A two-dimensional analysis of gallium arsenide junction field effect transistors with long and short channels

A two-dimensional analysis of gallium arsenide junction field effect transistors with long and short gates is presented. The short gate device is shown to have a negative resistance region at low drain voltages which is absent in the long gate device. The negative resistance region depends on the shape of the conducting channel and the distribution of the electron charge. A field dependent mobility is used, and its effect on current saturation is discussed. The electrostatic charge distribution and electric field distribution are calculated for both devices, and are shown in graphic form.     

A two-dimensional analysis of indium phosphide junction field effect transistors with long and short channels

A two-dimensional analysis of indium phosphide junction field effect transistors with long and short gates is presented. The two devices are shown to have properties similar to those of gallium arsenide JFETs. The short gate device has a negative resistance region at low drain voltages which is absent in the long gate device. The negative resistance region is determined by the distribution of the drift velocity along the conducting channel, and the resulting distribution of the electron charge. A field dependent mobility is used, and its effect on current saturation is discussed. The electrostatic charge distribution and electric field distribution are calculated for both devices, and are shown in graphic form.     

A minimum profile uniform current density electrode

Present methods of determining the safe injected charge levels for disk-type electrodes are given in terms of an average charge density, although the charge density is higher near the periphery of the electrode. This paper describes an electrode that produces an injected charge density that is uniform over the surface of the electrode and thus permits maximum utilization of the surface. Charge density is the time integral of current density, and the alteration of the current density is obtained by adding curvature to the electrode and recessing it within a cylindrical insulating well. A novel numerical method is used to determine the recession and curvature, and this numerical method is also presented. The benefit of this technique is that it permits a reduction in the electrode size while maintaining the maximum safe injected charge level of a disk-type electrode. A minimum profile uniform current density electrode and the algorithms used in its design are presented in this paper. Finally, a flat electrode that is recessed by as little as 1/10 of its diameter is shown to have an injected current density on the electrode surface that is superior to that of a flat surface mounted electrode.     

A new approach based on a combination of integral equation and asymptotic techniques for solving electromagnetic scattering problems

We introduce a new approach for combining the integral equation and high frequency asymptotic techniques, e.g., the geometrical theory of diffraction. The method takes advantage of the fact that the Fourier transform of the unknown surface current distribution is proportional to the scattered far-field. A number of asymptotic methods are currently available that provide good approximation to this farfield in a convenient analytic form which is useful for deriving an initial estimate of the Fourier transform of the current distribution. An iterative scheme is developed for systematically improving the initial form of the high frequency asymptotic solution by manipulating the integral equation in the Fourier transform domain. A salient feature of the method is that it provides a convenient validity check of the solution for the surface current distribution by verifying that the scattered field it radiates indeed satisfies the boundary conditions at the surface of the scatterer. Another important feature of the method is that it yields both the induced surface current density and the far-field. Diffraction by a strip (two-dimensional problem) and diffraction by a thin plate (three-dimensional problem) are presented as illustrative examples that demonstrate the usefulness of the approach for handling a variety of electromagnetic scattering problems in the resonance region and above.     

Determination of lines of constant phase in the near-field of ametallic cube and an airplane

A method is presented which enables one to calculate the scattered field very close to the surface of a perfectly conducting body as well as at the surface itself. The method is based on the representation of the scattered field by an integral over the surface current distribution. The integrand is treated by identity transformations that the singular terms can be integrated analytically, while the remaining nonsingular terms are integrated numerically. The surface current distribution is determined by the magnetic field integral equation. The theory is validated by experiments with the scattered field of a metallic cube with an edge length of a wavelength. The current distribution and the normal as well as the tangential electric field at the surface of the cube are measured by small probes, and the results are compared to those of the theory. The theoretical results of the current distributions are presented as gray value graphics-those of the near-field distribution of a cube and an airplane with the help of lines of constant phase     

Folded-cavity surface-emitting InGaAs-GaAs lasers withlow-threshold current density and high efficiency

