Conventional finite element method is continually used for the flaw evaluation of pipe structures to investigate the fitness-for-service for power plant components, however, it is generally time consuming to make a model of specific crack configuration. The consideration of a propagating surface crack is further accentuated since the crack propagation behavior along the crack front is implicitly affected by the distribution of the crack driving force along the crack front. The authors developed a system to conduct crack propagation analysis by use of the three-dimensional elastic-plastic extended finite element method. It was applied to simulate ductile crack propagation of circumferentially surface cracks in pipe structures and could realize the simultaneous calculation of the J-integral and the consequent ductile crack propagation. Both the crack extension and the possible change of crack shape were evaluated by the developed system.

View full abstract

A plane elastic problem for inhomogeneous and isotropic solids such as functionally graded materials is treated theoretically. An inhomogeneous slab with a Griffith crack subjected to uniformly distributed loading such as internal pressures on the crack surfaces is considered. It is considered that the Griffith crack is located on the middle plane of the slab. It is assumed that inhomogeneous material property such as the shear modulus of elasticity varies in a form of a power of a transverse coordinate, and the material inhomogeneity in the slab is symmetric with respect to the middle plane. The mixed boundary value problem is developed theoretically using a fundamental equation system for such inhomogeneous and isotropic solids derived in our previous paper. Thereafter, considering a variation of inhomogeneity in the shear modulus of elasticity, displacements and stresses in the slab, and a stress intensity factor for the mode I deformation at the tip of a crack are examined numerically. An effect of inhomogeneity in the shear modulus of elasticity on plane elastic behavior and stress intensity factor for the mode I deformation is discussed.

View full abstract

Strain range controlled low-cycle fatigue tests were conducted using ultrasonic method in order to investigate the effect of plastic strain range on the remaining life of austenitic stainless steel SUS316NG before the onset of crack growth in its early stages of fatigue. It was found that the decrease in ultrasonic back-reflection intensity from the surface of the material, caused by the increase in average dislocation density with localized plastic deformation at persistent slip bands (PSBs), starts earlier with increase in the plastic strain range. The amount of decrease in ultrasonic back-reflection before the onset of crack growth increases for larger plastic strain range. The difference in the cumulative plastic strains at the onset of crack growth and at the onset of decrease in the ultrasonic back-reflection remained constant over the range of tested plastic strain. This result can be used to predict the remaining life before the onset of crack growth within the plastic strain range used in this study. In addition, we present and evaluate another method to predict damage evolution involving ultrasound attenuation caused by PSBs.

View full abstract

Ultrasonic impact treatment (UIT), which is a peening method, is usually used as a post-weld treatment in order to improve the fatigue strength of welded joints. In this study, fatigue tests were carried out on web-gusset welded joints treated by UIT and the results were compared with the fatigue lives of as-welded joints in order to examine the effects of UIT on the fatigue lives of welded joints. The fatigue lives of web-gusset welded joints treated by UIT increased to more than ten times those of as-welded joints. The introduction of compressive residual stress, relaxation of stress concentration at a weld toe, and refinement of grains under the weld toes were considered as possible reasons for the improvement in fatigue life caused by UIT. Residual stress near weld toes was measured using the X-ray diffraction method. The stress concentration factor at the weld toes was analyzed using the finite element method (FEM). The grain size under the weld toes was measured using electron backscatter diffraction pattern (EBSD) analysis.

View full abstract

This paper describes an analysis method of the friction torque and weld interface temperature during the friction process for steel friction welding. The joining mechanism model of the friction welding for the wear and seizure stages was constructed from the actual joining phenomena that were obtained by the experiment. The non-steady two-dimensional heat transfer analysis for the friction process was carried out by calculation with FEM code ANSYS. The contact pressure, heat generation quantity, and friction torque during the wear stage were calculated using the coefficient of friction, which was considered as the constant value. The thermal stress was included in the contact pressure. On the other hand, those values during the seizure stage were calculated by introducing the coefficient of seizure, which depended on the seizure temperature. The relationship between the seizure temperature and the relative speed at the weld interface in the seizure stage was determined using the experimental results. In addition, the contact pressure and heat generation quantity, which depended on the relative speed of the weld interface, were solved by taking the friction pressure, the relative speed and the yield strength of the base material into the computational conditions. The calculated friction torque and weld interface temperatures of a low carbon steel joint were equal to the experimental results when friction pressures were 30 and 90 MPa, friction speed was 27.5 s^-1, and weld interface diameter was 12 mm. The calculation results of the initial peak torque and the elapsed time for initial peak torque were also equal to the experimental results under the same conditions. Furthermore, the calculation results of the initial peak torque and the elapsed time for initial peak torque at various friction pressures were equal to the experimental results.

