Transit Time Effects Research Articles

Noise temperature of n+nn+ GaAs structures.

The noise temperature of GaAs ${\mathit{n}}^{+}$${\mathit{nn}}^{+}$ two-terminal structures of micrometer and submicrometer lengths is theoretically analyzed as a function of frequency and applied voltage. Calculations are performed at a kinetic level, and are based on a mixed Monte Carlo hydrodynamic scheme. Different operation modes of the structure are considered, namely, when operating as a nonlinear resistor (passive device) as well as when operating as a Gunn amplifier or generator (active device). Under generating conditions the noise temperature at low frequency is found to go to infinity for voltages above the threshold value for the Gunn effect, as expected in bulk material. Under amplifying conditions the noise temperature at low frequency is found to remain finite even at applied voltages well above the threshold value for the Gunn effect. At high frequencies the noise-temperature spectrum shows structures associated with transit-time effects, carrier energy, and plasma-time effects. In particular, we calculate the noise figure of merit covering both amplifying and passive conditions. Very good agreement is found between theory and available experiments. \textcopyright{} 1996 The American Physical Society.

Angle dependence and transit time effect in Doppler flow measurement

Objective: In clinical practice, obvious differences exist between the Doppler theory which does not allow a frequency shift at right angles of insonation, and the fact that a frequency is generated. Possible technical and physical causes were evaluated in our study. Methods: A laminar, non-pulsatile flow model was insonated between 35° and 155° with pulsed wave (pw) Doppler sonography incorporated in a duplex scanner. Results: For all angles tested the maximum frequencies were higher than predicted by the Doppler formula. The frequency shift observed at perpendicular insonation cannot be explained by the Doppler theory. Conclusions: The results obtained can be explained by intrinsic spectral broadening (ISB) caused by the transit time effect or equivalently by the geometry of the ultrasonic beam. These effects have to be considered when the maximum frequency is employed to quantify flow volume or when spectral broadening is used as an indicator of disturbed flow.

Angle dependence and transit time effect in Doppler flow measurement

Objective: In clinical practice, obvious differences exist between the Doppler theory which does not allow a frequency shift at right angles of insonation, and the fact that a frequency is generated. Possible technical and physical causes were evaluated in our study. Methods: A laminar, non-pulsatile flow model was insonated between 35° and 155° with pulsed wave (pw) Doppler sonography incorporated in a duplex scanner. Results: For all angles tested the maximum frequencies were higher than predicted by the Doppler formula. The frequency shift observed at perpendicular insonation cannot be explained by the Doppler theory. Conclusions: The results obtained can be explained by intrinsic spectral broadening (ISB) caused by the transit time effect or equivalently by the geometry of the ultrasonic beam. These effects have to be considered when the maximum frequency is employed to quantify flow volume or when spectral broadening is used as an indicator of disturbed flow.

Radial acceletron, a new low-impedance HPM source

The transit-time effect in a coaxial structure was explored to develop a low-impedance high-power microwave (HPM) source that uses no external magnetic field and no confining foils. This source will work in the 1-20 GHz range and will have a power output of no less than 1 GW. The input will be a low-voltage DC pulse of only 350 kV or less with a flat top of approximately 200 ns. The DC pulse is launched into a coaxial structure that is the diode, the oscillator, and the buncher all in one. The source offers significant improvements in power, repetition rate, size, and efficiency. Because of the coaxial structure, the diode impedance may be reduced to a few ohms thus allowing larger input and output powers. With no foils to erode, the only factor limiting the repetition rate is the ability to maintain an adequate vacuum, and since there is no external magnetic field required, the device is simple, lightweight, and inexpensive. Because of the strong bunching, the efficiency is high. As is the case with all transit-time oscillators, the signal is stable and monochromatic. The device may be used as a buncher or as an oscillator. It is shown that a device based on this concept is indeed possible. In addition, it is shown that the gated emission of electrons, a process basic to high-power RF amplification, is a natural by-product of the mechanism used in this device.

Large-signal modeling of self-heating, collector transit-time, and RF-breakdown effects in power HBTs

A large-signal heterojunction bipolar transistor (HBT) model has been developed which includes self-heating, collector transit-time, and RF-breakdown effects. The model has a compact form which is based on a compromise between accuracy and utility. As such, the model can be readily extracted and verified with the aid of RF waveform measurements. Using the model in simulations, it was found that RF breakdown was dependent on base biasing and loading conditions. Therefore, with proper circuit design, the maximum output power of the HBT can significantly exceed the limit of open-base breakdown voltage.

