In order to study the yielding property etc. under shock compression of the Si3N4 ceramics, the Hugoniot curve measuring experiments were performed in the pressure range up to 35GPa. Two different kinds of sintered specimen (the normal and fine sintered ones) were used to examine the effect of fine structure on the yielding property. The Hugoniot parameters were measured by the inclined-mirror method combined with the keyed-powder gun. The Hugoniot-elastic limit (HEL) stresses of the normal and fine sintered ones were determined to be 10.2-15.5 and 15.0-22.0GPa, respectively, and both Hugoniot compression curves converge each other in the plastic region. As a result, it was found that, although these Si3N4 ceramics behave as elastoplastic solids under shock compression, the Hugoniot-elastic limits are strongly influenced by its fine structure.
High strain-rate deformation characteristics in some commercial aluminium alloys, or 1050, 3003, 3004, 5182 and 7N01 alloys, have been investigated at high strain rate over 103S-1. The main conclusions are summarized as follows; (1) The strain-rate sensitivity of flow stress in all commercial aluminium alloys has been observed in the high strain rate range. (2) Three types in typical change of ductility with strain rate have been observed in the high strain rate range; the first is that little change in ductility over strain rate as both 1050 and 7N01 alloys, the second is that ductility increases with strain rate as both 3003 and 3004 alloys, and the last is that ductility decreases with strain rate as 5182 alloy. (3) The alloys, that show larger ductility in the high strain rate range than that in the low strain rate range, are both 3003 and 3004 alloys, but both alloys of 5182 and 7N01 show lower ductility in the high strain rate range than in the low strain rate range, and 1050 alloy shows no change in ductility. (4) In commercial aluminium alloys, it is clear that one of the requirements to obtain larger ductility in the high strain rate range is an increasing value of strain rate sensitivity exponent m with strain.
The dynamic stress-strain curves of annealed aluminum alloy and solder alloy were measured by a more convenient impact test than the Hopkinson bar one. The plastic strain rate was expressed by an exponential function of the overstress, and the constant parameters in the function were estimated from the experimental relation between the overstress and the plastic strain rate. For the purpose of investigating the strain-rate effect on the flexual stiffness of the rate-sensitive material under a dynamic load, the elasto-viscoplastic behavior of an initially curved bar under longitudinal impact with a rigid hammer was analyzed numerically using the above-mentioned constant parameters. For the reason of difficulty in obtaining a closed form of solution due to geometric and material nonlinearity, the solid bar was replaced by a lumped-parameter model. The strain-rate dependent solution was compared with the strain-rate independent one in order to investigate the strain-rate effects on the residual strain and the residual deformation. As the result of this numerical analysis, it was shown that the impact caused the smaller residual strain (or deformation) in the rate-sensitive materials than in the rate-insensitive ones.
Though problems on a precise and concise evaluation method of strength behaviors of ceramics still remain open, extensive studies by many investigators have revealed the extremely low fracture toughness of engineering ceramics such as silicon nitride and silicon carbide. However, on the other hand, these engineering ceramics have several excellent properties; high compression strength, high strength at elevated temperatures and high wear resistance. Such opposite aspects in their mechanical properties must be taken into account when we intend to apply these creamics as a structural material. Furthermore, in order to establish the design concept for the application of engineering ceramics, the whole strength characteristics must be systematized by compiling the strength data covering mechano chemical reactions in aggressive environments and further the dynamic response to transientally varying load such as impact load. The aim of the present study was to clarify the impact strength characteristics of representative engineering ceramics at high temperatures. For this purpose, the impact fracture strengths of silicon nitride and silicon carbide at high temperatures upto 1300°C were investigated by using an impact 3-point bending load device. The results indicate that the sintered silicon nitride has excellent high temperature impact strength, i.e., the degradation of fracture strength at high temperatures is less under impact load than under static load. The impact strengths of two types silicon carbides were also described.
A novel impact bend test procedure is described for determining the dynamic fracture-initiation toughness, KId, at a loading rate KI of the order of 106MPa√m/s. The split Hopkinson pressure bar technique was used to measure dynamic loads applied to a fatigue-precracked bend specimen. The dynamic stress intensity factor history for the bend specimen was evaluated by means of a dynamic finite element code. The onset of crack initiation was detected using a strain gage near a crack tip. The KId value was determined from the critical dynamic stress intensity factor at crack initiation. A series of dynamic fracture tests were performed for a Ti-6246 alloy and a 7075-T6 aluminum alloy. The KId values obtained for both the alloys were compared with the corresponding values obtained under quasi-static loading conditions. The fracture surface morphology was examined with the help of a scanning electron microscope to rationalize the apparent differences in the measured fracture-initiation toughness in response to the change in loading rate. Furthermore, the validity of the KId determination was confirmed by measurements of the crack tip plastic zone size in the interior of the fracture specimens based on the microhardness tests.
