Smooth specimens of porous silicon carbide with porosity 37% showed nonlinear relations between load and strains under four-point bending. The relation between the true stress and the strain determined from non-linear relations was nonlinear for the tension side, while linear for the compression side. The bending strength of smooth specimens followed a two-parameter Weibull distribution. Single-edge notched specimens with the notch depth ranging from 0.1 to 2.8 mm were fractured under four-point bending. The fracture of specimens with a notch depth of 0.1 mm did not start from the notch ; the fracture strength of the notched specimen was equal to the smooth specimen. The notch depth of 0.1 mm is roughly equal to the size of silicon carbide particles in porous ceramics. The load versus strain record showed nonlinearity before the maximum load, suggesting the R-curve behavior of porous ceramics. The R-curve was constructed from the compliance change of the specimen, and was used for determining the maximum load point in bending tests. The intrinsic crack model was proposed to predict the effect of the notch size on the fracture toughness of porous ceramics. Fractographic observations showed the fracture path along the binder phase between silicon particles.

抄録全体を表示

An up-shock method was proposed as a new technique for estimating the thermal shock resistance of ceramics. The method uses a disk-shaped specimen which is heated by infrared ray and by a constant heat flux at the central area. This technique enables an evaluation of thermal shock strength, Rc, directly from the power charge, despite the fact that Rc consists of some of the properties of the tested material. In this paper, the thermal shock resistance of Al₂O₃ and Si₃N₄ are estimated by both the up-shock method and the improved water-quench method which was presented earlier. The results of the infrared radiation heating method show that the thermal shock strength of Si₃N₄ was 7 times as high as Al₂O₃'s one. In the water quench method the heat transfer coefficient of water was indispensable for the quantitative estimation. In this study we show that the up-shock method is more suitable for the evaluation of the thermal shock resistance for ceramics.

抄録全体を表示

Microindentation tests on ceramics (Si-SiC, SiC, Al₂O₃ and Si₃N₄) immersed in boiling 95 wt% sulfuric acid for 100 or 1000 hours were performed to evaluate the mechanical properties of their corroded layers. The thickness (T) of corroded layer on ceramics was evaluated in terms of characteristic depth (d) which was determined from the point of inflection on the Depth-Load/Depth curve by the microindentation test. The relationship between T and d was found to be given as T ≒ 10d. Finite element analyses were performed to validate the relationship and to clarify the effects of mechanical properties of corroded layer and the indenter tip radius on the relationship as well. The mechanical properties [Young's modulus (E_f) and yield stress (σ_yf)] of corroded layers of Al₂O₃ and Si₃N₄ were identified by fitting the predicted Depth-Load/Depth curve to the experimental data.

抄録全体を表示

Amoung advanced composite materials, alumina fiber (ALF)/epoxy composites are selected for the structural components of superconducting magnets because of their low thermal conductivity in addition to their higher specific strength and modulus. Since laminate structures are used as the load support system for superconducting magnets, the evaluation of interlaminar strength both under static and fatigue loadings at cryogenic temperature is essential. In the present study, mode II interlaminar fracture toughness and delamination fatigue crack growth behavior in liquid nitorogen (77K) were investigated with unidirectional ALF/epoxy laminates. Tests were carried out using end notched flexure specimens with special loading apparatus. The fracture toughness values at 77K were higher than those in laboratory air at room temperature (RT). The fracture mechanism under fatigue loading at 77K was controlled by the maximum stress, and was completely different from that at RT. Then, the increase of the fatigue crack growth resistance at 77 K from that at RT was observed only under stress ratio, R=0.1. The fractographic observation showed that the resin fracture was dominant for the fracture surfaces of fatigue fracture, suggesting that the microscopic mechanism was different from that at RT where the interfacial fracture was dominant.

