Recent development of the local approach to fracture based continuum damage mechanics and finite element method together with the related numerical problems is reviewed. After brief description of the concept, procedure and applicability of the local approach, the essential features and the causes of the mesh-dependence of numerical results are discussed. As regularization methods to avoid or to improve the mesh-dependence in time-independent materials, schemes based on localization limiters and nonlocal damage theory are discussed in some detail. The mesh-dependence problem in time-dependent damage, i.e., creep damage, are also discussed.
Viscosity of aqueous cesium chloride (CsCl) solution was measured in the ranges of 0.1-5.0molkg-1 and 0.1-375MPa at 283.2 and 323.2K. The Jones-Dole B coefficient of CsCl increased with increasing pressure and then decreased passing over a maximum as well as previously observed one at 298.2K. The maximum B became more remarkable with decreasing temperature from 323.2K to 283.2K. These phenomena were compared with the dielectric friction theory supposing solvent water as a continuum fluid. Inconsistency of the theory at low pressure suggested an existence of any structure in water. Pressure dependence of B is ascribed to a balance between the water structure and the dielectric friction effect.
The fabrication and processing of an actuator with a TiNi shape memory alloy (SMA) film are discussed and demonstrated. TiNi SMA films were deposited by using sputter deposition techniques, patterned and etched. These films were solution-treated and aging-treated, and then an inverse force layer of chromium was deposited on a SMA film. The frequency response of beam type actuator was measured by the resistance heating. The maximum amplitude was in proportion to inverse force, and was therefore controlled by chromium thickness. The maximum displacement decreased sligtly with an increase of frequency. The minimum displacement suddenly increased at more than 0.1Hz and was constant above 1Hz. The characteristic of the frequency response of the actuator was not decided by chromium thickness but TiNi thickness, because thickness of the latter was fifteen to forty times larger than that of the former. Because the frequency response depended on the heat radiation of an actuator, TiNi thickness was the most important factor to the actuation.
The residual stress in aluminum thin films sputtered on silicon substrates was measured by the X-ray diffraction method and the curvature method. Aluminum thin films have a fiber texture with a fiber axis of  direction perpendicular to the film surface. In the X-ray method, the residual stress in the films was determined from the strain measurements by using Al 222 and 311 diffractions. The residual stress of as-sputtered films was equibiaxial tension. The magnitude remained invariant with film thickness and was nearly equal to the value obtained by the curvature method. On the other hand, the residual stress measured by the two methods were different for the films annealed at temperatures above the sputtering temperature (423K). The stress value measured by the X-ray method decreased with annealing temperature; the value by the curvature method had a maximum for the case of annealing at 473K. Line broadening of X-ray diffraction profiles of as-sputtered films was found to increase with increasing thickness. Line broadening increased by annealing at temperatures above the sputtering temperature.
This paper describes an analysis of the damping properties of matrix hybrid composite by Lagrange's equation. The damping properties of matrix hybrid composite are calculated by using viscoelastic properties of homogeneous laminated composite. Two kinds of glass fiber reinforced laminated composites are employed. One is chopped strand glass fiber mat reinforced composites, and the other is unidirectional glass fiber reinforced composite. Hybrid composites of both samples consist of conventional and flexible matrix resins. This analysis method is shown to be effective in predicting the loss factor of hybrid composite with flexible matrix resin.
In this paper, a method for predicting compressive strength of notched AFRP (Aramid Fiber Reinforced Plastics) is proposed. The compression tests are performed with an end-tab compression method based on the NASA/Boeing standards. In the experimental procedure, both unidirectional and woven cloth AFRP specimens show the hole radius dependency of the compressive strength. It is also found that the failure modes can be classified by two failure criteria. Therefore, each failure criterion is applied to predict the compressive strength for each failure mode. A fracture simulation is carried out using FEM to predict the compressive strength of the notched AFRP. Fiber micro-buckling is represented by variation of the stiffness of the element at the elastic-plastic stage. The Tsai-Hill law is applied as a yielding criterion, and the maximum compressive strain theory is applied for fracture criterion. The load-displacement curve and compressive strength of the unidirectional AFRP are well simulated. On the other hand, it is found that the woven cloth AFRP is simulated only in the initial failure process. It can be concluded that the proposed fiber micro-buckling model is appropriate to explain the complicated buckling mechanism of laminated composites.
