Fracturre surface apppears to be a three dimensional image when stereo-pair photographs are examined in a stereoscope. However since such image is not quantitative it is necessary to build up accurate three dimensional topography associated with fracture surface in order to increase understanding of fracture process. In this study, it became possible to obtain 3-dimensional fracture surface topography from stereo-pair of SEM photographs. Position search algorithm in a stereo-pair of SEM photographs used was SSDA and MCC methods. After searching position it was calculated fracture surface height by using geometric equation. And then, we applied software to hydrogen embrittlement fracture surface of SUS329J1 duplex stainless steel for quantitative analysis. As a result of that, 3-dimensional image was well corresponded to fracture surface associated. Furthermore we also investigated influence of microstructure (austenite, ferrite, sigma phase) on hydrogen embrittlement behavior by applying both 3-dimensional analysis and microstructure observation. As a result of that it found sigma/austenite phase boundaries and sigma/ferrite phase boundaries are preferential fracture site, while ferrite quasi-cleavage fracture in as-received material was dominant.
Three-dimensional analysis of fracture surface by SEM (Scanning Electron Microscope) is inevitable to clarify the cause of fracture. It is effective to analyze with stereo matching. However, conventional template matching method to measure 3-dimensional profile of fracture surface usually employs fixed template size and has deficiencies such as mismatching and much processing time. In this paper, we develop a new stereo matching algorithm, which can save processing time and assure accuracy. In this new template matching method, template size is varied to assure accuracy and lessen mismatching points. When calculating the correlation coefficient of image template pixels to search the matching templates, we don't use all the pixels but choose one pixel every 2 or 3 pixels to save processing time. In order to investigate the precision and the usefulness of the proposed method, two examples of reconstructing fracture surfaces are successfully shown. It is practically realized to measure three-dimensional profile of fracture surface in personal computer.
In order to clarify the difference of impact fracture behavior of hot work tool steel SKD6 over a range from 300K to 873K, the Charpy impact test and the observation of the fracture surfaces were carried out using V notch specimens and without/with prefatigue crack specimens. The energy require to initiate a crack at room temperature is greater than the energy for the crack propagation, while at elevated temperature the tendency becomes reversed. Macroscopic fracture at room temperature occurred in a brittle manner. On the other hand, at 873K the wider ductile shear zone near V notch root and the shear lips appeared. But the specimens with prefatigue crack fractured without the ductile shear zone even at elevated temperature. In the crack propagate region the ripple pattern that consist of the dimple pattern and the slipped flat one was found. And the rate of slipped pattern increased with rising test temperature.
Tensile properties of Inconel 718 Ni-base superalloy were investigated in high-pressure hydrogen and in argon at room temperature. The effect of Nb-Ti carbide on hydrogen environment embrittlement (HEE) of the superalloy was also examined. The results obtained are as follows: (1) Hydrogen decreased elongation, reduction of area and ultimate tensile strength of the superalloy. HEE increased with decreasing the strain rate. Dimple rupture was mainly observed in argon, but brittle transgranular and intergranular fracture were observed in hydrogen. The crack initiation occurred at carbides both in argon and in hydrogen. Then the crack propagated in the matrix with ductile fracture in argon and with brittle fracture in hydrogen, respectively. (2) The finite element analysis (FEA) was applied to the crack initiation of the superalloy during the monotonic tensile process with the FEA model, which contained one particle at the center surrounded by the matrix. The effect of the shape of the particle on the maximum effective stress in the particle was simulated. The maximum effective stress in the particle increased markedly with increasing the average stress above the proof stress. The increasing order of the maximum effective stress depended on the shape of the particle. It was reasonable that the maximum effective stress in the carbide is high enough to break the carbide itself. It was also expected that the control of the shape of the carbide could improve HEE of the superalloy.
Crack growth characteristics due to stress corrosion cracking and its fracture surface were investigated for Si3N4 ceramic using indentation pre-crack. The effect of test environment on the crack growth was studied. The effect of water environment on crack propagation is detrimental. In the prediction of residual life the effect of stress corrosion cracking should be considered. The ratio of intergranular fracture surface to transgranular fracture surface is relatively small (20-30%). The dominant fracture mechanism is considered to be the transgranular fracture.
