An Accurate Combination of on-the-fly Interface Trap and Threshold Voltage Methods for NBTI Degradation Extraction

The Negative bias temperature instability (NBTI) is one of the most important reliability issues for modern CMOS technology. Accurate reliability prediction necessitates physically based models for NBTI and accurate methods for estimation of interface (?N _it ) and oxide trap (?N _ot ) generated under this degradation as well as mobility degradation (?μ _eff /μ _eff0 ). In this paper, we propose an accurate approach to estimate ?N _it , ?N _ot and ?μ _eff /μ _eff0 induced by NBTI degradation. This approach is based on combining on-the-fly interface trap (OTFIT) and on-the-fly threshold voltage (OTF-V_th) methods in the same time measurement setup, contrary to the classical combination where the two methods (OTFIT and OTF-V_th) are applied separately in two different measurements setups and using two transistors. In addition, the contribution of border trap to the charge pumping (CP) current in OTFIT is minimized using the high frequency signal and the scan band energy of the two combined methods is calibrated. Therefore, the data set of OTFIT and OTF-V_th can be directly comparable. The proposed approach can contribute to further understand the behavior of the NBTI degradation, especially through the mobility degradation and the threshold voltage shift contributions of interface (?V _it ) and oxide traps (?V _ot ).     

A two-dimensional elastic–plastic finite element analysis of friction effects on sliding crack surfaces in full or partial contact

Sliding crack surfaces are analysed, that are completely or partially in contact, using a two-dimensional plane-stress elastic–plastic finite element technique. Our in-house program was modified to account for the friction which acts between two rough mating surfaces. The analysis is applied to a cantilever beam cracked along its span through its centroidal plane. Twelve cracks with length-to-span ratios ranging from zero to 0.5 were analysed. The effect of friction was investigated by considering 0, 0.4, 0.8, 1.2 and 1.6 as values for the coefficient of friction with each crack length. The results show the influence of friction on the beam stiffness, strain energy release rate, modes of crack tip and surface displacements, and the development of plastic deformation. The present finite element outputs assist in the explanation of experimental events associated with mode II crack tip displacement data found in the literature.     

Analysis of weld-induced residual stresses and distortions in thin-walled cylinders

Circumferential weld specifically in thin-walled structures is a common joint type in the fabrication of structural members in aerospace, aeronautical and pressure vessel industries. This type of weld joint suffers various types of weld-induced residual stress fields (hoop and axial) and deformation patterns (axial shrinkage, radial shrinkage). These imperfections have negative effects on fabrication accuracies and result in low strength welded structures that can lead to premature failures. To precisely capture the distortions and residual stresses, computational methodology based on three-dimensional finite element model for the simulation of gas tungsten arc welding in thin-walled cylinders is presented. Butt-weld geometry with single “V” for a 300 mm outer diameter cylinder of 3 mm thick is used. The complex phenomenon of arc welding is numerically solved by sequentially coupled transient, non-linear thermo-mechanical analysis. The accuracy of both the thermal and structural models is validated through experiments for temperature distribution, residual stresses and distortion. The simulated result shows close correlation with the experimental measurements. This paper was recommended for publication in revised form by Associate Editor Dae-Eun Kim Naeem Ullah Dar received the B.Sc. and M.S. degrees in mechanical engineering from the Univer-sity of Engineering & Technology, Taxila, in 1989 and 2004 respectively. Presently, he is PhD scholar in mechanical engineering at UET, Taxila. His publications are over 25 in different Int. journals and conferences. His research includes manufacturing process (GTAW welding process, HSM process, abrasive waterjet process, incremental forming etc), welding simulations, optimization, and expert system. He spent more than 16 years in different mechanical manufacturing fields. He also received MBA degree in project management and six sigma black belt from SQII in 2005. Ejaz M. Qureshi is currently a graduate student working for his PhD in computational weld mechanics at National University of Sciences and Technology (Pakistan). After receiving his B.S. degree in Mechanical Engineering in 1997, he worked for five years in an industrial manufacturing setup producing hi-tech welded structures. Qureshi has published numerous technical papers in professional refereed journals and conferences of international repute. He has also been an active referee for several conferences and journals. His current research interests include: manufacturing processes simulation, structural integrity of welded structures and computational plasticity of cylinders/pressure vessels. M.M.I. Hammouda received the B.Sc. and M.Sc. degrees in mechanical engineering from Al Azhar University, Cairo, Egypt in 1970 and 1975 respectively. He received his PhD degree in mechanical engineering from Cambridge University, England in 1978. His has more than 35 publications in different International journals and conferences. The research includes mechanical behavior of engineering materials, linear elastic and elastic-plastic fracture mechanics, manufacturing process modeling and simulations. Presently being a foreign faculty professor in mechanical engineering department of UET, Taxila, Pakistan, he spent more than 25 years in academic and teaching. He is a member of the Editorial Board of the International Journal of Fatigue and Fracture of Engineering Materials and Structures. [www.blackwell-science.com].     

FRONT DEVELOPMENT OF A LONG FATIGUE CRACK DURING ITS GROWTH

Abstract— A 3-D elastic-plastic finite element analysis has been developed to simulate the deformation development along the front of a long mode I single edge crack in plates subjected to either monotonic or cyclic loading. Idealisations having both equal and unequal layers through the thickness of the plate were involved. Plane stress and plane strain 2-D finite element analyses were also performed and compared with the present 3-D solutions. The development of the monotonic and cyclic crack tip plastically deformed zones and opening displacements were traced and correlated to accommodate the effect of the plate thickness and the profile of the crack front. A previously developed crack tip deformation parameter was invoked to predict the effect of the specimen thickness on mode I fatigue crack growth and the associated change of crack front profile. Comparison of such a prediction and the experimental findings of the present work reflected the capability of that parameter in modelling fatigue crack growth through the plate thickness.     

