The physical, electrical and structural properties including optical band gap, microhardness, residual stress, IR spectra and resistivity of Si-B-N films made by plasma-enhanced CVD using a SiH₄-N₂-B₂H₆ gas mixture were studied. The films were mainly composed of mixtures of Si-N compounds and B-N compounds, and the compositional ratio of silicon to nitrogen in the Si-N compounds was almost constant, regardless of the quantity of boron added in Si-B-N films. The compressive stress increased and the microhardness decreased with increasing quantity of B-N compound in Si-B-N films. Boron atoms incorporated in deposited Si-B-N films with silicon-rich composition resulted in a decrease in the resistivity of the films. Good correlation between the resistivity and the optical band gap in the films was observed, by leading to the conclusion that the resistivity of Si-B-N films can be estimated fairly accurately measurement of the optical band gap.

View full abstract

We developed an optical birefringence measurement method that uses a photoelastic modulator and a polarized laser. A He-Ne infrared laser is adopted as the light source for measuring the optical birefringence in silicon wafers. In this paper we explain the theory and process of measurement of stress in Si wafers. The magnitude of principal stress difference as well as the directions are obtained simultaneously and quantitatively using our developed equipment. The optical birefringences of {100}, {111} and {110} Si stressed specimens were measured. From the experimental results, it was found that the photoelastic constant depended on the crystalline orientation and the birefringence direction did not coincide with the principal stress direction. By stress strain analysis of the Si single crystal, it was found that the birefringence direction coincided with the principal strain direction and the relation between the principal strain difference and the retardation was independent of crystalline orientation.

View full abstract

The relation between the static fatigue life at 1000°C and the thickness of a skin layer was investigated using silicon nitride specimens having an as-sintered surface. Using the same specimens, a room-temperature proof test was carried out, and the fatigue life distribution at 1000°C of the specimens subjected to the proof test was compared with that of the untreated specimens. The results obtained are summarized as follows. (1) The thickness of the skin layer influences the static fatigue life at 1000°C, and the fatigue life decreases with increasing thickness of the skin layer. (2) The minimum fatigue life of the specimens subjected to the proof test is not dependent only on the maximum flaw size that corresponds to the proof stress, but it is also dependent on the thickness of the skin layer.

View full abstract

In order to study the effect of the interaction and coalescence of multiple surface cracks on the static fatigue behavior of brittle materials such as glass and ceramics, a single crack and three collinear cracks were introduced on the surfaces of plate specimens of glass ceramics by Vickers microhardness indentation, and these specimens were subjected to a constant four-point bending load. It was found that the time-to-failure for a given applied stress significantly decreased with a decrease in the distance between the centers of the indentation cracks. Therefore, the effect of interaction and coalescence of multiple cracks was shown to be significant. On the basis of the crack growth parameters for the double torsion test, the time-to-failure for multiple cracks was predicted by taking the interaction and coalescence into consideration. It was demonstrated that the proposed multiple cracks coalescence model gives good predictions.

View full abstract

Cyclic fatigue of Si₃N₄/SUS304 joints with copper interlayer was analyzed using the thermoelastic-plastic finite-element method (FEM). It is found that the equivalent plastic strain (ε^^-p) is concentrated on the copper interlayer near the interface of the ceramic side due to the joining process. The equivalent plastic strain range (Δε^^-p) is increased and the residual stresses (σ_x, σ_y) of the ceramic side are relaxed due to cyclic fatigue. Stress singularity near the interface of the ceramic side was also studied. The exponent of stress singularity (λ) and intensity of stress singularity (K) are decreased due to cyclic fatigue. Therefore, fracture originates in the copper interlayer near the interface of the ceramic side due to cyclic fatigue. It is concluded that the cyclic fatigue strength and fracture mechanism of the ceramic/metal joints are influenced by the elastic-plastic behavior of the copper interlayer.