Effects and modeling of simultaneous switching noise for BiCMOS off-chip drivers

A model of simultaneous switching noise (SSN) for the conventional BiCMOS driver including the influence of negative feedback is presented. Level 1 SPICE-type MOSFET and Gummel-Poon BJT device models are used for the analysis. The model accounts for a reduction in forward current gain, /spl beta//sub F/, that is attributed to high-level injection effects. Closed-form expressions are presented that predict the maximum SSN encountered in low-level and high-level injection. Solutions of the first equation are combined with the second in order to model the complete SSN transient. Model predictions are compared to SPICE simulation results for 5 V, 1 /spl mu/m drivers and 3.5 V, 0.5 /spl mu/m drivers with close agreement. The effects of BJT parasitic resistance, junction capacitance, and forward transit time on reducing induced ground noise are also explored. With the introduction of these parasitics, SPICE predictions of the maximum SSN are approximately 20% lower than closed-form model results. Immunity of the quiet BiCMOS driver to noise present at the ground and power connections is also investigated. It is found that the quiet driver is highly immune to adverse effects caused by SSN at the power connection and partially immune to SSN present at the ground. Coupling between ground noise and the output occurs when the ground noise is greater than two forward diode drops above ground, in which case the output tracks the ground noise with a negative offset of one diode drop.

Frequency response of avalanche photodetectors with separate absorption and multiplication layers

We present analytical expressions for the frequency response of avalanche photodetectors (APDs) with separate absorption and multiplication regions (SAM). The effect of the electric field profile in the multiplication layer on frequency response is considered for the first time. Previous theories have assumed that the multiplication layer is very thin and the peak electric field, which corresponds to the effective multiplication plane, is positioned away from the absorption layer. This is a poor assumption for many devices, and in particular for silicon hetero-interface photodetectors (SHIPs). We present a theoretical model in which the thickness of the multiplication layer is arbitrary and the peak electric field may be positioned arbitrarily in relation to the absorption layer. We also consider the effects of parasitics, transit-time, and avalanche buildup time. Both front and back illumination from either multiplication layer or absorption layer are considered. The calculated results are compared with experimental results for existing SHIP's and performance predictions are also made for optimized SHIP structures. SHIP APDs with gain-bandwidth product in excess of 500 GHz are possible.

Design of Rapid-Transition Multiproduct Processes

In continuous processes where products are changed frequently, the time required to transition from one product to the next can significantly impact the overall operating cost by changing the cycle time and inventory profiles. The design of the plant influences the dynamics of this transition and consequently the overall cost. In this paper we present a methodology to incorporate the effect of transition time into the design of a continuous multiproduct process. The design of an exothermic continuous stirred tank reactor is used to demonstrate this algorithm. The resulting design is not necessarily the steady-state economic choice nor the design that makes the transition the fastest but rather an economic compromise of the two objectives.

Ultrafast, dual-depletion region, InGaAs/InP p-i-n detector

The design of a new kind of photodetector, the dual-depletion region p-i-n photodetector, is presented. This vertical detector has a parasitic capacitance and transit time that can be controlled semi-independently. This eases the classical tradeoff between these two speed limiting factors, allowing the design of large, fast detectors. A theoretical analysis of the transit time effect and the capacitance effect is made. This analysis is then used to compute optimum design parameters.

Notching as an example of charging in uniform high density plasmas

Numerical simulation was used to study both surface charging and ion trajectory distortion during submicron patterning in high density plasma etching. The plasma was assumed uniform and the cause for the surface charging was the directionality difference between ions and electrons. The role of ion transit time effects on the ion energy distribution function was also considered, while the effect of discharging currents such as through insulators was not included. Using a Monte Carlo sheath simulator, a Poisson equation solver, and an ion/electron trajectory simulator, the steady state potential distribution and ion trajectories were calculated for various line-and-space structures and plasma conditions where notching, which is a local sidewall etching, has been observed after the overetching part of polysilicon etching processes. The results show significant positive charging at the bottom of high aspect ratio spaces which depends on the ion energy distribution function. Notching at the bottom of an outermost polysilicon line before a wide space is the result of ion deflection toward the line which has the lower potential from receiving more electrons from a side facing the wide space. The simulator was validated using previously reported electron cyclotron resonance etcher notching results. It was found that the calculated potential difference between the bottom of a space and the adjacent line shows the same qualitative dependence on the wide space width as experimental notch depth results show. In addition, these potential differences are large enough to substantially increase the ion flux at regions where notching occurs for the ion energy distribution function for the plasma conditions used. Thus this shows aspect ratio dependent charging effects which are not dependent on initial plasma nonuniformity.

Inhibition of methanogenesis by human bile.