A long carbon fiber with 5.5GPa in average tensile stress-reinforced epoxy plastic (CFRP) has been studied to determine the environmental effect on the tensile stress and fracture behavior of the advanced CFRP. The cross-ply laminates (0°/90°plies) were used in this study. Unnotched and notched tensile tests were conducted using Instron machine from ambient temperature to 130°C. The tensile specimens were immersed in distilled water, 40wt.% nitric acid and 40wt.% sulfuric acid at ambient temperature for a month (designated as distilled-water, nitric-acid and sulfuric-acid specimens, respectively). The results were compared with those obtained by the non-immersed specimens. The tensile fracture stresses of unnotched and notched specimens increased in the following ascending order: non-immersion, distilled-water, nitric acid and sulfuric acid. However, there was little difference in the tensile stress between immersed and non-immersed specimens at 130°C for unnotched specimens and at and above 80°C for the notched specimens. The results are described and discussed in terms of fractography, EPMA and FT-IR analysis and so on.
Fretting fatigue tests were conducted on structural steels such as SCM415H, SCM440 and SAPH45 and cast iron FC25. Consequently, it was clarified that the stress over the contact area where fretting failure initiate was dependent on the Vickers hardness number of the steel surface. It was also clarified that the repeated bending stress at the contact area where fretting fatigue failure initiate linearly decreased as stress over the contact area increased. The stress over the contact area where fretting fatigue failure initiate was increased by eighty percent in application of Defric coating to SCM440. Striation, fretting debris and pit were predominantly observed on the surface of the contact area. Crack initiation mode was transcrystalline on SCM440 and SAPH45, and was intergranular on SCM415H. It can be concluded that the fretting fatigue crack initiate from a pit formed at the contact area due to slip movement.
Studies have been reported so far with regard to estimation of the stress acted on damaged members in actual machines under a constant load condition using the X-ray fractgraphy technique. However, the load applied on actually used machines and structures are variable. There are many unknown items with regard to application of the X-ray fractgraphy under such variable condition, and no report has been seen so far about such application. As a part of research to clarify the effect of the variable loading on the correlational expression of X-ray parameters and fracture mechanic parameters, we have examined the effect of the variable loading that the stress ratio and the maximum load vary once in the fatigue crack propagation process, then made it clear that the stress could be estimated with high precision by acquiring the information on the hardness just below the fracture.
In order to investigate the microstructure of dragline silk of Nephile clavata (Jorougumo), we used its super contraction behavior in urea/salt solvent to facilitate the observation of cross section. We proposed a model of the dragline consisting of three layers at least; outer coat, middle layer with micro fiber and inner core. The model is slightly different from the 4-layers model of Vollrath et al. The outer coat may play a roll as the resistance to aqueous environment such as rain and fog.
Ti-Ni shape memory alloy (NT-L) was brazed to commercial pure Ti (CPTi) with Ag-Cu filler (BAg-8) or Cu-Ti-Zr amorphous filler (MBF5004). The effect of brazing condition was investigated on the reaction layer formation as well as on the joint strength, and then their relation was discussed. When brazing was performed with BAg-8 under the brazing temperature lower than 1153K, four reaction layers were formed adjacent to brazing joints. The maximum joint strength was about 330MPa, and the fracture occurred around the Ti-Cu based intermetallic compound reaction layer isothermally solidified between CPTi base metals and BAg-8 filler during holding at brazing temperature. At an early stage of the isothermally solidified layer formation, the joint strength increased with its growth. When the thickness of the layer became larger, however, the joint strength decreased with the growth. For the joints brazed at the brazing temperature higher than 1193K with BAg-8, two reaction layers were formed. The maximum joint strength was about 350MPa, and the fracture occurred in the brazed layer consisted of αTi and Ti2(Ni, Cu). The thickness of the brazed layer was three times larger than that of BAg-8 filler, suggesting that NT-L base metal near the bondline melted during brazing by the diffusion of Ti to NT-L base metal. In the case of MBF5004, joint microstructure and fracture path were similar to the cases of BAg-8 being the brazing temperature higher than 1193K. The joint strength was larger for the joint using BAg-8 than that using MBF5004, and was almost the same as that of CPTi base metal.
To analyze the rate-controlling mechanism of Lüders deformation found in the  oriented Fe-30%Cr alloy single crystals, the activation volume and the activation enthalpy were obtained by a strain rate and temperature change method during the deformation from 293K to 511K and were found to be 94-176b3 and 0.8-1.4eV, respectively. Aging at 473K was made after extension by 23.5%. The Lüders band didn't move on further straining and the second Lüders band started from opposite side of the chuck. It was considered that the Lüders front was aged and locked by some solute atoms. The line intensity of carbon atom fairly increased around the Lüders front after aging. The interaction of moving dislocations and carbon atom is considered to be the most likely rate-controlling mechanism for the deformation. The activation distance was calculated to be 4.1-6.2 times larger than carbon diameter and would be reasonable value considering the moving dislocations overcome carbon atoms.