View full abstract

The effect of cyclic fatigue on residual stresses of Si₃N₄/SUS304 joints with copper interlayer was analyzed for 4-point bending specimens using the three-dimensional elastic-plastic finite-element method (FEM). The residual stresses were found to relax due to cyclic fatigue, which was consistent with the results of residual stress measurement by the X-ray diffraction method. The relaxation of residual stresses of the joints and equivalent plastic strain of the copper interlayer increase due to cyclic fatigue. Accordingly, the relaxation of residual stresses is thought to be caused by the elastic-plastic behavior of the copper interlayer. Bending tests were performed after cyclic fatigue and measurement of the residual stresses. The bending strength was found to increase with the decrease of residual stresses due to cyclic fatigue.

View full abstract

Cast defects such as microshrinkage reduce the for fatigue strength and reliabilty of ductile cast iron even though full attention is given for production of the test piece, and so HIP (hot isostatic pressing) is applied to FDI (ferritic ductile cast iron) in order to reduce microshrinkage. After that, rotating bending fatigue tests were carried out. HIP did not change the properties of graphite drains such as mean diameter, nodularity and nodule count. However microshrinkage in the HIP-FDI was completely suppressed. The reduction effect of cast defects by HIP treatment was shown by density measurement and the mesults of close observations of the fracture surface and the crack initiation site. The fatigue strength was improved at the whole region of fatigue life and the fatigue limit was about 1.32 times that of the material subjected to heat treatment with HIP.

View full abstract

Rotational-bending fatigue tests were performed on Co-based alloy thermally sprayed specimen with special focus on the effect of mixed NiCr weight ratio on coating microstructure and on fatigue properties. Effects of post heat treatment (fusing treatment) on coating microstructure and on fatigue strength after 10⁷ cycles were also discussed. The results obtained are as follows: (1) There is considerable scatter in observed fatigue life for specimens sprayed with Co-based self-fluxing alloy. This is because each specimen has its own fatigue strength associated with different porosities and microstructures of coatings. (2) Fusing treatment temperature strongly affects the characteristics of sprayed coating in terms of microstructure and porosity. The fatigue strength of a sprayed specimen without fusing treatment was much lower than that of a steel substrate. An improvement of fatigue strength has been obtained by well controlled fusing treatment.

View full abstract

By assuming that fracture toughness (crack resistance force R) is expressed by constitutive relation R=R₁(a)·(a^^./a^^.*)^m (a and a^^., crack length and crack velocity, respectively m, rate sensitivity parameter), the fracture behaviour under stable crack extension in a DCB specimen is analyzed for rate-sensitive materials. Analytical results show that an initial crack begins to extend before maximum load (P_max), and the maximum value of R appears after P_max. Alternate stable/unstable crack extension is obtained by introducing periodic disorder in R. The analytical work explains well the serrated P-δ curve which is often observed in delamination tests of laminated carbon composites.

View full abstract

Recently, a bending-type stabilizer for precracked ceramic specimens regulated by JIS was proposed. To investigate the behavior of crack growth in stable fracture of brittle materials, large specimens will be useful because of difficulties in accurate measurement in experiments using small specimens. In this study, therefore, a series of three-point bending tests was carried out using a newly developed large stabilizer and soda-lime glass speciments (100×20mm) with various precrack lengths. Crack extension during stable fracture was measured using video images. Numerical simulations of three-point bending for crack-rate sensitive, perfectly brittle materials were also performed using a rate constitutive equation. It was found that stable fracture was observed in the specimens with crack length larger than x₀=0.36 (x₀-a₀/W, a₀: initial crack length, W: specimen width). In stable fracture, the crack grows gradually and the stress intensity factor, K_t, increases even after a maximum load is observed. This newly observed experimental finding was also confirmed qualitatively by numerical simulations.