The factors that regulate methanogenesis in humans have not been established. The presence of bile acid, which is lost into the colon from the small intestine, may be an important regulatory factor of methanogenesis. To examine this possibility, the effect of human bile on methane production by faecal cultures, and the in vivo effect of biliary diversion on breath methane excretion in a methanogenic choledochostomy patient, were investigated. Faecal suspensions (0.1%) from five methanogenic humans were incubated anaerobically with bile (0.3-30%) from three choledochostomy patients, and headspace methane measured by gas chromatography. All biles inhibited headspace methane. Inhibition of methanogenesis was dose dependent, plateaued at 10-30% bile concentration, and was abolished by 0.6% cholestyramine. The maximum inhibition by bile, median (range), was 38 (0.9-56)% of control methane values. Reversal of the bile fistula in the fourth choledochostomy patient converted that subject from methanogenic to 'non-methanogenic' status, It is concluded that inhibition of methanogens in the caecum by bile acid could significantly reduce the number of methanogens in the colon. This and the effect of transit time could explain much of the known epidemiology of 'non-methanogenesis', which has been related to obesity, (comparatively) fast colonic transit in healthy persons, and to small intestinal Crohn's disease.

Acceleration and scattering of injected electrons in plasma beat wave accelerator experiments*

The results from experiments in which a two-frequency CO2 laser is used to beat-excite large-amplitude, relativistic electron plasma waves in a tunnel-ionized plasma are reported. The plasma wave is diagnosed by injecting a beam of 2 MeV electrons and observing the energy gain and loss of these electrons, as well as the scattering and deflection of the transmitted electrons near 2 MeV. Accelerated electrons up to 30 MeV have been observed. The lifetime of the accelerating structure as seen by small-angle Thomson scattering is about 100 ps, whereas the injected electrons are seen to be scattered or deflected by the plasma for several ns, with diffuse scattering occurring 0.5–1 ns after forming the plasma wave and whole beam deflection occurring at later times. A simple model, which includes laser focusing, ionization, transit time, and relativistic saturation effects, suggests that the wave coherence may be short lived while the wave fields themselves persist for a longer time. This may be the reason for the disparate time scales between the Thomson scattering and the electron scattering diagnostic. The whole beam deflection may be evidence for a Weibel-like instability at later times.

A statistical analysis of the received signal from blood during laminar flow

In order to determine the limiting velocity resolution that can be achieved using ultrasound, and to provide a model which can be generalized for the analysis of disturbed flow, a theoretical and experimental evaluation of the statistics of the received signal from laminar flow following the transmission of a train of short pulses is presented. The authors derive the autocorrelation function and determine the length of the correlated signal for various flow rates, comparing experimental measurements to theoretical predictions. High resolution experimental RF M-mode images are used to verify the theoretical model. Using a fluid with a density, viscosity, volume concentration, particle size, and speed of sound which is similar to that of blood, the authors show that the signal remains correlated for a long interval under many conditions of clinical interest. Including a comparison with experimental data, the effect of the lateral transit time through the sample volume and the axial velocity spread within the sample volume on the correlation of the received signal is evaluated. When a significant range of velocity components is present within the sample volume, this range is the limiting factor in the length of the correlated signal interval. Therefore, the use of a wideband signal, which reduces the sample volume size, produces a returned signal that may be correlated for a larger number of pulses, or for a longer time.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Time delay measurements using conventional E-beam technology

Abstract The ability to resolve sub-nanosecond phase shifts on internal IC nodes is central to the utility of electron beam test systems. In this work, the obtainable time delay resolution has been investigated both experimentally and via computer modeling. Initial time delay resolution results were obtained at working distances from 1 to 32 mm in a conventional electron beam test system. In order to obtain these results, a signal generation system was developed. This system proved to be adequate for resolution measurements down to 10 ps. Monte Carlo simulated waveforms were generated and used to develop a consistent method of determining the minimum detectable delay. This method, which is based on linear regression and confidence intervals, corresponded well with visual judgments of the resolution. The minimum detectable delays using this criteria ranged from 8 to 44 ps at a static beam current of 20 nA, a pulse width of 150 ps and a 100 kHz repetition rate. The degradation of resolution with increasing working distance was attributed to the transit time effect since there was no significant jitter noise and no increase in noise with working distance evident in the acquired waveforms.