View full abstract

This paper discusses the effects of pore arrangements on the dielectric property of porous low-k and high-k dielectrics. Higher performance large scale integration (LSI) requires lower dielectric constant to decrease line-to-line capacitance. Recently, porous low-k dielectrics are introduced for low-k dielectrics because of its ultra lower dielectric constant. However, their poor mechanical strength causes fractures of porous low-k dielectrics during Chemical Mechanical Polishing (CMP) process. Therefore, it is important to develop porous low-k dielectrics with high mechanical strength and low dielectric constant. On the other hand, porous high-k dielectrics are needed as ferroelectrics. It is also important to understand the dielectric property of porous high-k dielectrics. We studied the dielectric property of porous low-k and high-k dielectrics by finite element method and U^* theory. The index U^* is used to indicate load paths in a structure. We extended U^* theory to the electrostatic field problem and investigated the dielectric property of dielectrics. By using U^* in electrostatic field analysis, the dielectric performance of porous low-k and high-k dielectrics becomes more clear.

View full abstract

The mechanical and electrical properties (electromechanical coupling coefficient, piezoelectric constant and dielectric constant) of lead zirconate titanate (PZT) ceramics are investigated during mechanical static and cyclic loading. There are several failure characteristics which can alter the material properties of PZT ceramics. The elastic constant increases and electrical properties decrease with increasing the applied load. This is due to the internal strain arising from the domain switching. In this case, 90° domain switching occurs anywhere in the samples as the sample is loaded. It is also apparent that electrogenesis occurs several times during cyclic loading to the final fracture. This occurrence is related to the domain switching. The elastic constant and electrical properties can decrease because of crack generation in the PZT ceramics. Moreover, the elastic constant increases with increase of the mechanical load and decreases with decrease of the load. On the contrary, the opposite sense of change of the electrical properties is observed.

View full abstract

This paper describes stochastic analysis on a homogenized elastic property of a unidirectional fiber reinforced composite material considering uncertainty in material and geometrical properties of component materials. The influence of the microscopic uncertainties on the stochastic characteristics of a homogenized elastic property is analyzed with the perturbation-based analysis using the equivalent inclusion method. Especially, influence of a volume fraction variation on the stochastic characteristics is discussed in this paper. Also, as an example of a design considering the influence of a volume fraction variation, a limit of the random variation of the volume fraction of fiber for controlling a random variation of a homogenized elastic property of a composite to be less than random variations of elastic properties of component materials is discussed.

View full abstract

This paper describes stochastic homogenization analysis of a uni-directionally aligned short fiber reinforced composite material. In case of a short fiber reinforced composite material, length or orientation of the short fiber will be a random variable in addition to elastic properties of component materials, volume fraction or cross-sectional shape of fiber. Especially, a stochastic homogenization problem considering the length or fiber orientation variation is analyzed in this paper. For the purpose of this analysis, the first-order perturbation-based stochastic homogenization method is employed. Since the perturbation term with respect to the length or the orientation variation cannot be explicitly obtained in using the homogenization method, a finite differential technique is used for approximation of the perturbation term. From the numerical results, validity, effectiveness and problem of the perturbation-based analysis are discussed. Also, influence of the length variation on a stochastic characteristic of a homogenized elastic property of a short fiber reinforced composite material is discussed.

View full abstract

Carbon nanotube is nature's smallest fiber and predicted to have a range of unusual mechanical and electrical properties. One possible route to harnessing these properties for applications would be to incorporate nanotubes in a composite material. Here, we report the mechanical properties of multi-walled carbon nanotube (MWCNT) reinforced alumina composites made with a pristine MWCNT and an acid-treated version that have nanoscale defects on their surfaces from an acid treatment. It was demonstrated that surface modification of the MWCNT is effective in improvement of bending strength and fracture toughness of the MWCNT-reinforced alumina composites. On the basis of the results, we also prepared three sets of the acid-treated MWCNT-reinforced alumina composites having different sintering additives, in order to investigate the effects of sintering additives on their microstructures and mechanical properties. Mechanical properties of the composites were dependent mostly on the type of sintering additives and amount of MWCNT. The 0.9 vol.% acid-treated MWCNT-reinforced alumina composites with MgO sintering additive gave the highest bending strength (689.6 ± 29.1 MPa) and fracture toughness (5.90 ± 0.27 MPa·m^1/2), respectively.

View full abstract

Recent MEMS (micro electro mechanical system) fabrication techniques have made it possible to produce painless microneedles precisely enough to be inserted into epidermis layer penetrating the stratum corneum of human skin. This paper presents a testing procedure to evaluate the insertion characteristics of microneedle array using cultured human skin considering the tension and the curvature. First, the biaxial strain applied to the cultured human skin was measured by optical technique with image processing. It was found that almost constant strain could be successfully given within a certain area and that error factors in the experiment except the thickness variation of the cultured skin were negligible. Next, using a microneedle square array for brain machine interface (BMI) application, the effects of biaxial tension and the curvature on insertion characteristics were discussed. Within the above mentioned area with high strain, the needles were successfully inserted.

View full abstract