View full abstract

Scatter of cleavage fracture toughness K_Jc in the transition region can often be quite extensive. Approaches based on two- or three-parameter Weibull distribution have been used to predict the lower-bound fracture toughness from scattered data sets. However, for small data sets, an approach is often unavailable to estimate the lower-bound value with a high degree of confidence because of difficulty of determining the correct Weibull slope m. In this paper, based on a large number of experimental data, an empirical relationship among Weibull slope m, yield strength σ_ys, average value of fracture toughness K_Jc(avg) and specimen thickness B, has been developed as m=6.98+3.74 log[B/(K_Jc(avg)/σ_ys)²]. When the averaged slope value was determined and the lower-bound fracture toughness K_Jc0.03 was predicted using this relationship for A508 C1.3 and A533B steels tested in JSPS/MPC RRT, a more appropriate result was obtained than those obtained using the previous 2-parameter Weibull method and 3-parameter Weibull method in ASTM Draft with m=4 and K_min=20 MPa√(m).

View full abstract

The present paper shows a method of high-speed holographic microscopy which can take three successive photographs of a crack rapidly propagating in PMMA. The crack speed is of the order of 200m/sec. The optical system of the high-speed holographic microscopy consists of three Q-switched ruby lasers as light sources. By varying the angle of incidence of the reference beams, the optical system records the crack as three holograms on one photographic plate, that is angle multiplexing holography. The time interval between two of the three holography recordings is about 1 μsec or longer, and the spatial resolution of the reconstructed crack images is about 114 lines/mm. From the photographs, one can measure the running speed and the opening displacement of the crack and, know the crack propagation in the time scale of microseconds.

View full abstract

Welding residual stress in a multi-pass butt-welded pipe joint was calculated for various welding sequences by thermal elastic plastic analysis using the finite element method. The 5600-mm-diameter, 50-mm-thick pipe joint was prepared for both side grooves. The distribution of residual stress depends on differences in the sequence of welding passes. The optimal sequence considering practical use is determined such that the axial residual stress on the inner surface in the heat affected zone becomes a minimum, because the enviroment at this point for the stress corrosion cracking is severer. The production mechanism of welding residual stress at the pipe joint was studied in detail using a simplified method proposed in this paper. This method of estimating residual stress in pipe joints was validated by comparison with the computed results.

View full abstract

In bonded dissimilar materials subjected to cooling and heating, the stress singularity frequently occurs due to discontinuity of materials at the interface. Thermal stresses near the apex are represented by Kr_P-1, where P-1 is the order of the stress singularity and K indicates the intensity of the stress field. The stress singularity occurs when a root P of an eigenequation obtained from the boundary conditions of stress and displacement fields exists in the range of 0<Re(P)<1 and disappears when it exists in that of Re(P)>1. Also, in a previous paper, it was theoretically clarified that combinations of materials with wedge angle and mechanical properties, in which K becomes zero regardless of angle coordinate θ, exist, and a method for disappearing the singularity from K was suggested. After that, in case where Poisson's ratio ν₁ and ν₂ of each material is 0.3, the relationship yielding K=0 between the bonded wedge angle and the mechanical properties was shown on the k₁₂ (stiffness ratio)-φ1 (wedge angle of material 1) and φ1+φ2 (total wedge angles of materials 1 and 2)-φ1 planes, and it was examined how conditions under which K=0 vary with the bonded geometry and Young's modulus. In the present paper, the relation between conditions under which K=0 and poisson's ratio (ν₁=ν₂ and ν₁≠ν₂) is studied in detail, and the influence of poisson's ratio on the conditions under which K=0 is shown on the k₁₂-φ1 and φ1+φ2-φ1 planes.

View full abstract

The X-ray diffraction method was used to measure the residual stress in the compound and diffusion layers of gas-soft nitrided carbon steel (JIS S45C) and low-alloy steel (JIS SCM435). The X-ray elastic constant of the compound layer made of ε phase iron nitride was determined experimentally. The residual stress was tension on the surface and gradually changed to compression in the compound layer. The residual stress was compression in the diffusion layer near the interface with the compound layer. The compression zone extended about 0.7mm for S45C and 0.35mm for SCM435 from the surface. In this compression zone, an increase in Vickers hardness was observed. The tensile fracture strength of the compound layer measured by the X-ray method was about 600MPa for S45C, and 700MPa for SCM435.

