scispace - formally typeset
Search or ask a question

Showing papers in "Solid-state Electronics in 1988"


Journal ArticleDOI
TL;DR: In this paper, a deconvolution operation performed in the logarithmic time domain gives the "timeconstant spectrum" of the chip-case-ambient thermal structure.
Abstract: A new method has been developed in order to identify the thermal environment of a semiconductor device chip. The identification algorithm operates on the thermal transient response of the device recorded during a one-shot pulse measurement. A deconvolution operation performed in the logarithmic time domain gives the “time-constant spectrum” of the chip-case-ambient thermal structure. A further transformation leads to the “structure-function” that is the cross-sectional area of the heat conducting materials vs thermal resistance (related to the heat source). The structure function has a good and quantitatively evaluable correspondence to the physical chip environment and heat conducting structure. Separating the different regions of the heat-flow path (corresponding to the chip, bond, header, case) as well as the detection of eventual heat-transport irregularities (mounting errors) is possible.

419 citations


Journal ArticleDOI
TL;DR: In this article, the capacitance-time transients are digitalized, and the discrete Fourier coefficients are formed via numerical Fourier transformation, which can be used to calculate amplitude and time constant of the transients for discrete trap levels in various ways, thus giving control of the results.
Abstract: The advantages of a DLTS method—the DLTFS method—are presented. In this technique the capacitance-time transients are digitalized, and the discrete Fourier coefficients are formed via numerical Fourier transformation. These coefficients can be used to calculate amplitude and time constant of the transients for discrete trap levels in various ways, thus giving control of the results. This permits automatic control of measuring parameters (e.g. choice of period width and temperature values), automatic evaluation during measurement (e.g. impurity concentration, activation energy and capture cross section), and high measuring accuracy. In contrast to the usual DLTS technique a temperature dependence of the amplitude, leading to errors especially in CC-DLTS and current-DLTS measurements, does not present a source of error. The good noise suppression is a further advantage of the DLTFS method.

324 citations


Journal ArticleDOI
TL;DR: An analytical model for the power Insulated-Gate Bipolar Transistor (IGBT) is developed in this paper, which consistently describes the IGBT steady-state currentvoltage characteristics and switching transient current and voltage waveforms for all loading conditions.
Abstract: An analytical model for the power Insulated-Gate Bipolar Transistor (IGBT) is developed. The model consistently describes the IGBT steady-state current-voltage characteristics and switching transient current and voltage waveforms for all loading conditions. The model is based on the equivalent circuit of a MOSFET which supplies the base current to a low-gain, high-level injection, bipolar transistor with its base virtual contact at the collector end of the base. The basic element of the model is a detailed analysis of the bipolar transistor which uses ambipolar transport theory and does not assume the quasi-static condition for the transient analysis. This analysis differs from the previous bipolar transistor theory in that (1) the relatively large base current which flows from the collector end of the base is properly accounted for, and (2) the component of current due to the changing carrier distribution under the condition of a moving collector-base depletion edge during anode voltage transitions is accounted for. Experimental verification of the model using devices with different base lifetimes is presented for the on-state current-voltage characteristics, the steady-state saturation current, and the current and voltage waveforms for the constant voltage transient, the inductive load transient, and the series resistor-inductor load transient.

231 citations


Journal ArticleDOI
TL;DR: In this article, temperature dependence of noise power spectra in a series of commercial p-channel MOSFETs is studied. And the experimental results were compared to first-principle calculations which showed that the noise was due to the capture and emission of carriers by oxide traps through thermal activation.
Abstract: Detailed measurements of the temperature dependence of noise power spectra in a series of commercial p-channel MOSFETs are presented It was found that both the magnitude and the functional form of the voltage noise power spectral density varied greatly in the range between 60 and 260 K The experimental results were compared to first-principle calculations which showed that the noise was due to the capture and emission of carriers by oxide traps through thermal activation The process caused fluctuations in both the density and the surface mobility of the channel carriers through the modulation of the surface potential and the scattering rate respectively