Impact fracture toughness KId tests were carried out on a alumina and alumina-zirconia composite at temperatures up to 1200°C. The one-point-bend impact tester was used and the specimens were pre-cracked by the bridge indentation method. A simple formula was employed for evaluating the dynamic stress intensity factor KI(t) from the measured impact force. In order to examine the accuracy of this KId measuring system, the natural angular frequency of the pre-cracked specimen was measured by using a strain gage and a FFT analyzer, since it has a significant effect on evaluating KI(t). The measured values were found to be in good agreement with those calculated from the simple formula based on the Euler-Bernoulli beam theory. It is concluded that the natural angular frequency can be precisely estimated by the formula. The existence of the contact force acting on the crack surfaces was confirmed by comparing the static load-strain curves of the pre-cracked specimens with those of the notched specimens. The effect of contact force on the measured KId value was discussed in relation to the test conditions such as loading rate, specimen size and time-to-fracture. It was experimentally demonstrated that the effect of contact force was negligible in the present test conditions. It was found that the fracture toughness values of alumina and alumina-zirconia composite increased with the loading rate, and the KId of alumina was almost constant at temperatures up to 1200°C, while the KId of alumina-zirconia composite decreased with test temperature.
The free end of an anvil which is in contact with a sphere undergoes the normal collinear impact by traveling bar. If the length of the anvil is small, several wave reflections may occur in the anvil. In this paper, the impact load of sphere caused by the reflected waves was theoretically analyzed by a wave equation and the Hertz's contact theory. Since it is difficult to theoretically evaluate the reflection coefficient of wave at the discontinuous section where the anvil and the sphere are in contact, it was experimentally determined by observing the reflected wave passing through the anvil by a strain gage method. The particle velocity due to reflected wave was related to the load which the anvil transmits into the sphere. When the reflected wave arrives at the end of anvil before the load of sphere arisen by the former wave reaches the maximum value, impact load should be evaluated by using the particle velocity added to the first. On the other hand, when the reflected wave comes after the maximum value of load, the next particle velocity only affects the load transmitted into the sphere. When the circular cylinder with hemisphere end was used in place of the sphere in the experiment, the impact behavior depends only on the curvature of hemisphere and not on the shape of other part, because the ratio of length to diameter of the cylinder was not so large. As the number of repetitions of reflected wave increased, the maximum value of impact load grew up, but the acting period of force decreased. The analytical results of impact load were in agreement with the experiments.
It has been observed that a practical form of the constitutive equation, which is called a logarithmic or a power law type, well represents an experimental stress-strain curve up to the strain rate of about 103S-1. The both types show gradual dependency of the strain rate on the stress. On the other hand, the deformation drug is governed by the viscous drug of dislocation at a high strain rate above 103S-1. In the range of high strain rate, the relation between strain rate and deformation drug is linear and one of the Malvern types, where the over stress is a linear function of the strain rate, satisfies the relation described above. However, almost all of the constitutive equations have been investigated only in the small strain range up to now. In this paper, a new practical form of the constitutive equation which covers a wide strain range at high strain rates was investigated by comparing the experimental strain distribution with the numerical one for a lead specimen of finite length, which was impacted longitudinally. The results show that the new form, in which the logarithmic or the power law type is superimposed upon the Malvern type, is in good agreement with the experimental results. It is also important that the strain rate appears as a function of the strain in the form.
In the present study the repeated scratch tests were made on the (111) face of the αCu-Al single crystals in the ‹211› and ‹110› directions with a unidirectional or reciprocating motion. After the (111) face had been scratched, the face was electropolished to remove a thin layer of suitable thickness. The slip systems and the etch pit distributions were observed on the exposed surfaces. The results obtained are as follows: (1) The deformations produced by scratch in the  direction are found to be more pronounced than those in the  scratch. The active slip systems on both sides of the scratched track differ from those observed under the track. (2) The scratched tracks and the slip traces on the scratched faces and the dislocation distributions inside of the material show that the deformation produced by the unidirectional scratch is more remarkable than that produced by the reciprocating scratchs in both cases of ‹211› and ‹110› directions.
A numerical method has been developed to analyze buckling phenomena such as edge wave or center buckle often observed in hot strip rolling. The finite element technique for a non-linear plate problem was used and the wave number as well as the displacements were obtained as its solutions. The features of the method are as follows. (1) Since the high degree iso-parametric shape functions were used in the present method, the displacement, slope, bent and twist at the corner of each rectangular element could be solved directly. (2) Internal force caused by thermal expansion and transformation was integrated by an incremental method. Therefore the method is capable of predicting transient behavior of buckled wave observed in the thermo-mechanical control process.