Spike leakage of thin Si PIN limiters

Thin PIN diode limiters (10 /spl mu/m or less) are used to protect sensitive microwave components from fast-risetime microwave pulses having energies exceeding 1 to 10 /spl mu/J. This paper analyzes and experimentally confirms the performance of these PIN limiters. It is shown that spike leakage is a transit-time effect that is controlled by the mobility of the carriers. A p-type background I-region should yield less spike leakage energy for a given thickness. It is proposed that the hysteresis effect observed when limiters are operated under cw conditions is due to space charge effects and stored charges remaining after the reverse-biased half cycle. Detailed agreement between the measured and calculated device voltage waveforms requires accurate modeling of the circuit parasitics because of the high rate-of-change currents arising from avalanching.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Electrodynamics of fast beam blankers

Performance characteristics of an advanced electron beam blanker for lithography are presented. Various electrodynamic effects are discussed, which must be eliminated to achieve high beam placement accuracy during and after blanking. These electric and magnetic field effects have been measured over six orders of magnitude in time. The fast beam jitter characteristic of transit time effects in a double-deflection blanker is captured with nanosecond time resolution. Eddy current effects measured in the micro- to millisecond time domain are shown to be an inherent problem in earlier double-deflection blanker designs. A consequence is beam misplacement after the unblank transition, which can be 0.05 μm even after 500 μs. Several examples of pattern artifacts in purposely underdeveloped resist are given to illustrate graphically the lithographic consequences of the eddy current effect. All of these electrodynamic effects have been addressed with a new ‘‘virtual ground’’ blanker design. The MEBES IV-TFE maskmaker has been upgraded with this new blanker to meet its accuracy requirements throughout the nano- to millisecond time scale. This enables pattern generation at 160 MHz with 30 nm (3σ) critical dimension (CD) control for 2.0 μm features with addresses up to 0.25 μm. The CD control is tested with patterns that are fractured with a single pixel placed across the scan boundary. This situation provides the most extreme test of blanker accuracy at the nanosecond time scale. Blanker eddy current effects are eliminated from all exposures, including those in underdeveloped resist. The blanker design upgrade has facilitated the fabrication of reticles for production 64 Mbit DRAMs and developmental 256 Mbit DRAMs.

Large-signal numerical and analytical HBT models

Several large-signal heterojunction bipolar transistor (HBT) models are investigated to determine their usefulness at millimeter-wave frequencies. The most detailed model involves numerically solving moments of the Boltzmann transport equation. A description of the numerical model is given along with several simulated results. The numerical model is then used to evaluate two analytical HBT models, the conventional Gummel-Poon model and a modified Ebers-Moll model. It is found that the commonly used Gummel-Poon model exhibits poor agreement with numerical and experimental data at millimeter-wave frequencies due to neglect of transit-time delays. Improved agreement between measured and modeled data results b including transit-time effects in an Ebers-Moll model. The simple model has direct application to millimeter-wave power amplifier and oscillator design. Several measured results are presented to help verify the simple model.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Characterisation of time resolution in electron beam measurements

A method allowing the determination of time resolution in the electron beam testing of integrated circuits is presented, and experimentally validated. The effect of the extraction field on time resolution is clearly demonstrated. It is also shown that a simple formulation of the transit time effect agrees well with experimental data.

Design considerations for wide-band p-i-n/HBT monolithic transimpedance optical receivers

A comprehensive circuit and device model has been developed for the design of wide-band transimpedance optical receivers using heterojunction bipolar transistors (HBT's). Based on this model, we determine the device and circuit design that gives the highest combination of bandwidth, sensitivity, and stability. For optoelectronic integration, it is convenient to use the collector-base junction of the HBT device structure to fabricate the p-i-n detector. A resulting transistor transit-time effect is shown to cause shunt peaking in the closed-loop response or, at worst, instability. It is shown that the photodiode stray ca- pacitance is not a major source of sensitivity degradation in flip-chip transimpedance receivers. Optimum device structures are determined for InP- and GaAs-based HBT receivers with fine-line as well as relaxed geometries.

Transit time effect in electron beam testing voltage measurements

Various approaches to the reduction of transit time effect (TTE) induced voltage measurement errors in electron beam testing are considered: use of shielded test points or conductors, detection of high-energy secondary electrons, application of high extraction fields, and examination of secondary electron energy dispersion properties. A numerical technique based on secondary electron trajectory tracing is described and used in the study. Spectral dispersion is shown to be a marginal effect that is unlikely to be of practical consequence. It is emphasised that TTE measurement errors are fundamentally voltage errors that can arise from any conductor carrying fast signals in the vicinity of the probing point. Thus, in addition to self-induced errors on a conductor being probed, due consideration must also be given to measurement crosstalk from neighbouring conductors. Both aspects are demonstrated experimentally and compared with computer simulations. Recommendations on the design of test points with low TTE errors are provided.