78 citations


Journal ArticleDOI
TL;DR: In this article, the authors used low frequency noise vs temperature measurements for determining generation recombination (G.R.) trapping parameters in MOSFETs in a manner similar to that of Deep Level Transient Spectroscopy (DLTS).
Abstract: The technique of low frequency noise vs temperature measurements is shown to be a powerful diagnostic technique for determining Generation Recombination (G.R.) trapping parameters in MOSFETS. From computer controlled measurements of low frequency noise vs temperature, the trapping parameters are extracted in a manner similar to that of Deep Level Transient Spectroscopy (DLTS). The trapping parameters are also extracted by curve fitting of the low frequency noise vs temperature curves. These noise-determined trapping parameters are compared with those measured by DLTS. The agreement between parameters determined by the spot frequency noise Arrhenius plot with those determined by DLTS is close, while the agreement between those determined by noise curve fitting is reasonable for noise peaks near room temperature, but becomes poor for low temperature noise peaks. We have found that the low frequency noise vs temperature technique appears to show higher sensitivity than our DLTS measurements. By the use of PECVD silicon nitride as a passivation material, the sensitivity to measure G.R. traps by the low frequency noise vs temperature technique can be increased. This improved sensitivity is due to the reduction of interface state noise. The PECVD silicon nitride passivation had no effect on reducing the G.R. noise peaks thus indicating that these traps are probably located in the bulk space-charge region.

77 citations


Journal ArticleDOI
TL;DR: In this article, the drift velocity on electric field for electrons in n-type silicon inversion layers is determined from the d.c. drain-conductance measurement of polysilicon resistive-gate field effect transistors.
Abstract: The dependence of the drift velocity on electric field for electrons in n -type silicon inversion layers is determined from the d.c. drain-conductance measurement of polysilicon resistive-gate field-effect-transistors. Up to fields 1.5 × 10 4 V/cm this dependence is well approximated by an empirical relationship which satisfies Thornber's theory. The extrapolated saturation drift velocity, nearly independent of normal (or gate) electric fields up to 8 × 10 5 V/cm, is 6.0 × 10 6 cm/s. This value is used in a field-dependent model for electron mobility in silicon inversion layers implemented in a 2-D device simulator. Accurate numerical simulations of the resistive-gate device correctly predict the experimentally observed drain current-drain voltage characteristics.

72 citations


Journal ArticleDOI
TL;DR: In this paper, a new and accurate approach to conductance measurements on MOSFETs is presented, which can be used to study interface trap properties in most of the silicon band-gap by direct measurement on a single MOS-FET.
Abstract: A new and accurate approach to a.c. conductance measurements on MOSFETs is presented. It is shown that the conductance technique can be used to study interface trap properties in most of the silicon band-gap by direct measurement on a single MOSFET. The equivalent circuit is analyzed and the influence of the channel length on the inversion layer response is discussed in detail. It is shown that the channel time constant is mainly determined by the channel length. For small channel lengths L

64 citations


Journal ArticleDOI
Yoshitomo Sasaki1, K. Itoh1, E. Inoue1, S. Kishi1, T. Mitsuishi1 
TL;DR: The relationship between the Hall mobility and the hole concentration in p-type silicon was experimentally investigated in this article, where Boron and gallium were used as dopants; their doping was done by either the ion implantation or the diffusion technique.
Abstract: The relationship between the Hall mobility and the hole concentration in p-type silicon were experimentally investigated. Boron and gallium were used as dopants; their doping was done by either the ion implantation or the diffusion technique. It was found that the relationships for boron-doped specimens and for gallium-doped specimens for the heavily doped range are considerably different from each other. Both of these relationships are also different from the so-called Irvin curve which is well known as a standard relationship between the mobility of carriers and the impurity concentration, but the relationship for boron doping is almost consistent with that of Thurber et al., if the Hall mobility factor is reasonably taken into account.