View full abstract

X-ray stress measurement of a composite material using a diffraction ring detected by a two dimensional detector was studied. The merit of the method is the reduction of time required for the X-ray measurement. A theoretical background of the method for determining phase stresses, macro- and microstresses from one diffraction ring was given in the first section of the paper. An experiment on the applicability of the method to a damaged layer of ferritic and austenitic dualphase stainless steel after grinding was also described. The depth profiles of residual macro- and microstresses were obtained from phase stresses. Plastic strains, which can be measured using the Eshelby/Mori-Tanaka model, were also shown which results of the residual phase stresses were discussed. It was found that residual stresses in the surface layer which built up after grinding are mainly under the control of plastic strain rather than macrostress.

View full abstract

A microscopic Raman apparatus using an ultraviolet laser has been developed for measurement of stress in semiconductor devices. A minimum laser-spot diameter of 0.4μm is obtained at 300-nm wavelengh. This is half the size of the spot diameter obtained with a conventional Raman apparatus using a visible laser. Furthermore, coupled with the high-precision scanning system, we have succeeded to measure stress distributions with 0.05μm horizontal resolution. The penetration depth of the ultraviolet laser into silicon is about 0.07μm, which is a tenth of that of the visible laser. This enables to directly measure stress fields as thin as 0.1μm from the surface of the silicon substrates of semiconductor devices.

View full abstract

The optical method of caustics is extended to the bending problem of circular plates with clamped edges under uniform loading on their lateral surfaces. The complete theory of caustics for this problem including predictions of the initial curve, caustic shape and properties, as well as a corresponding measurement method, is presented and investigated. Using this method, it is possible to determine the existing lateral load of the plate, flexural rigidity of the plate, and the bending moments in the plate. Two kinds of plate (polycarbonate, stainless) are used for the research. It is shown that the experimental results for the bending moment are in good agreement with theoretital values, and measured values for lateral load agree well with values obtained using the precision pressure gauges.

View full abstract

A thin plate is heated locally by a laser beam and simultaneously, a large thermal elastic plastic stress-strain state is generated in the heated, thin surface layer. This state can lead to plate bending upon cooling. The bending configuration depends mostly on the plate geometry, and heating and cooling process. We presents some experimental results of bending the plate in V-, U- and concave shapes. A model of this process is constructed and the theoretical predictions are carried out based on FEM analysis using eleastic-plastic theory. Bending angles obtained experimentally and theoretically are in good agreement.

View full abstract

The internal contact surface of polycrystalline metals, as well as their free surface, is roughened during compressive plastic deformation. The effect of the combination of metals on the roughening of the contact surface, however, has hardly been studied. Roughening of the internal contact surface is important not only for the study of friction or bonding during plastic working, but also for the basic study of deformation of polycrystalline metals. In the present study, a rectangular block specimen and another block specimen comprised of different metals are prepared and pressed together. The roughening of the internal contact surface is measured using a stylus instrument. The combinations of copper, aluminium and lead are studied. The microscopic change of surface shape is also observed using a scanning electron microscope. It is found that the roughness of the contact surface of copper is dependent on the hardness of the contacting metals.

View full abstract

In plastic deformation of damaged materials, the influence of distribution of voids on macroscopic mechanical properties is an important issue. As the distribution of voids in damaged materials is heterogeneous, we have to evaluate the local damage state. In this paper, we propose a method of using stereology and voronoi tessellation to evaluate the local damage state and examine the relation between the distribution of voids and the local deformation of perforated sheets with randomly distributed holes as plane models of damaged materials. It is shown that the method of using stereology and voronoi tessellation is valid for evaluating the local deformation of damaged materials.