An existing FEM program for elastic-plastic problems was modified to the one which can be accommodated to the changing of stiffness matrix and the release of stresses at fractured elements. With this program, the fracturing processes of two different mechanical tests of wood were simulated: the JIS shearing test (JIS Z2114) and the JIS bending test (JIS Z2113). The results are summarized as follows: (1) The shearing strength obtained by the calculation was 8.6MPa. This value agreed with the actual testing data. The fracture initiated and propagated along the grains. This simulated result accorded well with the real fracturing process of the shearing test. (2) The bending strength obtained by the calculation was 127MPa, which was larger than the actual value because the finite element mesh was rough and thus the scale factors for yielding had been estimated larger. The fracture initiated at the bottom of the specimen, grew to the neutral axis and turned its direction to the grain. This simulated result accorded well with the real fracturing process of the bending test.
Mechanical properties and phase structure of yttria-stabilized tetragonal zirconia polycrystals aged in air at 200° and 300°C for 25000h (about 3 years) were investigated. The bending strength and dynamic hardness of the aged samples were reduced after the aging. The remarkable tetragonal-to-monoclinic phase transformation occurred in surface and sub-surface layers of the sample within 100μm from the surface. The monoclinic phase fraction of the aged samples, especially aged at 200°C, was much larger than that of the unaged samples. The bending strength and dynamic hardness decreased drastically when the monoclinic phase fraction exceeded 60%. Several samples were collapsed during the aging at 200°C of which monoclinic phase fraction finally exceeded 65%.
The effect of corrosion on fatigue fracturing of silicon nitride ceramics was investigated in 1N and 3N H2SO4 solutions. The crack propagation rate da/dt increased when the stress ratio R under cyclic loading was decreased. The higher concentrated solution enlarges the stress ratio dependence of da/dt. Therefore, it is considered that corrosion increases the effect of cyclic loading on da/dt. According to results of an energy dispersive X-ray spectroscopy (EDX) analysis, it was found that the sinterring additive components of Al and Y decreased at the crack tip. This suggests that the grain boundary phase at the region dissolved. In addition, a similar effect was also observed by scanning electron microscopy (SEM). Consequently, it is concluded that the dissolution of the grain boundary phase contributes to the increase of da/dt in H2SO4 solutions.
The compatibility of oxide whiskers with 6061Al alloy and the age hardening behavior of composites prepared from powder metallurgy were investigated. Interface reactants containing Mg were shown to be formed during hot pressing and hot extruding for every composite. The composites did not reveal age hardening behavior due to the depletion of Mg in the matrix at all. To keep whiskers from the interfacial reaction, electroless nickel plating onto whiskers was carried out. Although the composite with Ni plated whiskers revealed the age hardening, its hot workability was too poor to apply practical fabrication because of the formation of brittle intermetallic compound Al3Ni in the matrix.
Longitudinal creep tests of a 6-ply unidirectional SCS-6/Beta21S metal matrix composite with 38 percent fiber volume fraction were performed at three temperatures of 450, 500 and 550°C at four stresses ranging from 900 to 1350MPa. Elongation and acoustic emission in the gage section were measured during creep testing, and the specimens were subjected to observation of the fibers exposed by etching the outer matrix layers after creep testing. Also performed were creep tests of the matrix material and tensile tests of the fibers extracted from a virgin specimen. In the creep tests of the composite it was observed that the increase of temperature hastened greatly creep rupture while fiber damage and acoustic emission occurred with little dependence on temperature. It was shown that the activation of matrix creep with temperature is mainly responsible for hastening creep rupture of the composite. Besides it was shown that the fiber bundle strength determined from the tensile tests of extracted fibers is effective in estimating the level of applied stress at which the composite survived long in the creep tests.
An attempt was made to improve the electrical properties of SnO2-glass composites by dispersing Cu particles in the glass composites. Cu particles were precipitated from matrix glass containing Cu2O by adding LaB6 as a reducing agent. From the experimental results, it was found to be possible to improve the electrical properties of the SnO2-glass composites. However, the electrical properties and the microstructures of the glass composites were supposed to be influenced by the dispersion state as well as the content of a reducing agent. Therefore, it will be necessary to investigate the influential factors on the dispersion of the reducing agent such as particle size and a mixing condition in details for a further improvement.