59 citations


Journal ArticleDOI
TL;DR: In this article, the effect of the magnetic vector potential on the tunnelling electrons was investigated in a double barrier resonant tunnelling device with n-InP/(InGa)As.
Abstract: Negative differential conductivity (NDC) with a peak/valley ratio of 4.5:1 (4 K) and 2:1 (150 K) is observed in double barrier resonant tunnelling devices based on n-InP/(InGa)As. A transverse magnetic field applied in the plane of the tunnelling barriers ( J ¦ B ) significantly changes the current-voltage characteristics and eliminates the NDC for fields above −10 T. This behaviour is explained qualitatively in terms of the effect of the magnetic vector potential on the tunnelling electrons. The magneto-oscillations in the tunnelling current for J ‖ B are discussed in terms of a simple model of resonant tunnelling.

54 citations


Journal ArticleDOI
J.A. Kash1, J.C. Tsang1
TL;DR: In this paper, the initial relaxation of optically injected hot carriers in polar semiconductors is studied by Raman scattering from nonequilibrium LO phonons and anti-Stokes hot luminescence.
Abstract: The details of the initial relaxation of optically injected hot carriers in polar semiconductors are studied by Raman scattering from nonequilibrium LO phonons and anti-Stokes hot luminescence. Experiments on intrinsic and doped GaAs reveal carrier-phonon and carrier-carrier interaction times, the wavevector dependence of the hot phonon distribution, and the influence of holes on the LO phonon lifetime. Studies in AlxGa1−xAs probe the influence of alloy disorder on the generation of hot phonons. In addition, because there are two optic phonon modes in AlxGa1−xAs, we can experimentally measure the effect of ionicity (the frequency difference between LO and TO phonons) on the generation rate for the hot phonons.

51 citations


Journal ArticleDOI
TL;DR: In this article, a form of quantum transport theory has been developed to model the resonant-tunneling diode and similar devices in which quantum interference effects play a significant role, and the internal state of the devices is represented by the Wigner distribution function, with boundary conditions which model the effects of the electrical contacts to the device.
Abstract: A form of quantum transport theory has been developed to model the resonant-tunneling diode and similar devices in which quantum interference effects play a significant role. The internal state of the devices is represented by the Wigner distribution function, with boundary conditions which model the effects of the electrical contacts to the device. Inelastic scattering processes are approximated by a classical Boltzmann collision operator, and the effects of different scattering processes on the device characteristics are evaluated numerically.

Journal ArticleDOI
TL;DR: In this paper, the uniform plasma injection regime in reversely switched dinistors (RSD) was investigated and criteria of the quasi-diode operation were established in both the micro-and sub-microsecond range of operation.
Abstract: This paper deals with the uniform plasma injection regime in reversely switched dinistors (RSD) which is similar to the drift double-injection regime in p + nn + -diodes. Such a regime is distinguished as highly efficient and most stable. The anode current retroactive mechanism controlled by the plasma in the collector layer is investigated and criteria of the quasi-diode operation are established. The theory explains the main characteristics of the RSD in both the micro- and submicrosecond range of operation. It also allows the calculation of time dependence of current density and voltage drop. The criteria derived are in good agreement with empirical data.

Journal ArticleDOI
TL;DR: In this article, the effects of hot-carrier stress on MOSFET degradation were analyzed by separating the effect of generated fixedoxide charge ΔQF due to trapped electrons in the gate oxide and charge in the generated interface states ΔQi.
Abstract: An analysis of MOSFET degradation induced by hot-electron stress is carried out by separating the effects of generated fixed-oxide charge ΔQF, due to trapped electrons in the gate oxide and charge in the generated interface states ΔQi. Both influences are considered in the subthreshold and above-threshold bias regions using a model for the stressed device which divides its channel into two regions having different threshold voltages. A phyiscal explanation is offered for degradation of transconductance gm by considering the VG dependence of the charge in the interface states ΔQi and the threshold voltage VT. It is predicted that an increase in QF alone causes the magnitude of the threshold voltage VT to increase and the device turn-on to become more abrupt (transconductance gm increases), while an increase in Qi alone results in a softer turn-on (gm decreases) and makes the threshold voltage dependent on VG. The generated interface state density Dit and the fixed charge density ΔQF is extracted by fitting the measurements to the above threshold ID-VG characteristics (small drain voltages) calculated with the model. Furthermore, we find that while both QF and Qi increase monotonically with stress, the contribution of ΔQi to the change in VT is significantly larger than that due to the change in QF. This conclusion is independent of stress time or of the exact division of the channel (the assumed lengths L1 and L2). Interface state densities obtained from the measured slopes of subthreshold ID-VG plots (which are affected by states near the middle of the band gap) are appreciably lower than densities deduced from measurements made above threshold (which are sensitive to states near the band edges) using the divided-channel representation. An increased dependence of subthreshold current on VD after stress is attributed to short-channel behavior of the MOSFET in the vicinity of the drain. The effects of hot-carrier stress in p-channel MOSFETS are also interpreted in terms of the two-transistor representation.