Folded-cavity surface-emitting InGaAs-GaAs lasers (FCSELs) that employ high-quality internal 45° deflectors are demonstrated with low-threshold current density and high efficiency. A simplified process involving a stop etch to position the surface emitting output mirror close to the waveguide and ion-beam-etching (IBE) to form the 45° deflecting mirror is presented. FCSELs (cavity length 800 μm) with two 45° deflectors, are obtained with threshold current density as low as 112.5 A/cm² and surface-emission external quantum efficiency as high as 65% (0.82 W/A). The additional loss contributed by the folded-cavity design is estimated as 4.2 cm^-1     

Current/voltage characteristics, channel pinchoff and field dependence of carrier velocity in silicon insulated-gate field-effect transistors

It is shown that charge carriers moving in the surface channel reach their scattering-limited drift velocity as the current reaches its saturation value. A longitudinal velocity/field relationship is proposed for the surface channel, based on that which exists in the bulk lattice. This leads to simple expressions for current and transconductance which agree satisfactorily with experimental data. Theoretical characteristics are presented to illustrate the effects of field dependence of carrier mobility on operating characteristics.     

Surface Current Source Reconstruction for Given Radiated Electromagnetic Fields

A general analytic approach is presented for reconstructing: 1) the minimum energy source enclosed by a sphere, and 2) the surface current distribution on a sphere from the knowledge of the radiated fields. The surface current source is derived by adding proper non-radiating sources to the minimum energy source. In contrast to the minimum energy volumetric distribution, the surface current derived in this paper is practically realizable. Finally, we present a closed form formula for the reconstructed spherical surface current source. We will show that this spherical surface current is indeed the unique solution of the inverse source problem for square-integrable surface electric current on a sphere in a homogenous medium.      

Computation of excess capacitances of various strip discontinuitiesusing closed-form Green's functions

An efficient quasi-static method to compute excess (equivalent) capacitances of various strip discontinuities in a multilayered dielectric medium is presented. The excess charge distribution on the surface of a conductor is obtained by solving an integral equation in conjunction with closed-form Green's functions. A complete list of expressions of the closed-form Green's functions for a point charge, a line charge, and a semi-infinite line charge is presented. An open end, a bend, a step junction, and a T junction are considered as numerical examples     

Modeling of amorphous-silicon thin-film transistors for circuitsimulations with SPICE

A static and dynamic model for amorphous silicon thin-film transistors is presented. The theory is based on an assumed exponential distribution of the deep states and the tail states in the energy gap. Expressions are derived that link the density of the localized states and the temperature to the drain current and the distribution of the charge in the transistor channel. In addition the authors take into account parasitic effects such as channel length modulation, off-resistance, drain and source resistances, mobile and free charges in the insulator, surface states, and overlap capacitances. The model is incorporated into the circuit simulation program SPICE. Charge conservation problems are overcome by using a charge-oriented dynamic transistor model. Simulated and measured current-voltage characteristics agree well. A 96-b gate line driver for addressing liquid-crystal displays, which was successfully designed and optimized with the model, is introduced     

Carrier heating or cooling in semiconductor devices

A theory of charge and energy transport by hot or cold carriers in inhomogeneous semiconductor structures, developed earlier, is presented here in extended form under the assumptions of a simple parabolic band model, elastic or velocity-randomizing scattering processes and one-dimensional devices with steady-state flow of charge and energy. The treatment begins with the derivation of a relation between the two first terms in the Legendre expansion of the distribution function and the definition of the relaxation times for the two types of scattering processes mentioned. Subsequently, the expressions for the current density and the energy flux are obtained for an arbitrary shape of the isotropic part of the distribution function, so that they are valid for hot or cold carriers. In these expressions appear four transport coefficients, namely the mobility, the diffusion coefficient, the thermal mobility and the thermal diffusion coefficient. Furthermore, under the assumption of a Maxwellian form of the isotropic part of the distribution function, the obtained expressions are simplified. Generalized Einstein relations between the transport coefficients are derived and the boundary-value problem is formulated which has to be solved in order to determine quantitatively the behavior of a device if the carriers become hot or cold. Finally, a discussion is given of a similar theory by Stratton as well as of limitations, advantages and possibilities of applications of the present theory. The present synthesis is not available elsewhere.     

