Index of content:
Volume 58, Issue 1, 01 July 1985

The far field in terms of the spatial Fourier transform of the sources and its implications on the inverse problem
View Description Hide DescriptionIt has been known for some time that the range and phase normalized far field is proportional to the three‐dimensional spatial Fourier transform of the source distribution, taken on the surface of the Ewald sphere at a fixed frequency. Unfortunately, however, this very useful relationship is neither widely known, nor does there even exist for it a good, short, simple, and direct derivation in the open literature (thus making references to the relationship difficult). The purpose of this paper is to alleviate these difficulties by presenting a brief derivation of the relationship for both the scalar acoustic as well as the vector electromagnetic cases. A brief discussion of the implications of this relationship on the inverse source and inverse scattering problem is also presented.

Statistical procedure for deconvoluting experimental data
View Description Hide DescriptionA general procedure of the deconvolution of noisy experimental data is proposed. The deconvoluted solution is represented in terms of a cubic spline function f(x) with variable knots. The value of parameters and the location of knot points in the spline are determined by solving the nonlinear least‐squares problem of fitting g(x)=∫κ(τ)f(s(τ,x))dτ to the data, where experimental data g(x) represent the convolution of f[s(τ,x)] with an apparatus weighting function k(τ) and s(τ,x) is a function related to the configuration of the experimental system. The profile of the fitted solution varies widely with different choices of the number and position of knots in the spline. To obtain a good solution, the optimum number of knots is determined based on the data by applying minimum Akaike’s Information Criterion procedure. Practical application of the variable‐knot spline is generally considered to be rather intricate. However, by taking advantage of an advanced technique of a quasi‐Newton method, the nonlinear least‐squares problem becomes tractable, and excellent smoothing in the deconvoluted solution can be achieved.

Time‐resolved proton focus of a high‐power ion diode
View Description Hide DescriptionAn improved understanding of the factors that control the axial focus of applied‐B ion diodes was obtained from time‐resolved diagnostics of ion‐beam trajectories. This resulted in a new selection of anode shape that produced a proton focus of 1.3‐mm diameter from a 4.5‐cm‐radius diode, which is a factor of 2 improvement over previous results. We have achieved a peak proton power density of 1.5±0.2 TW/cm^{2} on the 1‐TW Proto I accelerator. The radial convergence of this proton beam, defined as the ratio of the anode diameter to focused beam FWHM, is 70. Time‐resolved information about virtual cathode evolution, the self‐ and applied‐magnetic‐field bending, and the horizontal focus of the beam was also obtained. In addition, the diffusion of the magnetic field into the anode plasma is estimated by measuring the horizontal focal position as a function of time. Finally, we discuss the effects of gas cell scattering on the beam focus.

A virtual‐cathode reflex triode for high‐power microwave generation
View Description Hide DescriptionMicrowavegeneration by a relativistic electron beam injected through a screen or foil into a cylindrical waveguide is investigated. The electron beam is generated by a field‐emission diode with an anode‐cathode gap of ∼1 cm. The electron density exceeds the space‐charge limit for the beam, which forms a virtual cathode inside the waveguide. The instability of the virtual cathode causes oscillations which couple to bunched reflexing electrons trapped in the potential well. Frequency locking between the oscillating virtual cathode and the reflexing electrons is found to generate a much purer, narrowband microwave output. Using diode mismatch to increase the beam impedance is shown numerically to be a useful technique to improve the microwavegeneration efficiency and spectral purity. The peak conversion efficiency is calculated to be 2.2% at 7 GHz. The numerical results are compared to experimental results, and the agreement is within 8% in frequency and 25% in peak power. Experimentally, we describe the qualitative effects of different cathode materials and dimensions. The anode materials used were screens and foils of various transparencies and thicknesses. Microwavegeneration efficiency dropped as the mean scattering angle increased with the thicker foils. Also, the increased transparency of fine screens caused an upward frequency shift due to the increased electron density.

Low threshold current density of GaInAsP visible injection laser diodes lattice matched with (100) GaAs emitting at 705 nm
View Description Hide DescriptionThe shortest possible lasing wavelength and the lowest threshold current density of the GaInAsP visible injection laser diodesgrown on (100) GaAs substrate by the liquid phase epitaxy using a two‐phase solution growth technique are discussed. The lasing wavelength and threshold current density at room temperature under pulsed conditions are 705 nm and 9.4 kA/cm^{2}, respectively. Light output power of 5–6 mW from one facet and maximum differential quantum efficiency of 38% are realized under the condition of fundamental lateral and transverse mode operation.