抄録全体を表示

Laminate structures of alumina fiber (ALF)/epoxy composites have been applied to the structural components of superconducting magnets because of their low thermal conductivity and higher specific strength and modulus. Since these structures are used as the load support system for superconducting magnets at cryogenic temperature, the evaluation of interlaminar strength both under static and fatigue loadings is important. In the present study, mode I interlaminar fracture toughness and delamination fatigue crack growth behavior were investigated with unidirectional ALF/epoxy laminates in liquid nitrogen (77 K). Tests were carried out using double cantilever beam specimens. Both the fracture toughness values and fatigue thresholds at 77 K were higher than those in laboratory air (RT). The analysis of the stress-ratio dependency showed that the fatigue mechanism in liquid nitrogen was only controlled by the maximum stress. The difference of the fracture mechanism owing to the test temperature was discussed on the bases of fractographic observation.

抄録全体を表示

The cross section of main frame structure is preferable to be a cylinder for plants and animals in order to adapt to their environment. There are some exceptionals. One of those is a hemp palm branch. It has an up side-down isosceles triangular shape. Since the hemp palm branch has unusually large leaves and a relatively slender branch stem as compared with other trees, the branches must be subjected a large bending moment. A triangular cross section must be a solution for that problem, which can reduce the induced stress about 30% under a bending stress. Furthermore, the branch is a composite material reinforced by strong fibers. It also constructs a functionally graded structure on fiber distribution. This structure provides a higher strength in the critical loded region than the induced stress there.

抄録全体を表示

This study is concerned with the reliability analysis of the composite plate with a hole. The reliability is evaluated by modeling the plate as a series system consisting of each ply failure based on a first ply failure criterion. The mode reliability is obtained by the first order reliability method. where material strengths and applied loads are treated as random variables. The reliability is evaluated at the most critical point of each ply. The strength distribution is evaluated by the finite element analysis by adopting Tsai-Wu criterion. The system reliability is approximated by Ditlevsen's bounds. Through numerical calculations, the effect of variations of the random variables is discussed. Then, the reliability-based design is compared with the deterministic strength maximization design. Finally, it is shown that reliability is an important measure for structural safety.

抄録全体を表示

Compression strength and stiffness of laminated CFRP structural components are severely degraded by delaminations. The delaminations are unfortunately invisible or difficult to inspect by naked eyes. This brings significance of development of smart structures for detection of delaminations of CFRP. In the present study, an electric resistance change method is adapted for identifications of delamination crack position and size of CFRP laminates, and possibility of identification of delamination cracks is experimentally investigated by using cross ply laminates. 3-point-bending test are conducted using specimens with multiple electrodes to generate delamination cracks. Firstly, response surface to predict electric resistance change are constructed from measured electric resistance change. Secondary, design of experiments are conducted for inverse problem to predict delamination crack position and size from electric potential changes. Finally, response surface to predict delamination crack position and size are constructed. Using the response surface method with design of experiments, delamination crack position and size are predicted within practical precision without consideration of fracture mechanism.

抄録全体を表示

In this study, a neural network is applied to optimization problems of material compositions for a functionally graded beam with arbitrarily distributed and continuously varied material properties due to a partial heat supply. Using the analytical procedure of a laminated composite beam model, the analytical two-dimensional transient temperature solution for the beam is derived approximately. Furthermore, the thermal stress component is formulated under the mechanical condition as being traction-free and making use of the elementary beam theory. As a numerical example, the finitely beam composed of zirconium oxide and titanium alloy is considered. And, as the optimization problem of minimizing the thermal stress distribution, the numerical calculations are carried out making use of neural network, and the optimum material composition is determined at arbitrary heat area and heat transfer coefficient. Furthermore, the results obtained by neural network and ordinary nonlinear programming method are compared.

抄録全体を表示

This paper is concerned with the theoretical treatment of transient thermal stress problem involving an angle ply laminated strip consisting of an oblique pile of layers having orthotropic material properties due to nonuniform heat supply in the width direction. We obtain the exact solution for the two-dimensional temperature change in a transient state, and thermal stresses of a simple supported strip under the state of generalized plane deformation. As an example, numerical calculations are carried out for a 2-layered angle-ply laminated strip, and the numerical results for the temperature change, the displacement and the stress distributions are shown in figures. Furthemore, the influence of heat function on temperature change and thermal stress distributions are investigated.

