The testing of metal injection molding of SUS316L powders using a PE (polyethylene)-PP (polypropyrene)-sesame oil binder system was carried out. When the green parts were immersed in a THF (tetrahydrofuran) solvent at 50°C for 2hr, the sesame oil was all extracted. The residual parts were debound in nitrogen gas. The debound parts were sintered at 1320°C in vacuum for 1hr. The relative density of the sintered specimen was 95.1%. The carbon and oxygen contents of the sintered specimen were 0.001wt% and 0.03wt%.
Bamboo has a composite structure since its cylindrical wall consists of bundle sheath and parenchyma. In this study, the relationships among its mechanical properties, the volume fraction of bundle sheath and the specific gravity were investigated. The experimental results obtained are as follows. (1) The volume fraction of bundle sheath Vf increased approximately parabolically toward the outer bark from the inner bark, and also increased relatively with height. (2) The specific gravity ρ and Young's modulus E also increased similar to Vf. Therefore, Vf could be evaluated by the measurement of ρ because a linear relationship existed. (3) The mechanical properties could be evaluated by the measurement of ρ because the rules of mixture existed between Vf and E or σB in the experimental region.
In this paper, numerical analyses of torsional stress and rigidity were carried out on two types of nonsymmetrical three-layered sandwich and covered beams subjected to torsion, by utilizing the exact and briefly arranged analytical solution. The following different features of these beams became clear from the numerically calculated results. (1) The maximum shear stress in the sandwich beam with hard core appeared at the middle point of the thicker layer of the outer elastic layers, and for the beam with soft core it appeared at the same point of the thinner outside layer. But for the covered beam, it appeared always in the thicker outside layer. (2) Shear stress in the core of the sandwich beam became highest at both the top and bottom interfaces, and it was highest in the symmetrical beam composition. The maximum stress for the covered beam, which did not appear at the interface, was lowest in the symmetrical beam composition. (3) An appropriate core layer thickness existed in order to reduce shear stress at the interface of two-layered beam by core insertion. In this case, the saturation of stress reduction was induced by inserting a thicker core, and moreover, higher stress than that of the original two-layered beam was induced sometimes at the interface. (4) Torsional rigidity of the symmetrical sandwich beam was lowest for the soft core and highest for the hard core. On the other hand, for the covered beam it was lowest in the symmetrical and highest in the two-layered beam composition. These characteristics for torsional rigidity are controlled by the increase of torsional rigidity of the elastic layer of each beam created by bonding.
The data of static mechanical properties of many kinds of ferrous metals were sampled from the database on the fatigue stength of metallic materials in JSMS. The correlations among various mechanical properties were investigated. When brinell hardness and Rockwell hardness were converted into the corresponding Vikers hardness, the relation between Vikers hardness and tensile strength can be expressed as linear. The relation between yield strength (yield stress or 0.2% proof stress) and tensile strength can be classified into two groups, each of which has a linear relationship. The first group contains low strength steels such as annealed carbon steels, normalize carbon steels and quenched and tempered carbon steels (<0.30%C) etc.. The second group contains relatively high strength steels such as alloy steels and quenched and tempered carbon steels (>0.30%C) etc.. The relationship between elongation and tensile strength in all irons and steels can be expressed as an exponential function. The data of austenitic steels and cast irons deviate from the relationship. The relation between reduction of area and tensile strength is expressed as three groups with linear relationship. The first group contains normalized and annealed carbon steels and quenched and tempered carbon steels (<0.30%C). The second group contains quenched and tempered carbon steels (>0.30%C). The third group contains alloy steels.
The surface roughening during tensile and compressive plastic deformations of polymers was investigated with a stylus instrument. The surface roughness of nylon, grey polyvinyl chloride and polyethylene increased with increasing applied strain. The surface roughness of polypropylene and transparent polyvinyl chloride increased in the early stage of tensile deformation, and it remained almost constant in the later stage. Surface roughening was hardly observed for polycarbonate. To investigate the mechanism of surface roughening, the deformed surface was observed by a scanning electron microscope. It was found that the development of fibrous structure, the break of the fibrous structure near the surface and the formation of craze contributed to the surface roughening.