Optical measurement of the photorefractive parameters of Bi_{1} _{2}SiO_{2} _{0}
View Description Hide DescriptionWe employ an optical technique requiring no electrodes to measure three parameters characterizing photorefractive traps in nominally undoped n‐type bismuth silicate at 515‐ and 488‐nm wavelengths. These parameters are (1) an effective density of active electron trap sites (∼10^{1} ^{6} cm^{−} ^{3}), (2) the contribution α_{ p }cm^{−} ^{1} of the photorefractive excitations to the total absorption coefficient (∼3.8 cm^{−} ^{1} or 54% at 488 nm and 1.4 cm^{−} ^{1} or 86% at 515 nm), and (3) the average distance over which an optically excited electron moves in the absence of electric fields before being retrapped (∼4 μm). With these parameters (and the crystal properties) standard models give predictions for all photorefractive effects, with or without applied electric fields and crystal motion, provided that there is negligible change in the density of occupied traps during an electron recombination time (∼microseconds). Our technique employs measurement of the light‐induced exponential decays of spatial ‘‘gratings’’ of trapped charges as a function of the grating period. The expected three‐parameter function is observed from which we determine the above three parameters.

Two‐beam coupling in photorefractive Bi_{1} _{2}SiO_{2} _{0} crystals with moving grating: Theory and experiments
View Description Hide DescriptionLarge values of the exponential gain coefficient Γ are obtained (Γ≂8–12 cm^{−} ^{1}) when recording with a moving grating in photorefractive BSO crystals (nearly degenerate two‐wave mixing; drift recording mode). The resolution of the Kukhtarev’s equations with a moving grating shows a resonance effect which at the optimum velocity makes the modulation of the photoinduced space charge field E _{sc} higher. An optimum of the grating spacing also exists: Λ_{opt}:2π(E _{0}/N _{ A })(με/eγ_{ R })^{1/2}. In such conditions, the space charge field is phase shifted by π/2 with respect to the incident fringe pattern; this allows an efficient beam coupling between the two recording beams. The dependence of the gain Γ versus the incident beam ratio β of the two interfering waves is interpreted by including the second‐order term in the Fourier development of E _{sc}. The conditions allowing one to obtain a reasonable agreement between the theory and experiments are presented and discussed, as well as the adopted values of the crystals’ parameters.

Diffusion theory of the low‐pressure positive column with two‐step ionization from a metastable state. I
View Description Hide DescriptionWe present a self‐consistent theory of a low‐pressure positive column with two‐step ionization from a metastable state in which the coupling between the continuity equations for the electrons and the metastables is appropriately taken into account. The corresponding boundary value problem is solved numerically for both cylindrical and planar geometries and vanishing electron and metastable densities at the boundary, for a wide range of the two parameters on which the problem is shown to depend. Curves of the electron and metastable density profiles, of the eigenvalues, and of conveniently defined generalized diffusion lengths for both types of particles are presented. We also show that the eigenvalues together with Boltzmann calculations of the electron rate coefficients and transport parameters provide similarity laws for a steady‐state positive column with radius R: ratio of the electric field to the gas density, E/N vs N R for I/R=const and ratio of the metastable density to the gas density, n _{ M }(0)/N vs I/R for N R=const, where I is the discharge current. An application to argon is discussed.

Diffusion theory of the low‐pressure positive column with two‐step ionization from a metastable state. II. Approximate analytical solutions for planar geometry
View Description Hide DescriptionWe present two approximate analytical solutions of the boundary value problem that models the low‐pressure positive column with two‐step ionization from a metastable state in the case of planar geometry and compare them with the numerical results of Part I.

Analysis of grating couplers for planar dielectric waveguides
View Description Hide DescriptionOutput and input grating couplers for step index planar dielectricwaveguides are investigated taking into account the interaction between a guided mode with a finite number of radiative and an infinite number of nonradiative space harmonics. For transverse electric and transverse magnetic modes, and for shallow gratings of arbitrary profiles located at either the upper or the lower guide surface, the governing equation for the slowly varying amplitude of the guided wave is deduced, and therefrom the real and the imaginary parts of the correction to the wave number of unperturbed guide are evaluated, the power conservation relation is established, and the output and the input coupling efficiencies are defined. For the input coupler, the excitation by a two‐dimensional radiation beam is also included. The role of the space harmonics which, like the guided wave, carry power along the guide in modifying the real and the imaginary parts of the wave number of the guided wave is discussed. The effect of the choice of the profile and location of the grating on the distribution of the outgoing radiation between the cover and the substrate, and between harmonics of different orders are examined. In a multimode coupler, significant improvements in input coupling efficiency can be achieved by a suitable choice of the grating profile and the excitation region. The effect of tapering the relative depth of the grating on the amplitude profile of the radiated beam of the output coupler, and on the coupling efficiency of the input coupler are investigated.