抄録全体を表示

Frequency dependence of group velocity is measured for longitudinal wave propagating along length of a steel wire. In a long ranged inspection of steel wires by using a broadband pulse, flaw detectability is greatly lowered by the waveform distortion due to geometric dispersion. To improve the flaw detectability, effective compensation for the dispersion should be carried out, which requires an accurate measurement of the dispersion curve. In this study, we suggest to use digital frequency filters to evaluate the frequency dependence of group velocity, which can then be translated into the dispersion curve of phase velocity. The group velocity can be determined as the velocity of the envelope of narrowband wavelet composed by filtering. Accuracy of the method was first examined by using the waveform simulated with given dispersion curve. We then applied this method to the echoes from artificial flaws and found that the signal to noise ratio is improved.

抄録全体を表示

Ultrasonic reflection coefficient was measured to evaluate the distribution of acoustic impedance in cortical bone of bovine tibia. The reflection coefficient, which is defined as the amplitude ratio of reflected wave to incident wave, depends on difference in acoustic impedance at the interface. In this study, we measured the amplitude of reflected wave from surface of cortical bone sample sunk in saline solution. Prior to the measurement, spot size of ultrasonic beam at the focal point was measured for preparing the flattened area on the bone surface. To calculate the intensity of incident wave, the standard sample, whose acoustic impedance is known, is used. We examined suitability of several materials for a standard sample. Accuracy of this method was evaluated by comparing the results with the product of density and bulk wave velocity. We revealed that the acoustic impedance of the cortical bone can be measured with sufficient accuracy by preparing the flat spot on the bone surface.

抄録全体を表示

The method of continuously distributed dislocations model is applied to the elastic contact problem near an elliptic rigid inclusion. In this method, the rigid inclusion is replaced with the continuous array of edge dislocations along its interface and the boundary conditions are expressed as singular integral equations of the second kind in which the contact pressure distribution and the dislocation density are unknown. The dislocation density functions are approximated as the Fourier expansion series with finite number of terms and the singular integral equations are made to be satisfied at the selected collocation points, which reduce them into a set of linear algebraic equations. Some typical cases are solved numerically using the present method and the distributions of the contact pressure and the stress around the inclusion are calculated. It has been clarified that the solution converges rapidly as the number of the collocation points increases.

抄録全体を表示

It is well known that multi-singularities would appear at an interface edge. It is useful to determine all stress singular orders for the strength evaluation of such a dissimilar material. However, the multiple singular order have been determined mainly based on theoretical analysis. Considering the engineering needs for the complicated problems which are not easy to analyze theoretically, it is strongly expected to develop a numerical method which can determine all stress singularities. This paper studied the nature of the singular stress fields, and then proposed a numerical method to determine all stress singular orders (including the oscillatory singularity), based on the numerical results obtained by FEM or BEM. From the results of the numerical analysis, it is found that the multiple stress singular orders obtained by the present method agree well to the theoretical solutions. Through these examples, it is proved that this method is very efficient and useful.

抄録全体を表示

The static tensile and fracture toughness properties of Ni-Ti-Nb shape memory alloy (SMA) were examined at room temperature and 561 K. Its chemical composition is 51 wt.% Ni, 38 wt.% Ti, and 11 wt.% Nb. The shape recovery temperature is around 350 K. The tensile strength and the reduction of area at 561 K are slightly higher than those at 293 K. And the 0.2% proof stress, tensile strength, and reduction of area are constant with increasing loading rate below 10000 MPa/s. With increasing temperature K_max slightly decrease below 413K, but slightly increases below 561 K, and K_Q slightly decreases below 561 K. With increasing loading rate below 10000 MPa√(m) K_max is constant, but K_Q increases. The fracture toughness of this SMA is lower than that of general structural elements, because, micro-cracks initiate from the Ti-rich precipitates at the bottom of a dimple. K (J_IC) was obtained as 51.7 MPa√(m), so the material crack was evaluated conservatively using at K (J_IC). In conclusion Ni-Ti-Nb SMA has sufficient mechanical and fracture toughness strength for structural elements at elevated temperatures.