View full abstract

A model of latent hardening due to dislocation interaction is proposed, which can express strain hardening properties of fcc single crystal in various tensile directions. Tensile properties of an aluminum bicrystal composed of crystals initially oriented for single slip and <100> tensile direction are analyzed on the basis of crystal plasticity theory incorporated with the proposed hardening model. In numerical results, several slip systems are activated near the grain boundary of <100> grain due to the local stress which arises from plastic incompatibility. Flow stress of the bicrystal is larger than the average of component single crystals due to such additional slips, but this difference is smaller than the experimental observation. As deformation proceeds, slip state of <100> grain changes from 8 symmetric slips to a smaller number of slips. This tendency is stronger for the latent hardening model than for the isotropic model.

View full abstract

Self-diffusion of atoms along a grain boundary often brings about fracture in an aluminum conductor an LSI (Large Scale Integrated circuit), which is called "stress migration". As the conductor is subjected to tensile hydrostatic stress at high temperature due to thermal mismatch, it is important to understand the effect of hydrostatic loading on the diffusion. However, this effect has not been examined yet since experiments under tensile hydrostatic stress are extremely difficult. In this study, the effect is investigated on the basis of a molecular dynamics simulation using the Nose-Hoover method and the Parrinello-Rahman one. The simulation enabled us to observe the motion of atoms under constant stress and temperature. The results obtained are summarized as follows. (1)The simulated coefficient of lattice diffusion under stress-free condition agrees very well with the experimental ones. This indicates the validity of the present simulation. (2)The motion of atoms is accelerated by tensile hydrostatic stress in the region of about 5 atomic layers near the grain boundary. (3)Tensile hydrostatic stress accelerates the diffusion along the grain boundary, while compressive stress suppresses it. (4)The coefficient is proportional to exp [-(ΔE)/{k(T/T_m)}] regardless of the magnitude of hydrostatic stress where T and T_m are the temperature and the melting temperature of the grain boundary under the corresponding hydrostatic stress, respectively.

View full abstract

The adaptively shifted integration technique is applied to the finite-element collapse analysis of beams and framed structures composed of brittle materials. Linear Timoshenko beam elements are employed for the analysis. Plastic deformation and cracking on a cross section are treated using the layered approach. Points at which plastic deformation and cracking occur along the element length are accurately determined by the adaptive shifting of a numerically integrated section in each element. Numerical studies for beams and frames with and without ductile reinforcements are carried out to demonstrate the validity of the proposed numerical technique.

View full abstract

A method for simulating microscopic shrinkage behavior of powder particles in sintering of a compact is proposed. In this method, the power particles are modelled as many circular elements undergoing viscoplastic deformation due to surface tension during the sintering, and the microscopic shrinkage is calculated by equilibrating forces acting on the elements. The change of the neck shapes in the boundaries between the elements due to the shrinkage is taken into consideration. Plane-strain shrinkage in sintering is calculated under regular and irregular dispositions of power particles. In the regular disposition of powers having the same diameter, the obtained shrinkage beravior is compared with the experimental one using glass rods and the calculated one by the viscoplastic finite element simulation. It is shown from the simulation of irregular disposition that the densification due to the sintering is heightened by mixing powders having different diameters.

View full abstract

Ultrasonic wave velocities in a plastically deformed medium are known to depend upon its microstructural properties. Therefore, one of the authors proposed a theoretical model of an ultrasonic nondestructive method in order to evaluate plastically deformed states. We obtained good agreement between the numerical and experimental results for an aluminum alloy specimen subjected to uniaxial tension and reported the influence of annealing upon plastic anisotropy and the correlation between the Lankford's value and the shape of yield surface of the specimen. In order to verify the proposed theory we simulated using the internal state variables of an anisotropic distortional yield model longitudinal wave velocities under combined stress states and the subsequent yield surfaces, which were determined to agree well with the experimental results of the longitudinal and transverse wave velocity changes under uniaxial tension test, and then compared the simulated results with experimental results of the longitudinal wave velocity and the yield surface of the aluminum alloy tested in combinations of tension or compression and torsion.