Numerical optimization of 3-D SAR distributions in cylindricalmodels for electromagnetic hyperthermia

Numerically optimized SAR (specific absorption rate) distributions in a source free 3-D multilayered concentric cylindrical (MCC) model are presented. The fields were expanded in the modes of the MCC. Cost functions which specify mathematically the relative weight assigned to differences between an SAR distribution and a desired SAR distribution were defined. The coefficients of the modes, which minimize the cost function, were obtained using gradient search optimization methods. The optimized SAR distributions shown were computed using three different cost functions and two different radial locations for the center of the region where the desired SAR is largest. A five-layered model, including the outer water layer for cooling and improved matching with the source, was used. The frequency was 70 MHz. The current and charge distributions computed on a perfectly conducting cylindrical surface just outside the model are also shown. The surface current and charge distributions depend strongly on the relative importance of the cost for acute heat and systemic heat. A technique is developed for generating a new set of basis functions for reducing the number of unknowns to be optimized. We suggest that the approach shown could be useful in designing hyperthermia applicators.     

Two-dimensional analysis of the surface state effects in 4H-SiC MESFETs

Two-dimensional small-signal ac and transient analysis of surface trap effects in 4H-SiC MESFETs have been performed in this paper. The mechanism by which acceptor-type traps effect the transconductance and drain current changes has been discussed. The simulation results show that transconductance exhibits negative frequency dispersion behavior, which is caused by the charge exchange via the surface states existing between the gate-source and gate-drain terminals. The current degradation behavior is also observed due to acceptor-type traps, acting as electron traps, in MESFET devices. A detailed study involving the density, ionization and energy level of traps reveals conclusive results in the devices analyzed.     

Electromagnetic characteristics of superquadric wire loop antennas

The analysis of antenna configurations in the form of a generalized superquadric loop, which includes circular, elliptical and rectangular loop geometries, is presented in this paper. Use of a Galerkin form of the moment method with piecewise sinusoidal subsectional basis and testing functions provides rapid numerical convergence and accurate representation of the antenna current. A convenient parametric representation for the superquadric curve is developed to allow a subsectional formulation using curved wire segments, rather than the commonly employed piecewise linear segments, to construct the geometry. Both magnetic frill and delta gap source models are implemented to allow a detailed study of input impedance, directivity, radiation pattern and current distribution as a function of various geometrical parameters. The results are shown to compare well with previous results for the special case of a circular loop antenna. Some useful curves are presented to aid in the design of practical superquadric loop antennas     

Influence of interface charge inhomogeneities on the measurement of surface state densities in SiSiO2 interfaces by means of the MOS a.c. conductance technique

The relationship between interface charge and surface potential of a MOS capacitor is examined when interface charge inhomogeneities are present. For practical values of the interface charge variance, the relation between interface charge and surface potential is found to be quite linear. High surface state densities and high impurity concentrations tend to damp the potential fluctuations and to increase the linearity. The magnitude of the potential deviation for a given charge deviation increases from flat band to weak inversion and decreases again in strong inversion, due to screening, but the linearity is found to be best in weak inversion.The original Nicollian-Goetzberger analysis of the MOS a.c. conductance technique uses a Gaussian potential distribution and an equivalent circuit consisting of an array of parallel surface state branches connected to a single oxide capacitance. We compare this model with a patchwork model, using a Gaussian interface charge density distribution and an equivalent circuit with distributed oxide capacitance. It is found that in depletion, for practical charge densities, the patchwork model interpretation of conductance peaks does not lead to a very different result than the random charge distribution model interpretation. Both models agree very well on surface state density and variance of the interface charge distribution, but a large discrepancy on the capture cross section of the surface states is possible.     

