approximately fixed relative ratio. The comparison further bears out that, while energetic photons (x and γ) and ion bombardment activate similar E'δ defects of axial symmetry, hole injection generates a more isotropic type, termed E'δ2. It reveals the tendency that as the damaging species gets "softer,'' the g matrix anisotropy fades together with the disorder-induced g spread.

Notes: We present the first Raman-scattering studies of the behavior of the intrasubband plasmon mode of a two-dimensional electron gas which is undergoing lateral drift in an applied electric field. The data clearly show the expected Doppler shifts of the modes traveling up- and downstream, together with the expected dependence on the wave vector, but at higher drift velocities, the expected linear shift is distorted because of electron heating effects.

Notes: The behavior of electromigration-induced voids in narrow, unpassivated aluminum interconnects is examined, using scanning electron microscopy. Some electromigration tests were interrupted several times in order to observe void nucleation, void growth, and finally the failure of the conductor line. It is found that voids which opened the line have a specific asymmetric shape with respect to the electron flow direction. Besides void nucleation and void growth, void shape changes can consume a major part of the lifetime of the conductor line. A first attempt to model these processes on the basis of diffusion along the void surface shows that voids with a noncircular initial shape tend to produce the fatal asymmetry due to electron wind effects, with the anisotropy of surface energy possibly playing only a minor role.

Notes: Core-level photoemission peaks taken on several insulators with spatial resolution of 30–300 μm rigidly shift in energy from place to place, most likely because of local charging effects. Such phenomena can affect the linewidth and the general lineshape of conventional, spatially integrated spectra even when a flood gun is used, which is one of the standard remedies against charging. This may require a critical revision of photoemission linewidths and lineshapes for many past experiments on insulators.

Notes: Cathodoluminescence in the scanning electron microscope has been used to investigate the relationship of point defects in CdTe and CdTe:V with luminescence bands at 1.40 and 1.13 eV. V has been found to inhibit the 1.40 eV luminescence. Annealing experiments indicate that Cd and Te vacancies are involved in the mentioned emission bands.

Notes: Techniques for visualization and quantification of the latent electric charge image on dielectric materials have been developed using Pockels effect of a BSO (Bi12SiO20) single crystal, a charge coupled device camera recording system, and a computer image processing technique. The Pockels device of the BSO single crystal was used as a transducer to transfer the surface charge density to retard the transmitted light beam. The basic principle of advanced electro-optical surface charge measurement is described and typical surface charge distributions of positive and negative discharge streamers produced by an ac (50 Hz) high voltage are demonstrated. The measured surface charge images are shown on a computer monitor as three-dimensional figures.

Notes: Time-resolved impact excitation and de-excitation processes of the 1.54 μm electroluminescence (EL) emission of Er3+-doped InP are investigated. Samples are impact excited by applying electrical pulses and the time response of the EL emission is measured in the temperature range from 77 K to 330 K. The decay of the emission proves almost exponential in all the temperature ranges and shows little thermal quenching with the decay time decreasing from 2 ms at 77 K to only 1 ms at 330 K. This result contrasts with the large thermal quenching and nonexponential characteristics of the photoluminescence (PL) time decay at high temperatures in the same sample, suggesting different excited Er3+ centers between EL and PL. Also measured is the emission rise time as a function of excitation pulse current, giving us the impact cross section of 9×10−16 cm2 for Er3+ ions in InP. The excitation and quenching processes as well as the efficiency of Er emission are analyzed. A model taking into consideration the presence of different Er centers explains the different behaviors in the time responses between EL and PL.

Notes: Tm3+ doped zirconium barium lanthanum chloro-fluoride, germanate, and tellurite glasses were prepared and their upconversion luminescence and excitation spectra were measured to investigate the effect of the host matrix glass on the branching ratio of 0.36 and 0.45 μm luminescences. Since the radiative emission probabilities of electric-dipole transitions are functions of Judd–Ofelt intensity parameters, Ωt(t=2,4,6), absorption spectra were also measured to obtain Ωt parameters. The branching ratio β of 0.45 μm luminescence in zirconium barium lanthanum chloro-fluoride glasses increases slightly with increasing Cl− doping, but the ratios were much smaller than those in tellurite and germanate glasses. The result could be explained by the smaller Ω2/Ω4 ratio in zirconium barium lanthanum chloro-fluoride glasses than those in oxide glasses. The correlation between the luminescence intensity ratio and the Ωt parameters suggests that the branching ratio can be controlled by the ligand field for Tm3+ in glass matrix.

