Time-Domain Simulation of Power Combining in a Chain of THz Gunn Diodes in a Transmission Line

Solid-state THz devices usually suffer from low power output. This, however, can be overcome by efficient power combining. In this paper, the power combining of a chain of active devices (longitudinal array of N Gunn diodes) in a semi-infinite transmission line has been simulated. Superlinear growth of power P is found in a line with inductive-capacitive elements: P is quadratic in N when P is small and becomes linear when P is large. Intricate auto-modulation due to coupling between the devices is shown to accompany power combining.     

Theory of dynamic charge and capacitance characteristics in MIS systems containing discrete surface traps

Theoretical studies have been made on the dynamic characteristics of the metal-insulator-semiconductor (MIS) capacitor containing discrete surface traps. It has been shown that when the surface traps are in dynamic equilibrium with the voltage ramp, the device exhibits steady-state charge and capacitance characteristics. When the surface traps are out of equilibrium with the voltage ramp, then the emission of trapped charge is a function of time only and not of voltage. Under such conditions, the characteristics are considered to be non-steady-state in nature. The emission of trapped charge in the non-steady state gives rise to a plateau or ledge in the C–V characteristics, while characteristics. The transition from the steady state to the non-steady state results in kinks in the characteristics.The physical processes involved have been stressed and simple expressions have been obtained for the charge and capacitance in terms of the trapping parameters, sweep-rate and temperature.     

A generalization of Freund's model for a repairable paired component based on a bivariate Geiger Muller (G.M.) counter

It has been envisaged in this paper to obtain the probability of the first system failure in a bicomponent parallel system. Following Freund (J. Am. Statist. Ass.56, 971–977, 1961), it has been assumed that the hazard rate of the surviving component suddenly jumps up on the failure of its counterpart. The system failure occurs when the surviving component with increased hazard rate fails while its counterpart is under repair for a fixed period π. Under this set up, the probability of the first system failure on the rth failure (r = 1,2,3…) of either of the two component has been worked out. Specific examples with Poisson inputs have been shown. It has been noticed as the process is a non-Markovian one, estimation of the parameters λ and λ′ (being the original hazard rate and the aggravated hazard rate), with the method of Moments having not been easily estimable; the process of embedded Markov chain has been utilized to estimate the parameters of the non-Markov process.     

Analysis of a non-uniformly doped MPN silicon Schottky barrier solar cell

An analysis of the solar cell properties of an AlPN Si Schottky diode has been made assuming a generalised impurity distribution in the P-type doped layer. The effect of the image force potential on the effective barrier height is considered. The effect of a constant built-in feld in the substrate has also been included in the analysis. Important solar cell parameters such as open circuit voltage (V_oc), short-circuit current (I_sc) and fill factor (FF) have been calculated as a function of the impurity concentration in the doped layer and its thickness. It is shown that the efficiency η depends strongly on the nature of the impurity profile, being highest for the Gaussian and lowest for the exponential. Finally, the presence of a built-in field (Eb = 4.5 V/cm) in the substrate is found to increase the resultant η which is about 6–8% higher than that of a diode without the drift field.     

A self-consistent extraction procedure for source/drain resistance in MOSFETs

A new method to determine source/drain series resistance has been developed for MOSFETs operated in linear region. The source/drain resistance (R_SD) is gate-bias dependent. Channel length reduction (ΔL) is extracted at low gate bias and chosen to be constant. All parameters extracted in this method are assumed to be independent of mask channel length for model simplicity. The method has been applied to devices with mask channel lengths of 0.23, 0.2, and 0.185 μm. The extracted parameters are consistent with the assumptions and have been validated by measured I-V characteristics.     

Theory of dynamic charge current and capacitance characteristics in MIS systems containing distributed surface traps

Theoretical studies have been made on the dynamic characteristics of the metal-insulator-semiconductor (MIS) capacitor containing distributed surface traps. It has been shown that when the surface traps are in dynamic equilibrium with the voltage ramp, the device exhibits steady-state charge, current and capacitance characteristics. When the surface traps are out of equilibrium with the voltage ramp, then the emission of trapped charge is a function of time only and not of voltage. Under such conditions, the characteristics are considered to be non-steady-state in nature.In the steady state, the emission of electrons from a continuum of surface traps accounts for the reduction in the slopes of the C–V curves from the ideal ones. Kinks are manifested when the traps just empty the last of their electrons.In the non-steady state, electron emission can be described by the non-steady-state (time-dependent) occupancy function derived herein, which is shown to be similar in shape to the Fermi function. This means that electron emission takes place from a narrow band of energy positioned near the uppermost-filled traps. Hysteresis effects are manifested in the C–V characteristics due to the non-steady-state emission of trapped charge. At the transition from the steady state to the non-steady state and vice versa, kinks are exhibited in the charge and capacitance characteristics, while step changes in current components are also predicted.The physical processes involved have been stressed and closed-form expressions have been obtained for the charge, current and capacitance in terms of the trapping parameters, sweep-rate and temperature.     