Hypervelocity-impact experiments were performed on the titanium alloys and aluminum alloys to study relation between the crater configurations and the static material properties. The copper projectiles were accelerated by means of a two-stage light-gas gun at averagely 2000m/sec, and were impacted on the zero oblique specimens. The crater configurations (crater depth, crater volume and mean crater diameter) were observed, and the relation to their static material properties (density, proof stress, elongation) were studied. As a result, several correlations between the deformation by the hypervelocity impact and the static material properties were found, as follows; (1) The crater depth of the both materials was dependent on their density and proof stress, and was expressed by the next equation. λ=37.8ρt-0.671σ0.2-0.175 (2) The crater volume of the both materials decreased with an increase of the product of proof stress by elongation. (3) The aspect ratio of the crater of the both materials decreased with an increase of proof stress. (4) The crater configurations of the titanium alloys were different from these of the aluminum alloys. Decreasing the product of proof stress by elongation resulted in the increase of the mean crater diameter of the titanium alloys, whereas for the aluminum alloys, it resulted in the increase of the crater depth and the mean crater diameter, particularly the crater depth.
The strength of directionally solidified Ni3Al-Mo alloys was investigated in tension at temperatures ranging from room temperature to 1273K. The 0.2% flow stress showed the positive temperature dependence. The flow stress increased with increasing temperature, reached a peak at 973K and then decreased. The effect of Mo fiber on the strength was dominant at temperatures lower than 973K, but it became weaker above 973K because the strength of Mo fiber decreased. The microstructures of deformed Ni3Al matrix changed at test temperatures. Below the peak temperature long straight dislocations, which appear be formed by cross slip, were found in the matrix. At temperatures higher than 973K three-dimensional networks of dislocations were formed, which indicates that the recovery process is more active.
A constitutive equation for shape memory alloys, which governs the thermomechanical as well as the transformation processes of the materials, was presented in the process of stress-induced martensitic transformation. A theoretical chain connecting the deformation of microstructures and the macroscopic behavior of alloys was well established by introducing two different microstructural levels of alloys; the microregion and the mesodomain. All material parameters in the constitutive equation are shown to be fully determined from the microscopic material data.
The authors have been systematically studying stress relaxation properties for typical heat resisting materials which are used in high temperature plants. In this study, the influences of test temperature and total strain on the long-term stress relaxation behavior for a 1Cr-Mo-V turbine rotor steel and a 2.25Cr-1Mo pressure vessel steel and the relationship between stress relaxation properties and creep properties were investigated, and a simple prediction method of residual stress was proposed. The residual stress at shorter times in stress relaxation curves for 1Cr-Mo-V steel showed the same dependence on temperature as that at longer times. The dependence of residual stress on temperature for 2.25Cr-1Mo steel was fairly small at longer times. The influence of total strain on residual stress for 1Cr-Mo-V and 2.25Cr-1Mo steels was small. It was found that the relaxed stress ratio at a specified time in stress relaxation tests was related with the rupture time in creep test at the same stress as the initial stress in the stress relaxation test, and the relationship between the relaxed stress ratio and the rupture time was nearly independent of test temperature, total strain and materials. A simple prediction method of residual stress in stress relaxation tests was proposed on the basis of the diagram of relaxed stress ratio versus rupture time obtained. The predicted residual stress from the proposed method agreed well with the observed one.
Vickers hardness test, four-point bending test and fracture toughness evaluation by two kinds of indentation (ISB and IF methods) were conducted on reaction sintered and pressure-less sintered silicon carbide ceramics, SiC-R and SiC-PL. The dependences of hardness Hv, bending fracture strength σf and fracture toughness KIC, on the indented load P were investigated. The effects of loading rate and atomosphere on the σf were also studied. The Hv decreased with increasing P for two SiC materials. The indented shapes were clear comparatively and four cracks initiated from vertices of the indentation for SiC-PL. For SiC-R, however, a part of the indented surface peeled off by the effect of lateral cracks initiated near the surface layer, and the fracture surface was not flat. For the smooth specimens, the mean value of σf of SiC-PL was higher than that of SiC-R. For the specimens pre-cracked by Vickers indentation, σf decreased with increasing P for the both materials, and the mean value and the variation of σf for SiC-R was larger than those for SiC-PL. For SiC-R, the dependences of σf on the strain rate and atomosphere such as in water and in silicon oil were not observed in the present experiment conditions. The KIC by the ISB method and by the IF method were almost identical in the same P condition. The KIC by the IF method decreased with increasing P for SiC-PL, while the KIC increased with increasing P for SiC-R. This might be caused by peeling-off of the indented surface layer of SiC-R.
