Design aspects of MOS-controlled thyristor elements: technology,simulation, and experimental results

2.5-kV thyristor devices have been fabricated with integrated MOS controlled n⁺-emitter shorts and a bipolar turn-on gate using a p-channel DMOS technology. Square-cell geometries with pitch variations ranging from 15 to 30 μm were implemented in one- and two-dimensional arrays with up to 20000 units. The impact of the cell pitch on the turn-off performance and the on-state voltage was studied for arrays with constant cathode area as well as for single-cell structures. By realizing MOS components with submicrometer channel lengths, scaled single cells are shown to turn off with current densities of several kiloamperes per square centimeter at a gate bias of 5 V. In the case of multi-cell ensembles, turn-off performance is limited due to inhomogeneous current distribution. Critical process parameters as well as the device behavior were optimized through multidimensional numerical simulation     

Hybrid numerical scheme for time-evolving wave fields

Many problems in geophysics, acoustics, elasticity theory, cancer treatment, food process control and electrodynamics involve study of wave field synthesis (WFS) in some form or another. In the present work, modelling of wave propagation phenomena is studied as a static problem, using finite element method and treating time as an additional spatial dimension. In particular, WFS problems are analysed using discrete methods. It is shown that a fully finite element-based scheme is very natural and effective method for the solution of such problems. Distributed WFS in the context of two-dimensional problems is outlined and incorporation of any geometric or material non-linearities is shown to be straightforward. This has significant implications for problems in geophysics or biological media, where material inhomogeneities are quite prevalent. Numerical results are presented for several problems referring to media with material inhomogeneities and predefined absorption profiles. The method can be extended to three-dimensional problems involving anisotropic media properties in a relatively straightforward manner. Copyright © 2006 John Wiley & Sons, Ltd.     

Abrasion Resistance of Field-Assisted Ion-Exchanged Glass

The fracture strength of HF-etched samples was measured as a function of field-assisted K-Na ion exchange depth, as well as the severity of abrasion. As expected, the more severe the abrasion, the greater the depth required to make the strength insensitive to abrasion. An abrasion-independent weakening mechanism that increased with exchange depth was also found to be operative.     

Attrition of spherical electrode carbon particles during batch fluidized combustion

Spherical electrode carbon particles prepared from carbon rods of dry cell batteries have been used to study the attrition behaviour in a bubbling fluidized bed combustor. Experiments have been conducted in a 40 mm I.D. and 1 m high fluidized bed combustor operated at 1 m/s superficial velocity. The bed was operated with nitrogen and with two different oxygen concentrations at 850°C to study the effect of combustion on attrition of these particles. The experimental technique used allowed the time resolution of attrited fines generation, providing detailed curves of attrition rates as a function of time. Attrition rate constants have been evaluated. Results show an enhancement of attrition due to combustion even for spherical, homogeneous and smooth particles.     

Efficient mining of discriminative co-clusters from gene expression data

Discriminative models are used to analyze the differences between two classes and to identify class-specific patterns. Most of the existing discriminative models depend on using the entire feature space to compute the discriminative patterns for each class. Co-clustering has been proposed to capture the patterns that are correlated in a subset of features, but it cannot handle discriminative patterns in labeled datasets. In certain biological applications such as gene expression analysis, it is critical to consider the discriminative patterns that are correlated only in a subset of the feature space. The objective of this paper is twofold: first, it presents an algorithm to efficiently find arbitrarily positioned co-clusters from complex data. Second, it extends this co-clustering algorithm to discover discriminative co-clusters by incorporating the class information into the co-cluster search process. In addition, we also characterize the discriminative co-clusters and propose three novel measures that can be used to evaluate the performance of any discriminative subspace pattern-mining algorithm. We evaluated the proposed algorithms on several synthetic and real gene expression datasets, and our experimental results showed that the proposed algorithms outperformed several existing algorithms available in the literature.     

Mechanical and dynamic mechanical properties of jute fibers–Novolac–epoxy composite laminates

A study was done of jute composite using a polymer matrix of epoxidized Novolac resin (ENR), diglycidyl ether of bisphenol A (DGEBA)–based epoxy, and their blends with different weight percentages of the resins. It was found that on blending ENR with DGEBA, the storage modulii at room temperature are enhanced by about 100% or more in the case of 30 and 40% ENR‐containing matrices, whereas the enhancement in the case of 20 and 12% ENR‐containing matrices is only 50% that of the pure matrix. It was also observed that the tan δ peak heights of the composites containing 30 and 40% ENR are closer to that of 20% ENR‐containing composite. The probable explanation drawn on the basis of experimental findings of DMA and mechanical analysis is that by blending ENR with DGEBA epoxy it is possible to manufacture jute composites with increased stiffness without sacrificing their ductility. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 85: 2800–2807, 2002     

Elevated temperature instrumented Charpy impact of a sintered silicon carbide

A commercially available, densely sintered alpha silicon carbide was tested in air from room temperature to 1700°C using a modified instrumented pendulum unit and standard Charpy size test specimens. The resistance heated silicon carbide specimens exhibited only elastic behavior, even at 1700°C. A compliance analysis of the test revealed a linear elastic decrease of the absorbed impact energy and also a decrease of the fracture stress of the silicon carbide with increasing temperature.     

Combustion of single char particles in a turbulent fluidized bed

The combustion of single bituminous char particles (4-12 mm diameter) was studied in a turbulent fluidized bed operated at 1098 K using air as the fluidising medium. Results indicated that particles burn with constant density following a shrinking sphere model. Burning rates are much higher than those observed in a bubbling fluidized bed. The rate of transfer of oxygen to the particle surface is also higher than that observed in bubbling beds. A model is proposed to calculate the Sherwood numbers of the burning carbon particles. Experimental values of the Sherwood numbers agree well with those predicted from the model.     

A Formal Methodology for Detecting Managerial Vulnerabilities and Threats in an Enterprise Information System

From information security point of view, an enterprise is considered as a collection of assets and their interrelationships. These interrelationships may be built into the enterprise information infrastructure, as in the case of connection of hardware elements in network architecture, or in the installation of software or in the information assets. As a result, access to one element may enable access to another if they are connected. An enterprise may specify conditions on the access of certain assets in certain mode (read, write etc.) as policies. The interconnection of assets, along with specified policies, may lead to managerial vulnerabilities in the enterprise information system. These vulnerabilities, if exploited by threats, may cause disruption to the normal functioning of information systems. This paper presents a formal methodology for detection of managerial vulnerabilities of, and threats to, enterprise information systems in linear time.