Journal ArticleDOI
TL;DR: In this paper, the authors measured the energy loss rates of hot carriers in bulk Ga.47In.53As and in Ga.53In.5As/InP quantum wells of widths 154 A and 14 A.
Abstract: We have measured the energy loss rates of hot carriers in bulk Ga.47In.53As and in Ga.47In.53As/InP quantum wells of widths 154 A and 14 A. We find that the energy loss rates are considerably lower than predicted by the unscreened carrier-LO phonon interaction. We show that the discrepancy may be resolved, at least in part, by invoking the presence of a non-equilibrium phonon distribution, the magnitude of the effect depending on the well width.

Journal ArticleDOI
TL;DR: In this article, a new method for determining the saturation voltage of a small-geometry MOSFET directly from the measured data is proposed and investigated, in which a special function G is formed and the drain-source saturation voltage is identified as the voltage of the peak point in a plot of G vs the drain source voltage.
Abstract: A new extraction method which determines the saturation voltage of a small-geometry MOSFET directly from the measured data is proposed and investigated. In this method, a special function G is formed and the drain-source saturation voltage is identified as the voltage of the peak point in a plot of G vs the drain-source voltage. Since the method is based on a general device theory, it is virtually independent of any device model and quite versatile and applicable for all MOSFETs. In addition, no given device parameters or iterations are required in the method. To verify the new method, SPICE MOS models are used as a calculation example. Moreover, the method is also applied to various fabricated MOSFETs to determine the saturation voltage. It is found that the saturation voltage can be definitely determined without ambiguity and the determined saturation voltage is quite close to that from the optimal extractions. Thus the method can be incorporated into the parameter extraction and the device modeling for small-geometry MOSFETs.

Journal ArticleDOI
TL;DR: In this article, a non-Maxwellian hot-electron distribution function derived from the Boltzmann transport equation is used to predict gate leakage current density in MOSFETs.
Abstract: A new equation, which does not require any fitting parameters, has been developed to predict gate leakage current density in MOSFETs. The equation is based upon a non-Maxwellian hot-electron distribution function derived from the Boltzmann transport equation, and utilizes a physically calculated local electron temperature. The model predicts gate currents for a sample submicron MOSFET that are in good agreement with experiment.

Journal ArticleDOI
TL;DR: The InGaAs-based hot electron transistors (RHET) as discussed by the authors achieved a peak-to-valley ratio of 19.3, with a maximum of 21.7.
Abstract: This paper reviews our current activities in hot electron transistors, and then describes recent advances in the RHET technology using InGaAs-based materials. The RHET's emitter common current gain is typically 10 to 17, with a maximum of 25, which is about four times greater than that of a GaAs-based RHET. The collector current peak-to-valley ratio reaches 19.3, with a maximum of 21.7, eight times that of the GaAs-based RHET. These are followed by theoretical and experimental analyses of the RHET's DC performance. It is found that theoretical and experimental results do not agree for the GaAs-based RHET but agree well for the InGaAs-based RHET.