Investigation of power Trench MOSFETs with retrograde body profile

We present here a power Trench MOSFET (T-MOS) with retrograde body doping profile. The channel length and trench depth are both shortened compared with conventional T-MOS. High energy implantation is used to form retrograde body profile. Electronic parameters of the new structure have been obtained by process and device simulation. The results show that the new structure has much lower specific on-resistance (R_ds,on) because of its shorter channel when compared with conventional T-MOS. As the trench depth is shallowed, the gate charge density Q_g is also reduced.     

Electrical properties of CdxHg1−x Te/CdZnTe heterostructures

Specific features of the electrical properties of p-type Cd_xHg_1?x Te/CdZnTe heterostructures have been studied at liquid-nitrogen temperatures. The adopted theoretical model satisfactorily describes the obtained experimental data. Errors in determining the concentration and mobility of heavy holes—arising when contributions to the electrical parameters from electrons and light holes are disregarded—are evaluated.     

Analysis of the interaction of an electron beam with a solar cell—III: The effect of spacial variations of the number density of recombination centers on SEM measurements

By means of an exactly soluble model the short circuit current I_SC generated by a scanning electron microscope in a P-N junction has been determined in cases where the trap density N_t is inhomogeneous. The diffusion length for minority carriers becomes then dependent on the spacial coordinates. It is shown that in this case the dependence of the I_SC on characteristic parameters as cell thickness, distance of the beam excitation spot from ohmic contacts etc., becomes very intricate. This fact precludes the determination of the local diffusion length in the usual manner. Although the model is somewhat simplified in order to make it amenable to exact solutions, it is nevertheless realistic enough to lead to the conclusion that SEM measurements of bulk transport parameters in inhomogeneous semiconductor material are impractical since they may lead to serious errors in the interpretation of the data by customary means.     

A new method for computer-aided optimization of solar cell structures

A new method for truly optimizing the design of intrinsic solar-cell structures has been established. To support the proposed method, a device modeling technique specialized for n⁺p solar cells with uniform doping has been developed, and the constrained optimization technique extended from Rosenbrock's algorithm has been implemented. All parameters to be optimized are simultaneously adjusted in a systematic manner, resulting in a truly optimum design. Several design examples for n⁺?p solar cells, such as maximizing cell efficiency and minimizing cell thickness while maintaining the acceptable characteristics, have been used to demonstrate the validity of the proposed method. Moreover, this method has also been successfully applied to the optimum design of other solar cell structures.     

Microwave spectra of CD3OH methanol

The microwave spectrum of CD₃OH has been studied over the 8 to 58 GHz region, and numerousb-type transitions have been assigned. Many of these belong toP subbranches which descend to the microwave region from subband origins lying much higher in the far-infrared, pass through zero frequency, and return upward again. Others are members ofb-typeQ branches which extend across the region. As well, variousa-typeK-doublet lines arising as transitions directly across the split levels of asymmetry doublets have been identified.     

High-efficiency LEDs based on <Emphasis Type="Italic">n</Emphasis>-GaSb/<Emphasis Type="Italic">p</Emphasis>-GaSb/<Emphasis Type="Italic">n</Emphasis>-GaInAsSb/<Emphasis Type="Italic">P</Emphasis>-AlGaAsSb type-II thyristor heterostructures

A new type of light-emitting diodes (LEDs), a high-efficiency device based on an n-GaSb/p-GaSb/n-GaInAsSb/P-AlGaAsSb thyristor heterostructure, with the maximum emission intensity at wavelength λ = 1.95 μm, has been suggested and its electrical and luminescent characteristics have been studied. It is shown that the effective radiative recombination in the thyristor structure in the n-type GaInAsSb active region is provided by double-sided injection of holes from the neighboring p-type regions. The maximum internal quantum efficiency of 77% was achieved in the structure under study in the pulsed mode. The average optical power was as high as 2.5 mW, and the peak power in the pulsed mode was 71 mW, which exceeded by a factor of 2.9 the power obtained with a standard n-GaSb/n-GaInAsSb/P-AlGaAsSb LED operating in the same spectral range. The approach suggested will make it possible to improve LED parameters in the entire mid-IR spectral range (2–5 μm).     

New Quaternary Sb-Se-Ge-In Chalcogenide Glasses: Linear and Nonlinear Optical Properties

Chalcogenide glasses find extensive applications in infrared (IR) devices and optical communication. Optical parameters of Sb₁₀Se₆₅Ge_25?y In _y thin films, deposited by the thermal evaporation technique, have been analyzed using ultraviolet–visible-near IR spectroscopy. The transitions in the forbidden gap are indirect. The effect of indium (In) alloying on the nonlinear optical parameters has been studied. A shift in optical absorption edge towards higher wavelength shows that the width of the localized states changes, which affects the optical parameters of the system. The high nonlinearity of these glasses makes them suitable for optical regeneration and Raman amplification.     