抄録全体を表示

Silica-particulate-filled epoxide resins are used as encapsulant materials in semiconductor packaging. In the present paper. the effects of temperature on the fracture toughness of two types of epoxide resin containing 70 wt.% silica particles have been studied. The bridge indentation method was employed in pre-cracking of the specimens, and three point bending tests were carried out to obtain fracture toughness. The fracture toughness of both resins was almost constant at the temperatures from room temperature to the glass transition temperature of the resin. The fracture toughness decreased remarkably above the glass transition temperature. Fracture surface observation by scanning electron microscope shows that epoxide resin matrix predominates in the fracture surface at temperatures above the glass transition temperature while cracking of silica particles is seen at room temperature. This suggests that differences in fracture toughness due to temperature are attributable to differences in crack propagation of the resin due to softening of the resin matrix above the glass transition temperature.

抄録全体を表示

Atom migration induced stress in an LSI conductor is analyzed on the basis of diffusion along grain boundary (GB) and interface. The focus is put on the network of GB and interface, which strongly affect the atom flux near the cavity. The numerical analyses are conducted for the conductors with polycrystalline network, bamboo structure and near-bamboo structure. Results obtained are as follows. (i) In a conductor with polycrystalline, the atom migration along GBs is activated near a cavity while there is little flow in a conductor without the cavity. This is due to the stress distribution, which is introduced by the cavity. The cavity grows by the atom transport. (ii) Interface difiusion between the conductor and the surrounded passivation activates the atom migration along GBs inside of a conductor for all simulations. This implies that the cavity grows under the interaction between GB diffusion and interface one. (iii) Growth rate of a cavity in a conductor with the near-bamboo structure, which contains a few triple junctions of GB, is faster than that in a one with the bamboo structure. This shortens the life of conductor, which is correspondent with experimental observation reported before.

抄録全体を表示

Long-term fatigue tests have been conducted on a high strength steel, SUJ2, in order to clarify the fatigue behaviour in long life region, more than 10⁷ cycles. The fatigue failure at the stress level under the ordinary fatigue limit occurred in long life region. Fracture mechanism in this region is fish-eye type, i. e. inner defect induced failure, while surface slip type fracture is found in short fatigue life region. At the center of fish-eye non-metallic inclusion is recognized, which is surrounded by an indefinite and grainy region. However, the fish-eye without grainy region was also observed exceptionally. Therefore, a fish-eye is divided usually into three parts, i. e. inclusion, grainy region and inner crack. Most of the inclusions generating the fish-eye were located within 0.1mm from the specimen surface. The sizes of inclusions calculated by the √(area) method are almost constant to the stress level and the number of cycles. The initial stress intensity factor range, ΔK_ini, at which crack was generated from the inclusion, is almost the same with increasing fatigue life, i, e. 2.3∼3.5 MPa √(m). On the other hand, the sizes of fish-eyes calculated by the √(area) method are scattered to the stress level and the number of cycles. Based on the results, a model of the fatigue failure at the stress level under the fatigue limit is proposed.

抄録全体を表示

The aim of this work is to investigate the effects of nitrogen ion implantation on fatigue crack initiation, early crack growth behavior and fatigue life for high tensile strength steel (HT60). Threepoint bending fatigue tests were carried out using an electro-hydraulic fatigue testing machine. The fatigue specimens used were edge notched specimen having a notch radius of 0.25 and 2 mm and smooth specimen. The fatigue crack initiation and the early crack growth were observed by using an optical and scanning electron microscopes adopting a plastic replica method. The main results obtained are summarized as follows. (1) The micro hardness of nitrogen ion implanted specimen was about 1.8 times greater than of non-implanted specimen. (2) The fatigue life was increased by nitrogen ion implantation when the maximum stress at the notch root was less than the static fracture stress of nitrogen ion implanted layer. (3) Comparing ion implanted specimen with non-implanted specimen, the fatigue crack initiation was delayed by nitrogen ion implantation. (4) The early crack growth behavior was sensitive to the microstructure so that a zig-zag pattern was observed at both sides of notched specimen on nitrogen ion implanted specimen.