Effect of melt superheat and chill material on interfacial heat-transfer coefficient in end-chill Al and Al-Cu alloy castings

Solidification of metal castings inside moulds is mainly dependent on the rate of heat removal from the metal to the mould. During casting solidification, an air gap usually develops at the interface between the solidfying metal and the surrounding mould or chill. This condition occurs in most casting geometries, except in some cases such as the cast metal solidifying around a central core. An overall heat-transfer coefficient, which includes all resistances to heat flow from the metal to its surroundings can be determined. The objective of this work was to determine the overall heat-transfer coefficient,h, using experimental and computersimulation results on commercial purity aluminium and Al-4.5 wt% Cu alloy solidifying in a vertical end-chill apparatus. The cast ingots had a cylindrical shape with 12.5 mm diameter and different lengths of 95 and 230 mm. It solidified at different superheats (ranging from 50–110 °C) against two different chill materials: copper, and dry moulding sand. A computer program solving the heat-conduction equation and taking into consideration the convection in the melt, was used to compute the temperature history at numerous points along the ingot length. Differenth values were assumed as a function of time, until agreement between experimental and computed cooling curves was obtained. The variation ofh as a function of time, surface temperature, specimen length for each melt superheat and chill material was found. The thickness of the air gap was also evaluated. The results indicate that the variation of heat-transfer coefficient with time followed a pattern of sudden increase for the first few seconds, followed by a steady state, after whichh decreased and reached another lower constant value. Theh values were also found to decrease rapidly when the liquidus temperature was reached in the melt. For longer specimen and higher melt superheat, the heat-transfer coefficient increased. It was also higher for a copper than for a sand chill.     

Thermal conductivity and viscosity of self-assembled alcohol/polyalphaolefin nanoemulsion fluids

Very large thermal conductivity enhancement had been reported earlier in colloidal suspensions of solid nanoparticles (i.e., nanofluids) and more recently also in oil-in-water emulsions. In this study, nanoemulsions of alcohol and polyalphaolefin (PAO) are spontaneously generated by self-assembly, and their thermal conductivity and viscosity are investigated experimentally. Alcohol and PAO have similar thermal conductivity values, so that the abnormal effects, such as particle Brownian motion, on thermal transport could be deducted in these alcohol/PAO nanoemulsion fluids. Small angle neutron-scattering measurement shows that the alcohol droplets are spheres of 0.8-nm radius in these nanoemulsion fluids. Both thermal conductivity and dynamic viscosity of the fluids are found to increase with alcohol droplet loading, as expected from classical theories. However, the measured conductivity increase is very moderate, e.g., a 2.3% increase for 9 vol%, in these fluids. This suggests that no anomalous enhancement of thermal conductivity is observed in the alcohol/PAO nanoemulsion fluids tested in this study.     

3-D finite element analysis of stress concentration factor in spot-welded joints of steel: The effect of process-induced porosity

The work presented in this paper utilises a numerical analysis for the computation of stress concentration factor generated by the presence in the weld nugget of a pore formed during the welding process. Welded structure containing porosity is subjected to uniaxial tensile stress. The effects of geometrical parameters of the pore and the interaction pore-defect on the stress concentration factor variation have been analyzed.     

Structure of semidilute single-wall carbon nanotube suspensions and gels

The microscopic network structure of surfactant-stabilized single-wall carbon nanotubes (SWNTs) in water was investigated as a function of SWNT concentration in the semidilute (overlapping) regime using small-angle neutron scattering (SANS). Most of the samples exhibit rigid rod behavior (i.e., Q(-1) intensity variation) at large scattering wavevector, Q, and a crossover to network behavior (i.e., approximately Q(-2) intensity variation) at low Q. The mesh size, xi, of the network was determined from the crossover of rigid rod to network behavior in the SANS intensity profile and was found to decrease with increasing SWNT concentration. When the dispersion quality of these associating rigid rods was degraded, only approximately Q(-2) intensity variation was observed at both high and low Q. Small-angle X-ray scattering measurements of the same stable dispersions were relatively insensitive to network structure because of poor contrast between SWNTs and surfactant.     

Deformation behaviour at the tip of a physically short fatigue crack due to a single overload

A two-dimensional elastic–plastic finite element analysis was utilized to investigate the transition behaviour of a physically short fatigue crack following the application of a single overload cycle. The deformation accommodated at the tip of a crack artificially advancing with a fully reversed load was considered. The development of the cyclic crack tip opening displacement was computed and then modelled to include the effects of the stress level of the base cycles, overload pattern and crack length at which the transient cycle was applied. The cyclic crack tip opening displacement was initially of a relatively high value. It decreased and then increased to match the behaviour under the base load cycles. The extent and location of both the minimum and matching points were dependent on the overload crack length and the stress compared with the material’s yield stress. In the case of the yield stress being exceeded by the overload, the minimum and the-return-to-normality points are identical. A previously developed crack tip deformation parameter was invoked to predict relevant experimental fatigue growth rates of short cracks reported in the literature.