Journal ArticleDOI
TL;DR: In this paper, the authors reviewed the transfer of electrons between semiconductor heterolayers and its importance and applications in new forms of heterolayer-structures are discussed, and interesting combinations of k-space and real space transfer effects can be achieved which have potentially new applications.
Abstract: Research on transfer of electrons between semiconductor heterolayers is reviewed and its importance and applications in new forms of heterolayer-structures are discussed. It is shown that the massive transfer of hot electrons transversal to the layers is interesting from both a device and a physics point of view due to the high speed of the effect and due to its dependence on the details of the electron energy distribution under conditions far away from equilibrium. Less visible but of no lesser importance for the understanding of transport phenomena is the spreading of hot electrons in confining fields such as the gate field in field effect transistors. This spreading can be viewed as a real space transfer effect and causes additional mobility degradation and generally the enhancement of effects of nonlinear transport, notably impact ionization. The transfer of minute electron numbers from silicon into silicon dioxide appears as a special case of real space transfer from this point of view. It will be shown, however, that a complete understanding of this latter effect involves complex quantum-transport principles. Finally, details of nonlinear transport of electrons over a well structure will be discussed. It will be shown that interesting combinations of k-space and real space transfer effects can be achieved which have potentially new applications. These effects vividly demonstrate the influence of the band-structure (band-gap engineering) on the energy distribution function of the electrons.

Journal ArticleDOI
TL;DR: In this article, a simple analytical model for the transfer admittance (dynamic transconductance) of MOS transistors is presented, which establishes a direct correlation between the transfer admission and the interface state admittance in an explicit analytical form.
Abstract: A simple analytical model for the transfer admittance (dynamic transconductance) of MOS transistors is presented. This model establishes, for the first time, a direct correlation between the transfer admittance and the interface state admittance in an explicit analytical form. Experimental measurements have been performed and the obtained results are in good agreement with our theory. Subsequently, we present a new method for characterising interface states in MOS transistors of channel lengths less than 10 μm from the measurement of the imaginary part of the inverse of the transfer admittance.

Journal ArticleDOI
TL;DR: In this article, a unique hot-electron transport model suitable for studying submicron GaAs device structures is presented, which is based on the semiclassical "hydrodynamic" conservation equations for the average electron density, momentum, and energy.
Abstract: A unique hot-electron transport model suitable for studying submicron GaAs device structures is presented. The model is based on the semiclassical “hydrodynamic” conservation equations for the average electron density, momentum, and energy. The model includes electron relaxation times, momentum relaxation times, energy relaxation times, electron temperature tensors and heat flow vectors as a function of average electron energy for the Γ, X and L valleys of GaAs. The relaxation times represent rates of exchange of electrons between valleys and rates of loss of average momentum and average energy between and within the individual valleys. The electron temperature tensor and heat flow vector depend on the electron velocity distribution about the average electron velocity and ultimately affect transport when spatial variations in average velocity and average energy exist. Transport parameters are calculated using the Monte Carlo method and the ergodic principle applied directly to the integral definitions for the parameters. Therefore, the model includes nonequilibrium transport effects such as velocity overshoot and nonuniform average electron energy. The new model should prove instrumental in optimizing electron transport through submicron structures for high-speed device applications.

Journal ArticleDOI
TL;DR: In this paper, the authors found that the barrier height of Al-Schottky diodes formed on surface oxidized GaAs was smaller than those on as-etched GaAs, probably due to reduction of the surface oxide layer by Al.
Abstract: The barrier heights of Al-Schottky diodes formed on surface oxidized GaAs have been found to be smaller than those on as-etched GaAs. They recovered to the as-etched value when the samples were annealed, probably due to reduction of the surface oxide layer by Al. The barrier height of Au-Schottky diodes formed on surface oxidized GaAs depend on the oxidation conditions. These variations of the barrier height can be explained by differences of the oxide compositions and bonding strengths of the metal oxides (AlO, GaO etc.). In accordance with decrease of the barrier height, the DLTS (Deep Level Transient Spectroscopy) signal of the midgap level EL2 decreased and the signal peak shifted to a lower temperature even for the AuGaAs Schottky barrier. The decrease of the DLTS signal can be explained by a change of the occupation factor due to change of the reverse saturation current density.