抄録全体を表示

Rotating bending fatigue tests were carried out for an aluminum oxide dispersion strengthening copper alloy in order to investigate the notch sensitivity based on linear notch mechanics. The fatigue limit for crack propagation decreased with decrease in notch radius, and the sensitivity for crack initiation was high. These facts were discussed based on properties of the crack initiation and propagation.

抄録全体を表示

In this study, low cycle fatigue life of squeeze cast aluminium alloy (AC4CH) was investigated. Fatigue life in AC4CH was shorter than that in its matrix material (A6061-T6) by approximately onethird. To clarify the cause of decrease in fatigue life, three controlling factors for low cycle fatigue life were compared. The results obtained are as follows : (1) Fracture in AC4CH starts from a high density zone of eutectic silicon particles that exists on the neighboring surface ; thus, the crack initiation length in AC4CH is longer than that in A6061-T6. (2) Fracture ductility in AC4CH is small and preexisting cracks originate from other eutectic silicon particles ; thus, the crack propagation limit length in AC4CH is shorter than that in A6061-T6. (3) The crack propagates in the interface between eutectic silicon particles and matrix ; thus, the crack growth rate in AC4CH is faster than that in A6061-T6.

抄録全体を表示

The purpose of this study is simultaneously to obtain plural pole figures using imaging plate (IP) which is a a two-dimensional X-ray area detector. Since more than one Debye-Scherrer rings are simultaneously recorded on an IP, plural pole figures are drawn from one X-ray measurement. Diffraction data recorded on IP are also detected as digital data, so that it is easy to analyze them by a computer. In the present paper. this method was applied to a textured dual phase stainless steel sheet which consists of coarse grains. Results are summarized as follows : (1) Plural pole figures are simultaneously obtained. (2) By applying a computer smoothing process to diffraction data containing spotty Debye-Scherrer rings, pole figures which elearly show the state of preferred orientation are obtained.

抄録全体を表示

In this paper, a complete pole figure measurement was examined by only the back-reflection method using an imaging plate (IP). In the present method, complete pole figures are obtained by only the back-reflection method without use of the transmission method. In particular, complete pole figures are theoretically obtained when its diffraction angle 2θ is 90°. In cases of other 2θ, complete pole figures are obtained by measuring two imperfect pole figures and combining them. In the measurement of imperfect pole figures, plural pole figures were simultaneously obtained by the two dimensional measurement ability of IP. In addition, a correction of diffraction intensity according to a shift of the pseudo-focusing condition is not necessary like the Schulz zeflection method. Pole density data of complete pole figures were effectively applied to analyze the three dimensional crystal orientation distribution of a textured dual phase stainless steel sheet.

抄録全体を表示

This paper deals with the problem of statistical uncertainties of safety index. Though a second moment safety index is widely used as a safety index for a probabilistic design code, it is impossible for this index to avoid to include intrinsically statistical variations since the random parameters are estimated only from the obtained field data or imperfect information. In this paper, the new measures for statistical uncertainties of safety index which is proposed as a confidence region is around the design point with the confidence level. These confidence regions of safety index are derived from an ellipsoid on super plane whose region corresponds to the value of confidence level. This paper shows that region for safety index can be mapped as the internal region of an ellipsoid on super plane. Therefore the region called as confidence ellipsoid.

抄録全体を表示

Thermal fatigue damage of solder joint is considered to be closely related to microstructural coarsening. In the present study, phase growth model is devised in order to express the microstructural evolution of Sn/Pb eutectic sobder. The model is based on results of systematic microstructural abservation using lap joint type shear specimen proposed by the authors. And the model is also derived from the vacancy diffusion theory in phase boundary. The phase growth is characterized by parameter S which is defined as average phase size to the 4 th power d⁴ and increases approximately proportional to time and total equivalent creep strain. Furthermore, the model is applied to evaluation of phase growth process in the solder joint of the specimen under thermal cyclic loading which realizes approximately thermal deformation state of printed circuit board (PCB). Consequently, it is clarified that phase growth proceeds at any temperature to which PCB is subjected and is mainly induced by strain rather than thermal equilibrium. And parameter S increases corresponding to total equivalent creep strain. This suggests that S would be evaluation parameter for thermal fatigue damage of actual solder joint.

抄録全体を表示