Notes: The deposition and properties of in situ boron-doped polycrystalline silicon (poly-Si) films grown at 550 °C were investigated using an ultrahigh vacuum chemical vapor deposition system. It is observed that, if the doping level is high enough, the boron incorporation would significantly reduce the deposition rate, impede the grain growth, and degrade the crystallinity of the poly-Si films. We also found that the preferential adsorption of boron atoms on the SiO2 surface at the initial stage of deposition shortened the incubation time of deposition. The property of trapping states at the grain boundary is also examined, and a density of about 4.7×1012 cm−2 is obtained for poly-Si films with a doping level less than 2.2×1019 cm−3.

Notes: Erbium-doped GaAs layers were grown by metalorganic vapor phase epitaxy using two new sources, bis(i-propylcyclopentadienyl)cyclopentadienyl erbium and tris(t-butylcyclopentadienyl) erbium. Controlled Er doping in the range of 1017–1018 cm−3 was achieved using a relatively low source temperature of 90 °C. The doping exhibits a second-order dependence on inlet source partial pressure, similar to behavior obtained with cyclopentadienyl Mg dopant sources. Equivalent amounts of oxygen and Er are present in "as-grown'' films indicating that the majority of Er dopants probably exist as Er-O complexes in the material. Er3+ luminescence at 1.54 μm was measured from the as-grown films, but ion implantation of additional oxygen decreases the emission intensity. Electrical compensation of n-type GaAs layers codoped with Er and Si is directly correlated to the Er concentration. The compensation is proposed to arise from deep centers associated with Er which are responsible for a broad emission band near 0.90 μm present in the photoluminescence spectra of GaAs:Si, Er films.

Notes: The layer thicknesses and composition of molecular beam epitaxy grown four period 200 A(ring)/100 A(ring) GaAs/InGaAs superlattice structures with nominal indium concentrations of 10%, 15%, and 20% were determined by transmission electron microscopy, Rutherford backscattering spectroscopy, double crystal x-ray diffraction (DXRD), photoreflectance (PR), and photoluminescence (PL). The results show that the indium concentration obtained by DXRD is a little low and that obtained by PR and PL is a little high, and that the discrepancies are larger for the larger indium concentrations. We show that both discrepancies can be accounted for by relaxation of the lattice, elastic relaxation as represented by a radius of curvature, and/or plastic deformation as represented by mismatch of dislocations. For the case of elastic relaxation the tetragonal distortion is less than it would be if the sample were perfectly pseudomorphic. The fractions by which it is reduced for the 10%, 15%, and 20% samples was 0.91, 0.86, and 0.77 as determined by DXRD and 0.80, 0.78, and 0.85 as determined by PR/PL.

Notes: Two Si atomic layer epitaxy schemes based on Cl/H exchange chemistry are compared by deposition of Si on Ge(100). Time-of-flight direct recoiling and reflection high-energy electron diffraction are used to characterize the very thin Si layers. It is shown that alternating exposure to SiCl2H2 and atomic hydrogen deposits Si in a process that is self-limiting. Growth of Si by this method results in either alloy formation or Si island growth as low as 465 °C on Ge(100), which prohibits a simple measurement of Si thickness/cycle. In contrast, alternating exposure to Si2Cl6 and Si2H6 below 500 °C results in Si deposition with uniform thickness. Plots of direct recoil intensity versus incident angle (with respect to the surface plane) are a sensitive probe of the thin layer morphology.

