Unstable Devices Research Articles

Stacking faults in3C−, 4H−,and6H−SiCpolytypes investigated by anab initiosupercell method

Recent attempts to make SiC diodes have revealed a problem with stacking fault expansion in the material, leading to unstable devices. In this paper, we present detailed results from a density-functional supercell calculation on the electronic structure of stacking faults which result from glide of Shockley partials in $3C\ensuremath{-}, 4H\ensuremath{-}$ and $6H\ensuremath{-}\mathrm{SiC}.$ It was found [Phys. Rev. B 65, 033203 (2002)] that both types of stacking faults in $4H\ensuremath{-}\mathrm{SiC}$ and two types of stacking faults in $6H\ensuremath{-}\mathrm{SiC}$ give rise to band states, which are strongly localized (confined within around 10 \AA{}) in the direction orthogonal to the stacking fault plane. Based on estimates of the band offsets between different polytypes and a simple quantum-well theory, we show that it is possible to interpret this one-dimensional localization as a quantum-well confinement effect. We also find that the third type of stacking fault in $6H\ensuremath{-}\mathrm{SiC}$ and the only stacking fault in $3C\ensuremath{-}\mathrm{SiC}$ do not give rise to states clearly separated from the band edges, but instead give rise to rather strongly localized band states with energies very close to the band edges. We argue that these localized near band edge states are created by stacking fault induced changes in the dipole moment associated with the hexagonal symmetry. In addition, we have also calculated the stacking fault energies, using both the supercell method and the simpler ANNNI (axial next nearest-neighbor Ising) model. Both theories agree well with the low stacking fault energies found experimentally.

Disciplining technology: Electronic reliability, Cold‐War military culture and the topside ionogram

This essay explores the relationship between cold‐war military electronics and the material culture of space science. Focusing on the world's first ionospheric research satellite, designed and built by the Canadian Defence Research Telecommunications Establishment at the height of the Cold War, it seeks to situate the practices and beliefs that underwrote the reliability of this instrument within more profound changes in electronics and electrical engineering during the 1950s. Rather than invoke a monolithic culture of reliability to explain the work of satellite technicians and engineers, this investigation identifies instead two principle approaches to the question of reliability in cold‐war electronics, their origin in the shortcomings of industrial electronics after the Second World War, as well as the place of the satellite project within them. By concentrating our attention on the conceptual and technical dimensions of electronic reliability, radier than on more traditional aspects of solid‐state research and industrial innovation, we can understand how technicians and engineers, developing weapons systems and scientific instruments alike, struggled to understand and use operationally and cognitively unstable electronic devices, and what these struggles suggest about the complex material and social legacy of the Cold War.

Impact of random facet phases on modal properties of partly gain-coupled distributed-feedback lasers

A first comprehensive study of impact of random phase at cleaved facet on modal properties of partly in-phase gain-coupled distributed-feedback (GC DFB) lasers is carried out both theoretically and experimentally. It is shown that, variation of grating phase of antireflective (AR) facet has a relatively small impact on lasing properties, while that of cleaved facet has a significant influence not only on modal properties such as threshold, efficiency, front-to-back facet efficiency ratio (FB ratio), side-mode-suppression-ratio (SMSR), and transient SMSR (TSMSR), but also on DFB lasers' long term stability against longitudinal mode-switching. Correlation among different laser properties is revealed for GC DFB lasers, and compared with that of index-coupled ones. It is found that FB ratio provides a good description of long-term longitudinal mode stability against CL facet phase variation and equivalent external feedback in GC DFB lasers, while TSMSR reveals their dynamic properties for a fixed facet phase. It is shown that devices with a possible stable operation will exhibit both a large TSMSR and a large FB ratio in GC DFB lasers. Lasers that are located within the transition region between left Bragg mode region and the stably operated right Bragg mode region in phase-space will have either a low TSMSR or a small FB ratio, and, therefore, are susceptible to longitudinal Bragg mode-switching. The combination of TSMSR and FB ratio as a screening criteria provides a simple and yet powerful tool to single out the potentially unstable devices due to random facet phase distribution in AR/CL GC DFB lasers.