Computer model and charge transport studies in short gate charge-coupled devices

A computer model has been developed that simulates charge transport of carriers in a surface channel charge-coupled device. This model is based on the charge continuity and current transport equations with a time dependent surface field. The device structure of the model includes a source diffusion an input gate and transfer gate. The present model is the first real simulation of the input scheme of the surface-channel CCDs. The scooping and spilling techniques associated with the charge injection process are simulated by the input diffusion which is included in the model.As an application to a CCD practical problem the present model has been used to study the linearity of the electrical charge injection into surface channel charge-coupled devices. The generated harmonic components of a sinusoidal input are calculated using the transfer characteristics of the input stage obtained from the computer simulation.Using this model the spatial variations of the self-induced fringing field and total currents under the storage and transfer gates were computed. The charge transfer mechanisms for short-gate (L ≤ 8 μm) CCDs was investigated. It was found that for short gates the charge transfer efficiency is governed mainly by the fringing field and self-induced current mechanisms. The results of this study help to clarify the mechanism by which the signal-charge level and gate length affect the charge transfer efficiency.     

Surface Barrier Models of ZnO

For a low surface barrier, the energy band, barrier height and width of the space charge region at the surface of relatively large grains of ZnO are presented analytically on condition that the electron distribution obeys the Boltzmann statistics. It is shown that the temperature in the space charge distribution factor has an important effect on the energy band, barrier height and width of the space charge region. The depletion approximation is a model in which the temperature in the space charge distribution factor is zero. Our results are better than the depletion approximation.     

Three-dimensional derivation of the electrodynamic jump conditions and momentum-energy laws at a moving boundary

The purpose of this paper is to set forth a comprehensive three-dimensional derivation of the electromagnetic jump conditions at a moving boundary. Also presented are corresponding derivations for the electromagnetic surface traction and power transfer formulas. The derivations are based on a variation of an old technique which dates back to Lorentz' introduction of kinematic axioms for electrodynamics. The surface of electromagnetic discontinuity is allowed to move and deform in an arbitrary manner. In addition to the Helmholtz vector-flux theorem which Lorentz employed, two kinematic theorems are utilized in conjunction with Maxwell's equations. Surface charge and surface current are retained in the derivation. These have a marked effect on the current boundary condition and on the surface traction and power transfer formulas. A clear distinction is made between the material velocity and the abstract velocity symbol appearing in the Lorentz integral axioms. New modifications based on this distinction permit derivation of the boundary conditions in a general form applicable to any moving surface of electromagnetic discontinuity, irrespective of the ambient motion of the material medium.     

Scattering by periodic metal surfaces with sinusoidal height profiles--A theoretical approach

A theory of scattering by periodic metal surfaces is presented that utilizes the physical optics approximation to determine the current distribution in the metal surface to first order, but modifies this approximate distribution by multiplication with a Fourier series whose fundamental period is that of the surface profile (Floquet's theorem). The coefficients of the Fourier series are determined from the condition that the field radiated by the current distribution into the lower (shielded) half-space must cancel the primary plane wave in this space range. The theory reduces the scatter problem to the familiar task of solving a linear system. For certain basic types of surface profiles, including the sinusoidal profile considered here, the coefficients of the linear system are obtained as closed form expressions in well-known functions (Bessel functions for sinusoidal profiles and exponential functions for piecewise linear profiles). The theory is thus amenable to efficient computer evaluation. Comparison of numerical results based on this theory with data obtained by recent numerical schemes shows that for depths of surface grooves less than a wavelength and for unrestricted groove widths, reliable and comparable, if not more accurate, data is obtained, in many cases at considerably cheaper computational cost.     

Theoretical and practical investigation of the thermal generation in gate controlled diodes

The different components of thermal generation in a gate controlled diode are studied theoretically and experimentally. Expressions for the generation current in the space charge layer, the diffusion current from the quasi-neutral bulk and the surface generation current are derived for a gated-diode. The width of the generation zone within the space charge layer is calculated as a function of the energy level of the trap and the diode reverse voltage. This leads to a characteristic of the leakage current as a function of the space charge layer width. It is pointed out that the diffusion current can influence the leakage current and cannot be neglected in structures with a low dark current. In the second part the gate controlled diode is used to characterize the thermal generation in structures with a homogeneous and low dark current. A generation lifetime of 5.5 msec and a surface generation velocity at a depleted surface of 1.5 cm/sec is derived. The generation lifetime is found to be constant as a function of depth into the substrate. A considerable diffusion current is measured which is comparable to the generation current in the space charge layer.