Characteristics of light amplifier of AlGaAs semiconductor diode laser
View Description Hide DescriptionThis paper reports the experimental and analytical studies of the light amplification of the Fabry–Perot type lasers. Particularly, for high bias current level of the laser amplifier, say more than 90% of the threshold current, the waveform distortion of the amplified output becomes apparent. Considering the following facts that the time division spectra of the input pulse to the amplifier changes from shorter to longer wavelengths due to temperature increase in the active layer after the pulse current is turned on and the peaks of the amplification factors of the Fabry–Perot type amplifiers depend on the wavelength for the different bias current level, we can show that the waveform distortion is caused by the peaks of the spectrum shift of the input and the amplifier. We also observe an important fact that the effective bias current to the amplifier is not the same as the actual bias current, which causes the wavelength shift of the peak gain of the amplifier, because of the reduction of the carrier concentrations in the active layer due to stimulated emission by the input light pulse. Taking into account these facts, the analytical and experimental results are in good agreement with each other.

Cavity supermodes for gain‐saturated diode laser arrays
View Description Hide DescriptionWe present a numerical method for the solution of the cavity modes of a multichannel array of laser diodes. The analysis is based on coupled‐mode theory and includes the appropriate boundary conditions for the electric field at each reflecting facet. Saturable gain is also included, with saturation arising from both the forward‐going and reflected waves. We present results which differ from a recent analytic analysis for the case of two channels with different gains and propagation constants. The present work predicts ‘‘cavity supermodes’’ which are similar to the ‘‘symmetric’’ and ‘‘antisymmetric’’ waveguide supermodes previously described in the literature. These cavity modes are characterized by equal intensities in both channels even when the channel gains are grossly different.

Photoluminescence and stimulated emission in Si‐ and Ge‐disordered Al_{ x }Ga_{1−x }As‐GaAs superlattices
View Description Hide DescriptionPhotoluminescence and absorption data are presented on Al_{ x }Ga_{1−x }As‐GaAs superlattices(SLs) disordered into bulk‐crystal Al_{ y }Ga_{1−y }As (0≤y≤x) by Si or Gediffusion. The bulk‐crystal Al_{ y }Ga_{1−y }As produced by impurity‐induced disordering (by Al‐Ga interchange) is determined by transmission electron microscopy, absorption measurements, and photoluminescence to be homogeneous, with an alloy composition (y) that agrees with the average Al concentration of the SL. For low enough Al concentration (y≊0.23<y _{ c } =0.44, the direct‐indirect crossover), in absorption the Ge‐ or Si‐disordered SL exhibits (4.2 and 77 K) the bulk‐crystal exciton, which is characteristic of homogeneous alloy (Al_{ y }Ga_{1−y }As). Stimulated emission (4.2 and 77 K) in bulk‐crystal Al_{ y }Ga_{1−y }As is observed ΔE≤50 meV below the band edge via photopumping for both Si‐ and Ge‐disordered SLs of Al concentration yielding y∼0.23 and y∼0.39. Shallow hydrogenlike donor or acceptor states are characteristic of Al_{ x }Ga_{1−x }As‐GaAs SLs disordered with Ge or with Si. For the Si impurity (i.e., an Al_{ x }Ga_{1−x }As‐GaAs SL disordered with Si), however, much deeper states (transitions) are observed that saturate at higher photoexcitation levels. These states are attributed to nearest‐neighbor or extended Si‐Si pairs since similarly disordered Al_{ x }Ga_{1−x }As‐GaAs SLs doped with Ge do not exhibit deeper states.

Static stability of mercury mirror light modulators
View Description Hide DescriptionThis paper analyzes the static stability of a mercurydroplet held in a geometry consisting of two parallel gaps connected by a short tube. This arrangement finds application as a high‐efficiency optical modulator which works in reflection, with the mercurysurface acting as a mirror. Theoretical calculations of the pressure balance and the system energy are presented which lead to conclusions concerning the design of these devices.

Moderate‐power cw chemical oxygen‐iodine laser capable of long duration operation
View Description Hide DescriptionDesign and operation of a cw, transverse‐flow oxygen‐iodine chemical laser are reported. Reactant recirculation in the O_{2}(^{1}Δ) chemical generator allows long duration (up to 2 h) operation. Peak powers of 460 W, and stable outputs of 80 W have been demonstrated. Power extraction efficiencies of up to 65% were measured for 80‐W operation, and the implications of this observation for the I _{2}dissociation mechanism are discussed.