Journal ArticleDOI
TL;DR: In this article, a model for the ohmic operation of a MOS transistor at very low temperature (4 −40 K) is presented, based on a quantum treatment of the inversion layer and on a specific low temperature mobility law.
Abstract: A model for the ohmic operation of a MOS transistor at very low temperature (4–40 K) is presented. The model is based on a quantum treatment of the inversion layer and on a specific low temperature mobility law. It enables a good description of the MOSFET transfer characteristics (field effect mobility, drain current) as a function of gate voltage and predicts the temperature dependence of the maximum field effect mobility and of the threshold voltage.

Journal ArticleDOI
TL;DR: In this article, the electrical behavior of the Al-Si contact and its effect on the performance of solar cells has been studied and fabrication methods are proposed to overcome the contact problems associated with low doping concentration at the metal-semiconductor interface.
Abstract: The electrical behaviour of the AlSi contact and its effect on the performance of solar cells has been studied. When the sheet resistance ϱs of the n-layer is greater than about 200 Ω/□ the contact will function as a rectifier and the dark I–V characteristics of the solar cell is mainly limited by the reverse biased MS contact. This accounts for the large ideality factor (> 2) observed experimentally. In addition the illuminated I–V curve of the solar cell in the fourth quadrant has an S-shape with an inflection around zero cell current, and the conversion efficiency is considerably reduced. In fact this inflection is a diagnostic tool for a rectifying MS contact. A circuit model has been developed to describe the behaviour of a solar cell with any internal construction and under all operating conditions. This model is confirmed by the electrical behaviour. Finally fabrication methods are proposed to overcome the contact problems associated with low doping concentration at the metal-semiconductor interface.

Journal ArticleDOI
TL;DR: In this paper, the Schottky diode, a surface barrier diode and an n+p junction on 3 × 103 Ω cm material (C) were studied and the results showed that the dispersive process, known as Low-Frequency Dispersion (LFD), which may go over into negative capacitance under forward bias, is seen only in interfacial devices.
Abstract: Dielectric studies, equivalent in many respects to the familiar admittance spectroscopy, are reported on three silicon barrier devices: a Schottky diode on 10 Ω cm n type (A), a surface barrier diode on 103 Ω cm n-type (B) and an n+-p junction on 3 × 103 Ω cm material (C). The response in the frequency range 0.01–104 Hz and in the temperature range 10–325 K shows three principal features in A and B: the d.c. conductance, a strongly dispersive behaviour at low frequencies and high temperatures, a secondary loss peak associated with the d.c. conduction, and a high-frequency loss peak which is distinctly broader than Debye. The dispersive process which is known as Low-Frequency Dispersion (LFD) and which may go over into negative capacitance under forward bias, is seen only in interfacial devices. The n+-p junction shows only the d.c. process and the high-frequency loss peak which is almost Debye-like. Several of these features have been seen previously in other devices, especially LFD in GaAs, and it is significant that they are now seen in silicon barrier devices implying that they are not simply a consequence of the compound nature of the material. They are related to the presence of electrochemical interaction in the oxide layer under the metal contact.

Journal ArticleDOI
TL;DR: In this paper, a comparative study of d.c. magnetron sputtered and evaporated Ti/p-InP Schottky barriers was made, by measuring the barrier height increase, the shift of the Mott-Schottky lines and the change of the ideality factor due to sputtering, it was shown that donor-type defects were introduced by this deposition technique.
Abstract: A comparative study was made of d.c. magnetron sputtered and evaporated Ti/p-InP Schottky barriers. By means of measuring the barrier height increase, the shift of the Mott-Schottky lines and the change of the ideality factor due to sputtering, it could be shown that donor-type defects were introduced by this deposition technique. An electrostatic model based on a rectangular defect concentration profile enabled to calculate both the depth (0.06–0.4 μ m) and the doping concentration (1021–1023 m−3) of the n-type layer formed beneath the surface. Both quantities were found to depend on the original substrate doping concentration. The model was supported by the observed changes in the photovoltaic behaviour caused by sputtering. An important increase of the open circuit voltage occurred (0.35–0.70 V) together with a decrease of the spectral response in the short-wavelength region. It could be concluded that magnetron sputtering increases the photovoltaic efficiency (from 3.5 to 9.5%) and leads to a Schottky barrier height increase (0.2 eV) which is comparable to that obtained by more elaborate techniques (e.g. ion implantation).