Fracture under mixed Modes I and II was investigated experimentally on MDF (medium density fiberboard) and DuPont Corian. The experimental results were compared with various fracture criteria hitherto proposed in the literature. The fracture criterion obtained for MDF follows approximately the criterion by Amestoy et al., and the one obtained for DuPont Corian follows approximately the maximum stress criterion and the criterion by Amestoy et al. On the other hand, the direction of crack growth obtained for MDF and DuPont Corian do not follow any of these criteria.
The residual stresses of TiC layer and WC phase in substrate, and strength of WC-10 mass% Co alloys coated with TiC by CVD were studied as a function of carbon content of the alloy. It was found that the residual stress of TiC layer determined by X-ray indicated tensile stress and that of WC phase in substrate indicated compressive stress. The stress of TiC layer increased with increasing carbon content of substrate. The formation of η phase was effective for stress relaxation of TiC layer, but the strength decrease was unavoidable.
Silicon nitride ceramics were diffusion-bonded to Nimonic 80A, a Ni-based metal, by use of Ni-plate insert between them. By applying FEM techniques a variety of stresses developed in this ceramics-metal combination was estimated. The strength of joints under residual stress conditions was then calculated to evaluate the effect of insert introduced for stress release. The results satisfactorily agreed with the measured tensile strength for various thicknesses of Ni insert: the Ni thickness has an optimum value to yield a maximum tensile strength.
The main objective of this study was to clarify the strength degradation and the strain behavior of engineering ceramics at high temperature during long-term range. For this purpose, a high temperature creep testing machine for ceramics was newly developed together with a displacement measuring device. To begin with, alignment of load transmission train of this testing machine was carefully checked and sufficient performance of this testing machine was confirmed. Then, experiments were carried out by using sintered silicon nitride and sintered silicon carbide at several temperature levels up to 1723K and during long-term up to 1000 hours. The results indicated that the creep rupture curves of both ceramics showed a straight line relationship on log-log diagram of applied stress vs. rupture time at respective temperature levels, and furthermore that the scatter of the rupture lives may be negligible when compared with that of cyclic fatigue strength of these ceramics. Discussion on the creep rupture life of sintered silicon nitride from the viewpoint of the creep strain behavior revealed that the Monkman-Grant's law governed the rupture time of this type of ceramics analogous to the case of metallic materials.
This paper describes the notch effect in multiaxial creep-fatigue loadings for a type 304 stainless steel at an elevated temperature. Push-pull and reversed torsion creep-fatigue tests with hold-times of 3, 10 and 30min. were carried out at 873K for three types of notched specimens. The effect of stress concentration was more significant in torsion than in tension. The notch root local displacement (NRD) approach in combination with the frequency modified fatigue life equation provides a satisfactory data correlation for the creep-fatigue crack initiation life in tension and torsion.
Y-Ba-Cu-O ceramics (mixing atomic ratio (Y:Ba:Cu)=(1 to 4):(1 to 5):(1 to 6)) were prepared by solid state reactions at 950°C for 17hrs in O2, starting with Y2O3, BaCO3 and CuO. When kept in water at 20°C, most of the superconducting black ceramics became white within 1 to 2 days, due to the H2O attack. However, the ceramics made of mixing atomic ratio 1:1:2 was still black and had Meissner effect up to 20 days. Jc of the 112 ceramics (25A/cm2) was smaller than 39A/cm2 of 123 ceramics. The mixing composition of the 112 was the same as that of the mixture of 123 and YCuO2.5 (20.5wt%). It is expected that the water resistivity and superconducting properties of the ceramics composed of 123 and YCuO2.5 (3 to 15wt%) may be better than those of the 123 or 112 ceramics. The Tc (zero) and Jc (77K) of these ceramics (123-YCuO2.5) were about 92K and 28-60A/cm2, respectively.However, the Jc of these ceramics when kept in water at 20°C for only 1 day decreased below 5A/cm2, in spite of these ceramics being black and having Meissner effect. When kept in salad oil, these ceramics were still black, and Tc (zero) and Jc were about 92K and 18-53A/cm2 even after 90 days. It is concluded that the H2O attack is very sever, but the superconducting properties are not so much damaged when kept in oil, due to the lack of water.
The distinct element method (DEM) have the potential applicability for analysis of fissured rock structures and granular materials. Before its practical application, a suitable procedure for determining element constants should be developed. Simple evaluations of element constants K and η by elastic contact theory and inelastic attenuation of rebounding, respectively, are proposed, and the results of laboratory tests, such as a free-fall rebounding test and Schmidt hammer test are described in detail. Finally a slope stability of blocks are submitted to the DEM analysis and the relationships among the element constants K, η, φμ, the angle of inclination and the block dimensions are discussed.