Optoacoustic generation and electrical detection of subnanosecond acoustic pulses
View Description Hide DescriptionThe optoacoustic generation and electrical detection of acoustic pulses of about 0.5‐ns duration in quartz and in several polymer materials is demonstrated. The ultrasonic pulses are generated by thermoelastic effects and/or ablation due to absorption of 70‐ps laser‐light pulses in an opaque coupling layer deposited on one of the sample surfaces. Detection of the acoustic pulses is possible by using electrically charged, piezoelectric, or biased samples which yield electrode signals upon transit of the pulses. The acoustic pulses have also have been used to measure ultrasonic velocity and attenuation in various materials.

Scattering matrix approach to thermal wave propagation in layered structures
View Description Hide DescriptionIn this paper we describe a new technique, based on Fourier optics, to explain the propagation, as well as loss, of three‐dimensional thermal waves in isotropic, homogeneous materials. Using this, the dependence of temperature distribution at any arbitrary infinite parallel plane on the aperture distribution is derived. In addition, the temperature distribution at the aperture plane, due to a given perpendicular source is formulated, by applying the boundary conditions in the spatial frequency domain. A scatteringmatrix theory is developed to analyze the propagation of thermal waves in multilayered structures. This directly relates the heat source characteristics to the temperature distribution at any level. The contrast mechanism in the subsurface mode of operation is explained, and the dependence of the response of a typical system on depth is explored. In addition, the theoretical amplitude and phase images of a cylindrical void are presented; the results are in good agreement with the published experimental subsurface images of voids. Finally, transform techniques in both temporal and spatial frequency domains are employed to analyze the pulse response of layered structures. It is found that the decay rate of the surface temperature is strongly influenced by the presence of inclusions within a sample.

Vacuum surface flashover: A high‐pressure phenomenon
View Description Hide DescriptionDesorbed gases from insulator surfaces may be much more important in flashover initiation than properties of the dielectricsurface. Ambient neutral densities, immediately above the insulator surface, in the initial stages of flashover have been calculated to lie in the range from 2×10^{1} ^{7}/cm^{3} to 2×10^{2} ^{1}/cm^{3} with a mean of about 1×10^{1} ^{9}/cm^{3}, whether flashover occurred on surfaces in air at atmospheric pressure or in vacuum as low as 10^{−} ^{6} Torr. The evidence presented indicates that, for dc and microsecond pulse voltages, surface flashover is a local, relatively high‐pressure phenomenon. Comparison of field strengths of surface flashover for various insulating materials perpendicular to the electrodes, operating pressures, and voltage waveforms appears to indicate that flashover values range well within an order of magnitude (∼18–80kV/cm). A filamentary nature of the flashover discharge is shown and is consistent with gas breakdown phenomena. Ionization wave front velocities are examined and these compare favorably with those obtained from gaseous breakdown studies.

A comparison between arc drops in ignited thermionic converters with and without ion reflections at the emitter
View Description Hide DescriptionThe output performance of two thermionic energy converters is compared. One converter has a normal emitter, working with zero field at the emitter which is close to the optimum working point, and the other has a low work function emitter and ion reflection at the emitter. A simple model of the plasma and the sheaths shows that a converter working with a low work function emitter and ion reflections gives a worse performance than a similar converter with a normal emitter.

Surface flashover of solid insulators in atmospheric air and in vacuum
View Description Hide DescriptionThe surface flashover of Teflon, plexiglass, quartz, Pyrex glass, Macor glass‐ceramic, and sapphire solid insulators has been measured in vacuum (∼10^{−} ^{8} Torr, ∼10^{−} ^{6} Pa) and in atmospheric air using dc, ac (60 Hz), and 1.2/50‐μsec lightning impulse voltages. The dependence of the flashover voltage on the following parameters is investigated: (1) spacer material, (2) diameter of the spacer, (3) spacer length, (4) number of spacers stacked in series, (5) air pressure in the range 10^{−} ^{6}–10^{5} Pa, (6) electrode material, (7) spark conditioning, and (8) the external resistance in series with the gap. At a fixed insulator length the flashover voltage decreases with increasing spacer diameter. The withstand voltage of spacers stacked in series increases with increasing the number of spacers. The dc flashover voltage of different insulating materials is theoretically calculated in vacuum as a function of the length of the insulator and compared with the experimentally obtained results. Good agreement is obtained.