In the present investigation, open die forging tests have been conducted for AZ31 and ZK60 Mg alloys. Microstructure and mechanical properties of the forged Mg alloys have been investigated. Cylindrical specimens with a diameter of 18mm and a height of 18mm were forged in a temperature range of 323 to 673K. The forged specimens of both Mg alloys had good surface quality at high forging temperatures of more than 573K. Microstructural observation revealed that grains were recrystallized by hot forging, though unrecrystallized grains were partially found. The recrystallized grain area for AZ31 was larger than that for ZK60. The mean size of the recrystallized grains decreased with decreasing forging temperature. In particular, a very small grain size of about 3μm was attained by forging at 473K. As a result of tensile tests, it has been found that ultimate tensile strength and 0.2% proof stress of the forged Mg alloys increased with a decrease in forging temperature. This was related to grain refinement by hot forging.
A series of ratchetting deformation tests was conducted on modified 9Cr-1Mo steel at 550°C under uniaxial and multiaxial stress conditions. Ratchetting behavior depended on various parameters such as mean stress, stress/strain rate and those range, hold time and prior cyclic deformation. Under uniaxial conditions, untraditional ratchetting behavior was observed; the ratchetting deformation rate was the fastest when the stress ratio was equal to -1, while no ratchetting deformation was predicted by conventional constitutive models. In order to discuss the reason for this untraditional ratchetting behavior, a lot of monotonic compression tests were conducted and compared with tension data. The material showed a difference of deformation resistance of about 30MPa between tension and compression at high strain rates. Furthermore, the authors' previous model and Ohno-Wang model were applied to the test conditions to evaluate their description capability for ratchetting behavior of the material. It was shown that the authors' model has a tendency to overestimate the ratchetting deformation and that the Ohno-Wang model has a tendency to underestimate the uniaxial ratchetting deformation at small stress rates.
Based on the authors' previous model and Ohno-Wang model, a viscoplastic inelastic constitutive model considering tension-compression asymmetry was developed to describe ratchetting deformation behavior of modified 9Cr-1Mo steel adequately. In the new constitutive model, the first invariant of stress was introduced into the flow rule in order to express the asymmetry. Furthermore, static recovery term was introduced into the Ohno-Wang type kinematic hardening rule to express rate/time-dependent deformation accurately, and strain range dependency of cyclic softening behavior was introduced to express effect of cyclic softening on ratchetting deformation precisely. Through the application to various test conditions, it was demonstrated that the present inelastic constitutive model is capable to describe uniaxial and multiaxial ratchetting behavior of the material with reasonable accuracy, including unordinary ratchetting behavior under vicinity of zero mean stress as well as the asymmetric monotonic and cyclic deformation behavior.
Attenuation characteristics of ultrasonic transverse waves in unidirectional carbon fiber-reinforced plastics are examined based on a theoretical model recently put forward by the authors. The model accounts for energy losses due to wave scattering by the fibers as well as viscous absorption in the matrix, and gives the composite attenuation coefficient as a function of frequency. In the present study, two modes of ultrasonic transverse waves, i.e. with polarization direction normal and parallel to the fiber direction, are considered for the common propagation direction normal to the fibers. Moreover, the model is extended to deal with composites with highly concentrated fibers using a differential scheme. Based on the results of the numerical analysis, attenuation behaviors of the two transverse wave modes are discussed in detail. The results show that the attenuation coefficients of the composite are smaller than that of the epoxy matrix, for a frequency range relevant to practical situations. Furthermore, attenuation in CFRP is more pronounced in the case when it is polarized parallel to the fibers than in the case when it is polarized perpendicularly. Attenuation characteristics of transverse waves in CFRP are also compared to those of the longitudinal wave analyzed previously.