Journal ArticleDOI
TL;DR: In this article, an analytical theory has been developed for the light emission of a reverse-biased silicon p-n junction based on the theory of ionization and indirect recombination of electrons and holes under high-field conditions, the relative intensity of emitted light has been derived.
Abstract: An analytical theory has been developed for the light emission of a reverse-biased silicon p-n junction. Based on the theory of ionization and indirect recombination of electrons and holes under high-field conditions, the relative intensity of emitted light has been derived. It is observed that the normalized results of intensity are in good agreement with the experimental results for an electron ionization length l ioe = 60 A , and a hole ionization length l ioh = 70 A . Thus it confirms the validity of this theoretical model for light emission in a reverse-biased silicon p-n junction.

Journal ArticleDOI
TL;DR: AuGeNi/InP contacts with excellent surface morphology and edge definition were obtained for alloying temperatures between 360°C and 460°C, with contact resistance around 0.02-0.07 μ · mm.
Abstract: AuGeNi ohmic contact formation on Si-implanted InP has been investigated. Ohmic contacts suitable for application to field-effect transistors are obtained at alloying temperatures between 360°C and 630°C, with contact resistance around 0.02–0.07 μ · mm. A liquid phase appears at an alloying temperature of about 460°C. in consequence, contacts with excellent surface morphology and edge definition are obtained for alloying temperatures between 360°C and 460°C. Four different regimes in the alloying temperature behavior of the AuGeNi/InP system are observed and their physical origin is discussed. A close to quadratic correlation exists between the achieved contact resistance and the underlying semiconductor sheet resistance.

Journal ArticleDOI
TL;DR: In this article, a Monte Carlo method for the solution of the Boltzmann equation is presented, where the electron state k at which the ditribution function is evaluated at time t is chosen at the beginning of the procedure and the electron paths are generated backward in time from t to the time t = 0 of the (known) initial condition.
Abstract: We present a basically new Monte Carlo method for the solution of the Boltzmann equation. One of the major features of this method is that the electron state k at which the ditribution function is evaluated at time t is chosen at the beginning of the procedure and the electron paths are generated backward in time from t to the time t=0 of the (known) initial condition. The possibility of fixing arbitrarely the value of k at which ƒ is evaluated, makes the method particularly appealing for problems where rare regions of ƒ are of particular interest.

Journal ArticleDOI
TL;DR: In this paper, the 1/ f spectrum from the collector is attributed to the scattering of carriers during diffusion through the forward biased base rather than their recombination, and agreement and disagreement are found between α Hp and α Hn as given by the theory of acoustic phonon scattering and the experimental Hooge's constants.
Abstract: For the first time flicker noise in the collector of p + - n-p Ge82-185 silicon bipolar transistors is observed unobstructed by the amplified base noise sources. The 1/ f spectrum from the collector is attributed to the scattering of carriers during diffusion through the forward biased base rather than their recombination. Both agreement and disagreement is found between α Hp and α Hn as given by the theory of acoustic phonon scattering and our experimental Hooge's constants. No evidence is found to indicate either Umklapp or intervalley scattering as possible sources for the 1/ƒ noise.

Journal ArticleDOI
TL;DR: In this paper, graded composition AlGaAs layers are used to generate a non-equilibrium distribution of electrons, in order to spatially control the transfer of electrons from the Γ to L valleys.
Abstract: The use of semiconductor multilayers for hot electron injection and intervalley transfer have recently been demonstrated. In this study graded composition AlGaAs layers are used to generate a non-equilibrium distribution of electrons, in order to spatially control the transfer of electrons from the Γ to L valleys. The consequent growth of charge and field instabilities has been investigated both experimentally and by Monte-Carlo simulation. Particular emphasis has been placed on the exploitation of these hot electrons in transferred electron devices. Gunn diodes consisting of such graded gap layers, and transit regions of 1 μm have been fabricated by molecular beam epitaxy. They have been observed to oscillate at a frequency of 95 GHz. Experimental results on these structures, together with their Monte-Carlo simulations, are discussed.