Notes: The growth rates RSi of Si films deposited on Si(001)2×1 from Si2H6 by gas-source molecular-beam epitaxy were determined as a function of temperature Ts (500–950 °C) and impingement flux JSi2H6 (0.3–7.7×1016 cm−2 s−1). RSi (Ts,JSi2H6) curves were well described using a model, with no fitting parameters, based upon dissociative Si2H6 chemisorption followed by a series of surface decomposition reactions with the rate-limiting step being first-order hydrogen desorption from Si monohydride. The zero-coverage Si2H6 reactive sticking probability in the impingement-flux-limited growth regime was found to be 0.036, more than two orders of magnitude higher than that for SiH4. B doping concentrations (CB=5×1016–3×1019 cm−3) from B2H6 increased linearly with increasing flux ratio JB2H6/JSi2H6 at constant Ts and decreased exponentially with 1/Ts at constant JB2H6/JSi2H6. Secondary ion mass spectrometry analyses of modulation-doped samples revealed sharp profiles with no detectable B segregation. Hole mobilities in uniformly doped samples were equal to bulk values.

Notes: We reevaluate the current model of noise gain and photoconductive gain for AlGaAs/GaAs quantum well infrared photodetectors. An experimental study is carried out on samples covering different types of intersubband transitions, i.e., bound-to-continuum, bound-to-quasibound, and bound- to-bound cases. A general difference between gains measured by dark current noise and by responsivity was found for all three cases, while agreement between them was found within some ranges of bias voltages for the bound-to-quasibound and bound-to-continuum samples. The thermal noise contribution to the total white noise is explicitly modeled and discussed.

Notes: Four-terminal microcontacts between metallic electrodes develop nonlinear current-voltage dependencies both in the source and control channels as well as between the channels. Theory is presented of the nonlinearity caused by the reabsorption of nonequilibrium phonons emitted in the contact by injected electrons. Temperature of the lattice due to heating by the current is of the order T∼eV/4, which results in substantial increase of the resistance both in the bias direction and in the direction perpendicular to the bias. Performance characteristics of such a device at low temperature compared to the Debye temperature are quite promising for frequencies below 109 Hz.

Notes: We investigate the effect of a Rb overlayer on the cubic SiC(100) surface oxidation by core level photoemission spectroscopy. The presence of Rb monolayer on the β-SiC(100) surface enhances the oxygen uptake by four orders of magnitude at room temperature and results in the selective formation of high silicon oxidation states. Furthermore, the Rb catalyst could be desorbed from the surface by rapid thermal annealing at moderate temperature, leading to the formation of a Si oxide layer on SiC(100) with no carbon atom present on the surface and/or in the oxide layer.

Notes: We present a new type of Josephson transmission line (JTL) in which discrete multilayer Josephson junctions are employed instead of the standard single tri-layer junction technology. It is shown via numerical simulation that such a structure supports the propagation of multiple flux quantum solitons, which are represented in this system by a 2πn kink in the phase difference across a multilayer junction of n superconductor-insulator-superconductor layers. As a practical aspect, parallel and series connected multilayer junctions in a JTL can significantly reduce the need for microstrip line inductance, thereby providing the optimal conditions for large scale integration of JTL-based logic and memory circuitry. In this article we investigate in detail through numerical simulation the properties of JTL designs with multilayer junctions, incorporating expected variations in junction characteristics. A brief description of the current fabrication of multilayer junctions for flux quantum circuitry will be provided.

Notes: Transport critical current measurements have been carried out on melt-processed thick films of YBa2Cu3O7−δ on yttria-stabilized zirconia in fields of up to 8 T both within grains and across grain boundaries. These measurements yield Jc values of ∼3000 A cm−2 at 4.2 K and zero magnetic field and 400 A cm−2 at 77 K and zero magnetic field, taking the entire sample width as the definitive dimension. Optical and scanning electron microscopy reveals that the thick-film grains consist typically of a central "hub'' region ∼50 μm in diameter, which is well connected to radial subgrains or "spokes'' which extend ∼1 mm to define the complete grain structure. Attempts have been made to correlate the transport measurements of inter- and intra-hub-and-spoke (H-S) critical current with values of this parameter derived previously from magnetization measurements. Analysis of the transport measurements indicates that current flow through H-S grains is constrained to paths along the spokes via the grain hub. Taking the size of the hub as the definitive dimension yields an intra-H-S grain Jc of ∼60 000 A cm−2 at 4.2 K and 0 T, which is in reasonable agreement with the magnetization data. Experiments in which the hub is removed from individual grains confirm that this feature determines critically the Jc of the film.

