Thermal effects in Ni/Au and Mo/Au gate metallization AlGaN/GaN HEMT's reliability

AlGaN/GaN high electron mobility transistors (HEMT) are key devices for the next generation of high-power, high-frequency and high-temperature electronics applications. Although significant progress has been recently achieved [1], stability and reliability are still some of the main issues under investigation, particularly at high temperatures [2-3]. Taking into account that the gate contact metallization is one of the weakest points in AlGaN/GaN HEMTs, the reliability of Ni, Mo, Pt and refractory metal gates is crucial [4-6]. This work has been focused on the thermal stress and reliability assessment of AlGaN/GaN HEMTs. After an unbiased storage at 350 o C for 2000 hours, devices with Ni/Au gates exhibited detrimental IDS-VDS degradation in pulsed mode. In contrast, devices with Mo/Au gates showed no degradation after similar storage conditions. Further capacitance-voltage characterization as a function of temperature and frequency revealed two distinct trap-related effects in both kinds of devices. At low frequency (< 1MHz), increased capacitance near the threshold voltage was present at high temperatures and more pronounced for the Ni/Au gate HEMT and as the frequency is lower. Such an anomalous “bump” has been previously related to H-related surface polar charges [7]. This anomalous behavior in the C-V characteristics was also observed in Mo/Au gate HEMTs after 1000 h at a calculated channel temperatures of around from 250 o C (T2) up to 320 ºC (T4), under a DC bias (VDS= 25 V, IDS= 420 mA/mm) (DC-life test). The devices showed a higher “bump” as the channel temperature is higher (Fig. 1). At 1 MHz, the higher C-V curve slope of the Ni/Au gated HEMTs indicated higher trap density than Mo/Au metallization (Fig. 2). These results highlight that temperature is an acceleration factor in the device degradation, in good agreement with [3]. Interface state density analysis is being performed in order to estimate the trap density and activation energy.

Theory of a probe in a strong magnetic field

An asymptotic analysis of the Langmuir-probe problem in a quiescent, fully ionized plasma in a strong magnetic field is performed, for electron cyclotron radius and Debye length much smaller than probe radius, and this not larger than either ion cyclotron radius or mean free path. It is found that the electric potential, which is not confined to a sheath, controls the diffusion far from the probe; inside the magnetic tube bounded by the probe cross section the potential overshoots to a large value before decaying to its value in the body of the plasma. The electron current is independent of the shape of the body along the field and increases with ion temperature; due to the overshoot in the potential, (1) the current at negative voltages does not vary exponentially, (2) its magnitude is strongly reduced by the field, and (3) the usual sharp knee at space potential, disappears. In the regions of the C-V diagram studied the ion current is negligible or unaffected by the field. Some numerical results are presented.The theory, which fails beyond certain positive voltage, fields useful results for weak fields, too.

Objective optical classifier for tobacco leaves.

The main objective of the current research was to search the optimum method to segregate the most frequent color commercial quality classes of tobacco leaves (c.v. "Virginia"). These color classes cover the whole continuous color scale, between "Pale Lemon" and "Oxidated Brown". With the usual expert classification there exists a significant level of uncertainty . Within this research, several methods for data discrimination were tested, in order to solve uncertainty. Classification errors below 5% were obtained with this proposed classifier along two different seasons (1994&1995).

Clasificador óptico para hojas de tabaco.

Las tres características fundamentales que definen la calidad y madurez del tabaco son el color, la textura y los aromas, aunque es color es el factor más extendido para su determinación Sin embargo, el color varia de forma continua desde el verde hasta el amarillo y naranja a lo largo del proceso de maduración, de manera que las distintas categorías comerciales de color muestran amplias zonas difusas entre ellas. El materia empleado en e! presente trabajo fue provisto por CETARS A en las campañas l994 y 1995 En ambos casos, las muestras fueron previamente clasificadas por sus expertos en 12 categorías comerciales de color. Las determinaciones objetivas de color se llevaron a cabo empleando un espectrofotómetro (Monolight). Posteriormente el espectro fue recalculado afectándolo del las características del íluminante habitualmente empleado por los expertos (F7). A través de la integración del espectro se determinaron asimismo las coordenadas triestímulo de color (X.Y.Z), equivalentes al proceso de asimilación del color por el ojo humano. Los resultados obtenidos con el presente estudio pueden resumirse de la siguiente manera. 1.- las categorías comerciales de color provistas por los expertos mostraron un alto índice de solapamiento entre ellas, para las coordenadas XYZ, debido a la ausencia de. homogeneidad características de toda muestra biológica (c v. de 20.14% y 10.08% en 1994 y 1995). 2.- a través de un clasificador no-supervisado sobre observaciones de color se ha creado una nueva base de color similar en valores medios a las de CETARSA pero con menor variabilidad intragrupo (6.31% y 5.81% en 1994 y 1995 este sistema permite agrupar observaciones de color similares independientemente de las hojas a las que pertenezcan. 3.- empleo de un clasificador bayesiano sobre la nueva base de color permite clasificar los puntos de color de una hoja con errores inferiores al 4% en ambas campañas. 4.- se ha desarrollado un criterio global de clasificación de hojas en función del color y la variabilidad de color de distintas muestras dentro de una hoja El sistema será validado "in situ" en las próximas campañas.

Revising the Emergency Management Requirements for new generation reactors

The paper presents the application of a new risk-informed methodology for the identification of the Emergency Management Requirements (EMR) to a Generation II, Large size Reactor and a Generation III+ Small Modular Reactor. The results obtained in this test case demonstrate that the actual EMR is conservative, as expected, for the GenII reactor, while the new methodology could be applied for the definition of EMRs for the new generation Nuclear Power Plants, with a possible reduction of the emergency area without loss of safety level. By adopting both probabilistic and deterministic approaches, the study addresses possible accidents and corresponding release scenarios for the two types of reactor, calculates the areas where the accidents have an impact on the population and defines the new EMR considering the health effects on the population.