Phosphorus diffusion in gallium arsenide

The formation of monocrystalline GaAs_1?xP_x by the solid state diffusion of phosphorus into monocrystalline GaAs at a phosphorus pressure of 35 atm has been confirmed by X-rays and reflectivity measurements. Phosphorus distribution in GaAs specimens diffused in the temperature range 800–1100°C has been determined by reflectance studies. The diffusion coefficient and the activation energy computed from the known phosphorus distributions in the diffused specimens are found to depend on the content of phosphorus in the diffused region. The variation of the depth of GaAs_1?xP_x-GaAs junction with time (at a fixed temperature, 1000°C) and temperature (for a fixed time, 24.5 hr) has been studied.     

Interface States of Fe3O4/Si Interfacial Structure and Effect of Magnetic Field

Electronic transport across Fe₃O₄/Si interfacial structure has been studied with and without the application of magnetic fields along the interfacial plane, up to 8 kG. Current–voltage (I–V) and capacitance–voltage (C–V) characteristics across the junction have been recorded for various bias voltages, frequency and magnetic field. The interfacial parameters, such as, ideality factor (n), barrier height (? _B0), series resistance (R _S) and donor concentration (N _D) etc. have been estimated from the characteristics. The interface state density (N _SS) and their energy distribution have been estimated by using the interfacial parameters. It has been observed that the N _SS decreases as the energy increases from the conduction band edge towards the valence band. A magnetoresistance (MR) of ～40% has been estimated from the I–V–H data along with its variation with magnetic field. The change of interface state density with the magnetic field shows a similar variation as MR versus H. From the observed variations, the interface states seem to be related to electronic spins. The possibility of an interfacial magnetic silicide or magnetic ions in the interfacial region has been invoked for the observed interface states.     

A dual-polarization InSb receiver for 461/492 GHz

A dual-polarization InSb hot-electron bolometer-mixer receiver has been built for the James Clerk Maxwell Telescope, for operation at 461 and 492 GHz (the frequencies of theJ=4→3 rotational transition of CO and of the³ P ₁→³ P ₀ transition of neutral carbon). Receiver noise temperatures of 500K have been obtained at 461 GHz, in observing bandwidths of 3 MHz. The receiver was designed as a “common-user” or facility instrument. Here we describe those aspects of the design and construction which enabled this goal to be realized.     

Analytical model of body factor in short channel bulk MOSFETs for low voltage applications

An analytical model to study the degradation of body factor (γ) and subthreshold factor (S) of short channel bulk MOSFETs have been developed and the expression for body factor of short channel devices is derived. Modified relation between S and γ factors has also been obtained and a correction factor γ_s is found which maintained the conventional relation in spite of the degradation of these two parameters in short channel devices. Results obtained are found in good approximation with 2D simulated data.     

Study of a buried sin+P homojunction solar cell

An analysis of the solar cell properties of an ITO-I-N⁺-P (InP) Schottky diode has been made. Important solar cell parameters such as open circuit voltage (V_oc), short-circuit current (J_sc), fill factor (FF), efficiency (η) have been calculated as a function of the impurity concentration in the doped layer and its thickness. It has been shown that efficiency η depends strongly on the properties of the doped layer. Finally, theoretical results have been compared with the experimental results available in the literature.     

The effect of deposition conditions on structure properties of radio frequency reactive sputtered polycrystalline ZnO films

Polycrystalline ZnO films were prepared by radio frequency reactive sputtering technique with Zn target on glass wafers under different pressure ratio K (defined as P_O2/(P_Ar+P_O2)) and substrate temperature T_s. X-ray diffraction technique (XRD) was employed to analyze the effect of deposition conditions on the structure of the films. With increase in T_s from room temperature to 400°C, the films appear to have a preferred orientation with the c-axis perpendicular to the substrate. Pressure ratio K has an effect on orientation of the films, which have highly preferred orientation when K=40%. The grain size is in the range 13.2–42.8 nm calculated from XRD data. The packing density, the compressive stress in (0 0 2) plane and the average optical transmittance in visible region of the films deposited under the optimal condition are 97%, −1.06×10⁹ N/m² and 92%, respectively. It is found also that carbon is the main contamination in films observed by X-ray photoelectron spectroscopy. The films are close to stoichiometric relation.     

A theoretical analysis together with experimental data of inhomogeneous Schottky barrier diodes

We have shown that the current at lower temperatures may exceed the current at higher temperatures using the approximation of the apparent BH in the inhomogeneous SDs, as determined by some authors in the literature. Then, in the simulated forward I-V curves of non-interactive inhomogeneous SDs using the approximation of the apparent BH, it has been showed that the current at lower temperatures does not exceed the current at higher temperatures when considering the bias voltage dependence of apparent BH. In the calculations, the parameters from the experimental forward bias I-V characteristics, in the temperature range of 60-300 K, of the Ni/n-GaAs Schottky contacts with the inhomogeneous BH have been used by means of the thermionic emission theory of inhomogeneous Schottky contacts, considering a Gaussian distribution of barrier heights with a linear bias dependence.