Notes: Real-space transfer devices using strained InxGa1−xAs (x=0.15, 0.25) channel and unstrained GaAs channel are fabricated. The electron transfer and negative differential resistance of InGaAs channel devices are enhanced. The InGaAs channel devices with a larger indium mole fraction show lower leakage currents and can be operated at a higher collector voltage. For operation in the negative resistance field-effect transistor mode, the drain current peak-to-valley ratios of strained InGaAs channel devices are larger than that of unstrained GaAs channel devices. The peak-to-valley ratio can be increased more than two orders of magnitude by using strained InGaAs channel layers.

Notes: Beneficial features of boron multiple delta doping (MDD) in synthetic diamond thin films are studied analytically and compared with MDD in GaAs. In spite of a deep boron level (∼0.3 eV), MDD greatly helps the thermal excitation of holes via elevation of the Fermi level toward the acceptor boron level. Thus, a hole excitation 6–7 times higher than that of the uniformly doped one is obtained. Furthermore, more than 90% of holes are in the spacer layer (i-diamond) where the mobility is high, resulting in a film conductance of the MDD structure more than 20 times higher than that of the conventional one when the same amount of boron is uniformly doped.

Notes: A ferro-antiferroelectric Li0.7Na0.3NbO3 solid solution crystal was grown from the melt by the Czochralski method. It has been found from the high temperature x-ray diffraction that the crystal is orthorhombic with a unit cell of dimensions a=5.374(9) A(ring), b=5.152(1) A(ring), and c=16.738(0) A(ring) at 25 °C, and undergoes a structural phase transition from orthorhombic to tetragonal at 480 °C. It was confirmed from the observation of the D-E hysteresis loop and the pyroelectric measurements that the mixed crystal Li0.7Na0.3NbO3 undergoes a first order ferroelectric phase transition at 480 °C. In the dielectric measurements, the dielectric behavior in the vicinity of the phase transition temperature reveals a diffusive character and also a pronounced dielectric dispersion due to an ionic hopping observed over the wide temperature range near 298 °C.

Notes: An optical band-gap shrinkage was found in highly transparent and conducting thin films deposited by the Pyrosol process. The narrowing appears at carrier concentrations which are consistent with a metal-semiconductor transition. The gap-shrinkage dependence on carrier concentration is consistent with the model proposed by P. E. Schmid [Phys. Rev. B 23, 5531 (1981)] for Si:As and Si:B.

Notes: We present results from a numerical solution of a resonant tunneling diode structure, using a time-independent Schrödinger equation without considering scattering. Intrinsic circuit parameters such as conductance, capacitance, and inductance are extracted from the numerical results and compared with experimental data obtained from dc and microwave characterization of GaAs/AlGaAs devices. A study of the influence of different lightly doped spacer layer thicknesses on the emitter side reveals the relationship between spacer layer thickness and the circuit parameters. From these results, we provide explanations for the behavior of the biased diode circuit parameters. The simulated data show good qualitative agreement with experimental results; possible explanations for deviations between experimental values and simulated data, including inelastic scattering and series resistance effects, are discussed.

Notes: Copper-containing polymer thin films sandwiched between metal electrodes were prepared by plasma polymerization of metal compounds in a radio-frequency glow discharge. Current-voltage (I-V) characteristics reveal the switching phenomenon, characterized by an abrupt impedance variation from the off state (about 108–1010 Ω) to the on state (about 10–102 Ω). The I-V characteristics exhibit an ohmic behavior in the on state. The pressure dependence of the switching threshold voltage (Vth) in the range of 0.1–80 MPa was investigated. It was assumed that Vth variations were due to the sample compression effect by the pressure. Results are analyzed by using Abeles' model.