View full abstract

As their in communication and electronic industry expands, it is important to study the shielding effectiveness (SE) of carbon fiber reinforced thermo plastics (CFRTP) against electro magnetic (EM) radiation. In this work, the SE of CFRTP was investigated experimentally in an electronic shielding room using carbon fiber (CF) and glass fiber (GF) reinforced thermo plastics. The resin materials used were polycarbonate (PC), polypropylene (PP), polyetherimide (PEI), polymethylmethacrylate (PMMA) and polyamide (PA). Experimental results indicated that CF was a good EM shielding material. The shielding effectiveness of the composite was found to increase as the number of laminated layers of CFRTP was increased. Other SE characteristics depended on the material used for the resin matrix, the distance of the antennas and the noise frequency band.

View full abstract

This study deals with an economic analysis of structural systems on using a reliability-based design method. Although the expected total cost which is composed of the initial cost and the cost of failure for the structural system has been investigated, it is crucial to consider the annual prediction cost with life cycle cost (LCC) during the life duration. It is also important to evaluate the cost not only on structural failure but also on designer's faults. Furthermore, it is necessary that the cost of the reliability test is taken into account in determining the expected total cost because uncertainties of parameters are strongly dependent upon the quality control or material test. In this paper, a new expected total cost concept which includes the cost mentioned above is proposed. In the proposed formulation, an optimization problem for economic evaluation during the life duration with the capital rate of interest and rate of price fluctuations can be solved by applying the present currency and annual prediction concepts. For a comparative evaluation, from the numerical analyses of a truss bridge structure, it is pointed out that not only optimization by present value but also that by annual value is useful for decision making in practical design.

View full abstract

In this paper we present a numerical analysis method for the boundary design of continua in order to control the displacement mode to a prescribed mode, which is known as the homology design. As an objective functional, we introduce a squared displacement error norm on the prescribed boundaries. Using the Lagrange multiplier method, a shape identification problem subject to a volume constraint is formulated. The shape gradient function derived using the material derivative method and the adjoint variable method is used in the traction method. With the traction method, the domain variation that minimizes the objective functional is numerically and iteratively determined. The state equation and the adjoint equation are solved using a general-purpose FEM code to evaluate the objective functional and the shape gradient function. The calculated results of 2D and 3D problems show the effectiveness and practical utility of the proposed method to control the displacement mode.

View full abstract

The effect of mechanical properties and deformation of surrounding tissues on vascular resistance was investigated experimentally and theoretically for canine coronary arteries. In the experiment the hydraulic resistance was measured in epicardial coronary artery segments for two transmural pressures and four stretching conditions, i.e. 0% stretch, and 10% parallel, transverse, and biaxial stretch with respect to the vessel axis. The experimental results showed that the stretch in the direction of the vessel axis increased the resistance significantly for low transmural pressure. Finite element analyses were also performed. In the models the vessel was surrounded by a lipid layer and myocardial tissue. The simulation results showed that the existence of a lipid layer reduced both the resistance and the resistance change due to deformation of the myocardium. It was also shown that the resistance change due to myocardial deformation was smaller in the vessel near the surface compared to at the center of the myocardium.

View full abstract

Observing the human upper limb from the point of view of mechanics, every degree of freedom of motion is actuated and controlled by more than two muscles. Many biomechanical researchers have investigated the principles of the complex workings of human muscles using a mechanomathematical model, to obtain more knowledge about the musculoskeletal system control. In this paper we describe a method of numerical analysis of muscle force in a human upper limb during flexion of the elbow joint. Because no muscle force can be calculated from only the equilibrium equations of force or moment, the optimization method using Lagrange multipliers which was proposed by Raikova was applied to this analysis. Muscles were modeled by a straight line from the origin to the insertion. Parameters in the object function were selected to express the muscle properties acting only in the direction of the contraction. Therefore, the synergist and the antagonist could be separated analytically. Moreover, the optimizing point in the wide region of the origin of the muscle depended on the joint angle of the elbow. The results obtained from this analysis were confirmed by experiments using electromyography.

View full abstract