Defects and Impurities in Mercuric Iodide Processing

AbstractIn the fabrication of mercuric iodide (HgI2) room temperature radiation detectors, as in any semiconductor process, the quality of the final device is very sensitive to the impurities and defects present. Each process step can change the effects of existing defects, reduce the number of defects, or introduce new defects. In HgI2 detectors these defects act as trapping and recombination centers, thereby degrading immediate performance and leading to unstable devices. In this work we characterized some of the defects believed to strongly affect detector operation. Specifically, we studied impurities that are known to be present in typical HgI2 materials. Leakage current measurements were used to study the introduction and characteristics of these impurities, as such experiments reveal the mobile nature of these defects. In particular, we found that copper, which acts as a hole trap, introduces a positively charged center that diffuses and drifts readily in typical device environments. These measurements suggest that Cu, and related impurities like silver, may be one of the leading causes of HgI2 detector failures.

A new drain-current injection technique for the measurement of off-state breakdown voltage in FETs

A simple three-terminal technique for measuring the off-state breakdown voltage of FETs is presented. With the source grounded, current is injected into the drain of the on-state device. The gate is then ramped down to shut the device off. In this process, the drain-source voltage rises to a peak and then drops. This peak represents an unambiguous definition of three-terminal breakdown voltage. In the same scan, a measurement of the two-terminal gate-drain breakdown voltage is also obtained. The method offers potential for use in a manufacturing environment, as it is fully automatable. It also enables easy measurement of breakdown voltage in unstable and fragile devices.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Spectral ageing pattern of carefully handled silicon photodiodes

Temporal stability is one of the predominant requirements for standard detectors. Si photodiodes are widely used standards for photometric and radiometric measurements. However, recently they have turned out to be very unstable devices, even though carefully handled. The ageing of their responsivity is a spectral effect. Some — very common — types of diodes show ageing rates of up to 12% per year, even in the visible range. Only within a limited spectral range do they exhibit sufficient stability. Similar results for various samples of the same type suggest a very typespecific spectral ageing character. ‘Stable’ diodes show ageing rates of less than 1% per year over the whole spectral range investigated. The results are based on measurements of the absolute spectral responsivity of various Si photodiodes 10 mm in size, from different manufacturers, in the wavelength range from 250 nm to 1200 nm over several years. There are indications that photodiodes with thinner SiO 2 coatings tend to be more stable. The physics of this phenomenon are discussed here.

Microwave Amplifier Design with Potentially Unstable FET's

The concept of maximally efficient gain, as applied to the design of microwave transistor amplifiers with potentially unstable devices, is illustrated by an example, utilizing the parameters of a typical small-signal microwave FET.

Maximally efficient gain: a figure of merit for linear active 2-ports

A new power/gain invariant is proposed as a high-frequency figure of merit for linear active 2-ports. Using this figure of merit, it is feasible to design amplifiers with potentially unstable devices which have power gains comparable in magnitude to Mason&apos;s U function and are optimum in the sense of maximising the 2-port added power.

Power Gain and Stability of Multistage Narrow-Band Amplifiers Employing Nonunilateral Electron Devices

This paper gives an analysis of the transducer power gain and stability of a multistage, narrow-band amplifier employing nonunilateral electron devices. The amplifier is assumed to consist of n identical stages with the input and output terminations. The individual amplifier stage consists of a general active two-port device, such as the transistor, characterized by its four short-circuit admittance parameters, plus a two-terminal interstage network and an ideal coupling transformer. Both the individual amplifier stage and the over-all cascade of n amplifier stages, considered as a composite active two-port, are also characterized by their short-circuit admittance (Y) parameters. Relations between these Y parameters of the individual amplifier stage and the Y parameters of the over-all iterative amplifier have been derived. The transducer gain of the amplifier as a function of the interstage and the terminating network parameters has been studied. The transducer gain is optimized with respect to the external passive terminations and is expressed in terms of a design parameter \gamma , which is directly related to the terminating conductances of the amplifier. It is shown that for an amplifier employing inherently stable active devices, there is a value of \gamma which gives maximum transducer power gain; for an amplifier employing potentially unstable active devices, the optimum transducer power gain of the amplifier will, in general, be a monotonically decreasing function of \gamma . In any case any prescribed value of \gamma determines the maximum gain obtainable from the amplifier. The amplifier's margin toward instability is prescribed through prescribing a number \rho_{l} . For \rho_{l} greater than unity, the amplifier will be stable. Control of \rho_{l} is effected through appropriate choice of the design parameter \gamma . A relation relating \rho_{l} , and \gamma has been derived. Some fundamental considerations in the design of multistage, narrow-band amplifiers employing general active two-port devices are given. Results of experimental two-and three-stage transistor amplifiers are presented which show excellent agreement between the theoretical and the experimental results.