Notes: The bolometric nature of the optical sensitivity of YBCO (YBa2Cu3O7−δ) thin-film microstrips in the superconducting state is demonstrated. Below the critical temperature Tc the critical current temperature dependence of the device provides a suitable temperature sensor. A theoretical thermal model that gives reliable forecasts for the sensor time constant and sensitivity over the 10–90 K temperature range has been developed. Sensitivity measurements on high-quality YBCO films deposited on MgO are in quite good agreement with the model, showing that the observed signals are entirely bolometric. The effective heat capacity of the sensor is that of the film in the irradiated area. Owing to the high thermal conductivity of MgO, the substrate stays at the heat sink temperature and the main resistance to heat flow is the film/substrate interface. The heat capacity of the irradiated area of the film, strongly coupled to the heat sink through the thermal boundary resistance gives high-speed bolometers behaving as first-order systems. For SrTiO3 or ZrO2 substrates, their lower thermal conductivity would lead to higher time constants and more complicated behaviors. The precise origin of the measured sensor noise has still to be established. On a 0.2×10×15 μm3 microbridge at 85 K (below the critical temperature), a noise equivalent power of 120 pW/(square root of)Hz and a time constant τ of 20 ns were measured; this corresponds to a noise equivalent temperature of 60 nK/(square root of)Hz and a specific detectivity D* of 1.8×106 cm (square root of)Hz/W. A more relevant criterion for comparing optical sensors is the D*/(square root of)τ ratio. In the case presented, this ratio is equal to 1.3×1011 cm/J which is one of the best values published at this time. Such performances should give competitive composite or antenna coupled bolometers compared to existing sensors for wavelengths greater than 20 μm.

Notes: A comparison between the optical transmission and the photoconductive response of p-type quantum well infrared photodetectors is demonstrated. The dependence on polarization was found to be different for these two processes. Light polarized perpendicular to the growth layers was more efficiently absorbed by the quantum wells, while the photoconductive response was stronger for the parallel polarized light. The difference is attributed to the different vertical transport behavior of light and heavy holes and the mixed nature of hole subbands.

Notes: The third-order nonlinear optical susceptibility χ(3)(3ω) due to the intersubband transitions in the four-level AlInAs/GaInAs compositionally asymmetric coupled quantum well (CACQW) is investigated theoretically. The subband eigenenergy En of the CACQW structure could be designed to form an equally spaced energy-level ladder. Since the eigenenergy spacing could be designed to resonate with the pumping source, the third-order nonlinear optical susceptibility could be greatly enhanced through the triple resonance. Based on the theoretical calculations, a magnitude of ||χ(3)(3ω)|| as high as 2.2×105 (nm/V)2 can be achieved for the CACQW structure. This is a more than eight orders of magnitude enhancement as compared to that of the bulk value in GaAs. In addition to the design of CACQW structure, the triple resonance can also be achieved by biasing the CACQW under a proper electric field due to the large Stark effect of the CACQW structure.

Notes: The amorphous silicon hydrogen alloys (a-Si:H) prepared by plasma-enhanced chemical-vapor deposition are implanted with various impurities, i.e., phosphorus, silicon and boron, followed by 250 °C thermal annealing and hydrogen passivation. A critical phosphorus dosage of 5×1015 cm−2 is found, beyond which the doping effect becomes evident and the electrical properities of the films are comparable to the n+ a-Si:H deposited by glow-discharge decomposition of SiH4 and PH3. The silicon and boron implantation have less effect since the dosage is below 1015 cm−2. It is found that the conductivity of the implanted sample after annealing is determined by three competing mechanisms, i.e., annihilation of the implant-induced defects, defect creation due to hydrogen evolution, and the impurity activation. From the IR spectra of the high dosage phosphorus-implanted samples, a broad peak due to PHx radicals located from 2100 to 3500 cm−1 is found. In addition, most of the voids are found to be filled by extra P ions during annealing, so very few hydrogen atoms can be driven into the film after hydrogen passivation. In photoluminescence spectra, the high-energy peak at 1.4 eV drops very quickly and the low-energy shoulder at 1.25 eV becomes evident after implantation. It is found that Brodsky's quantum-well model can be successfully applied to explain the observed results.