Mixed mode I+II delamination crack growth tests at elevated temperature (473K) were carried out on unidirectionally carbon fiber reinforced thermoplastics (CFRTP) under static creep condition. Asymmetric end loaded split (AELS) specimens were used in order to investigate the effect of mixed mode ratio, GI/GII, on crack growth. The creep crack growth rate, da/dt, became higher in terms of mode I component of total energy release rate, GI, as GI/GII decreased. This implies that the crack propagation rate was accelerated by mode II component, GII. The measurement of the load point displacement rates due to mode I or mode II component of total load revealed that the creep deformation at the crack tip by mode II component was larger than that by mode I component and the size of creep deformation area depended on GI/GII. It is considered that the creep deformation by mode I component, which was perpendicular to carbon fibers, was constrained strongly by fibers, while the creep deformation by mode II expanded along carbon fibers. The crack growth rate under mixed mode loading was governed by the creep J integral, J, regardless of the mixed mode ratio. Fractographic observation showed that matrices were elongated along carbon fibers and fracture occurred at the interface between fibers and matrix.
WC-Co cermet was coated on the steel substrates (JIS: SKD5, SKD62 and SS400) with different hardness by two types of spraying methods using high-velocity oxygen fuel (HVOF) and high-pressure high-velocity oxygen fuel (HP-HVOF). Tensile tests as well as Edge-indent tests were carried out to obtain interfacial fracture toughness GC12 and delamination energy Ed, respectively. The GC12 and Ed of the coatings deposited on the SS400 substrate with smaller hardness revealed larger values, and the GC12 and Ed of the coatings deposited by HP-HVOF were higher than that by HVOF with the same hardness of substrate. The reason can be that the sprayed powder with higher speed more easily shot into the soft substrate than the hard substrate so that the mechanical bond effect along interface between coating and substrate was stronger.
Fatigue tests of smooth specimens cut from squeeze cast Al alloy car wheels were carried out to clarify the fatigue behavior under the constant stress amplitude. In addition to this, fatigue tests under the two-step loading were performed to study the effect of stress change on fatigue damage. To minimize the scatter in fatigue tests due to the fluctuation of laboratory atmosphere, all the tests were made in controlled air (t=30°C, h=60%). Fatigue cracks were initiated principally from the eutectic Si particles and sometimes from a slip band in the matrix. But, no initiation from microscopic defects such as pin-holes and shrinkage porosities was observed. The lnl vs. N/Nf relation under constant stress amplitude exhibited stress dependency, namely the crack length initiated at same N/Nf was larger in higher stress. With regard to the fatigue damage under the two-step loading, the cumulative cycle ratio Σ(N/Nf) for the low to high block loading was smaller than unity, however it was larger than unity for the high to low block loading. The value of Σ(N/Nf) was closely related to the growth characteristics of a small crack.
The FRP adhesion method which uses carbon or aramid fiber sheets for the strengthening of the flexural concrete members of a building or bridge has lately become a subject of special interest. The bonding properties between the FRP sheets and concrete influence the structural properties of the concrete members which are strengthened by this method. And these properties may be classified into three types according to the direction of the debonding force; the shear bonding property, the cleavage bonding property and the tensile bonding property. This paper concerns the cleavage bonding properties, which have not yet been published in any other reports. In order to obtain the properties, compact tension tests were conducted, varying the type of treatment method concerning concrete surfaces, the compressive strength of the concrete and the type of fiber sheet. The bond softening diagram and fracture energy were derived from the data as the cleavage bonding properties. The bond softening diagram was obtained by applying the poly-linear approximation method to the load-displacement curve measured from the test. Also, the fracture energy was given by integration along a softening diagram. The conclusions obtained from this study are as follows: 1) roughening of the concrete surface remarkably improves the cleavage bonding properties, if the surface treatment is carried out without loosening bond between the aggregate and the mortar matrix, 2) increase of the concrete compressive strength improves the cleavage bonding properties, and 3) the type of fiber sheet has little effect on the cleavage bonding properties.
The strength of composite materials depends on absorbing moisture. The aircraft structural integrity should be demonstrated by considering every factor which degrade its strength. The composite aircraft structural components should be tested after moisture absorption equivalent to the component lifetime exposure. However, the thick laminate needs a long period to absorb moisture in a conventional manner, such as high temperature and 100% humidity but ambient pressure. Then the rapid moisture absorption technique has been studied using a pressure chamber. The purposes of this study are cutting the period for moisture absorption and adjusting the moisture distribution inside components so that much higher reliability of the aircraft, the cost reduction of development and weight reduction of the components will be obtained. Some moisture absorption tests have been carried out. From this study, following results were obtained. 1) Under some high pressure and temperature condition, moisture absorption speed can be accelerated. 2) New analytical method can be suggested. 3) By using this technique, the period for 20mm thick laminates can be reduced by a factor of 3.5 under a 20 year moisture absorption condition 4) The appropriate moisture distribution inside of the thick laminate can be obtained by this technique.