In order to classify the fracture mechanism of composite materials, the PET resin and the model FRP test specimens were prepared and the relation between the detected AE frequency and the damage mode of those test specimens was examined by performing tensile tests. The effects of sensor position and tensile speed on the frequency characteristics were also examined by using FRPET test specimens. The results obtained are as follows: (1) The AE frequency ranges corresponding to the resin cracking in the PET test specimen, the fiber debonding and pulling out and the fiber breakage in the model test specimens were 60-240, 240-330 and 350-420kHz, respectively. (2) When the distance between the sensor and the center line of FRPET test specimen was changed from 20mm to 30mm, the AE signal also changed due to the wave attenuation; the peak frequencies which correspond to the fiber breakage and the fiber debonding and pulling out shifted to the higher frequency side, and the power spectrum at the frequency which corresponds to the resin cracking became lower. (3) When the tensile speed of FRPET test specimen was changed from 0.2mm/min to 0.05mm/min, the peak frequencies which correspond to the fiber breakage and the fiber debonding and pulling out shifted to the lower frequency side and the power spectrum became lower. The power spectrum at the frequency which corresponds to the resin cracking was affected by the tensile speed and became higher.
Electronic equipments and devices are being exposed to much more severe temperature environments than before. In the case of an equipment used in space, it is necessary to predict the thermal cycling life of solder joints under a low temperature environment below -100°C. Moreover, in the life assessment of solder joints, it is necessary to consider not only the simple shearing load condition but also the tension-compression and mixed mode load conditions in many cases. An attempt was made to investigate the influence of low temperature and load conditions on the low cycle fatigue properties of a solder material by carring out the tension-compression and torsional fatigue tests on the specimens of 36Pb62Sn2Ag solder at room temperature (20°C), -100°C and -140°C. The results obtained are as follows. (1) Although the ductility of 36Pb62Sn2Ag decreased at low temperatures (-100, -140°C) considerably in comparison with that at room temperature, no noticeable decrease in low cycle fatigue life was observed. (2) At low temperatures, the universal slope method and the Langer's method can be used effectively to predict the low cycle fatigue life. At room temperature, however, the test data indicate shorter life than the predicted value, and so the method suitable for predicting the high temperature low cycle fatigue life should be used. (3) All the tension-compression and torsional fatigue data at low and room temperatures can be unified approximately by using the equivalent strain of Mises' type. It is possible, therefore, to evaluate the low cycle fatigue life under a mixed mode load condition by use of the equivalent strain of Mises' type.
The purpose of this study is to investigate the influences of sample size n, level interval d and number of levels L on the distribution properties of the estimates, that is, the mean μ and the standard deviation σ, of the fatigue enderance limits obtained by Probit and staircase methods, by Monte Carlo simulation tequnique. Major conclusions are summarized as follows; (1) The distribution of the sample mean μ estimated by both methods agrees well with the normal distribution. (2) The distribution of the sample standard deviation σ can be approximated by the log-normal distribution. (3) There is little influence of the level interval d and the number of levels L on the mean μ of the sample mean μ by Probit method, and μ is in the range of ±0.01 in all cases, when the population mean is zero and its standard deviation is unity. This is also the case of μ by staircase method. (4) The mean σ of the sample standard deviation σ approaches the population value of unity from the larger side (σ>1) with an increase in sample size n in the case of Probit method, but σ approaches to unity from the lower side (σ<1) with an increase in n in the case of staircase method.
This paper deals with the effect of each factor in the reinforcement on the fatigue limit of welded joint. Fatigue specimens were made by using an electrospark machining for changing only one factor of the reinforcement without leaving any residual stress, increasing hardness and so on. As the toe radius of welded joint is very small, the fatigue limit of welded joint is decided by the condition that a non-propagating crack starts to propagate. Therefore it was investigated whether the fatigue limit is given by a critical stress intensity factor or not. The stress intensity factor was calculated by use of a program of boundary element method. The following conclusions are obtained: The fatigue limit of welded joint is predictable by using a stress intensity factor. The height of reinforcement (h) does not affect the fatigue limit when h is greater than 2mm in case of plate thickness being 4mm. The effect of width in the reinforcement on the fatigue limit depends on the loading condition of fatigue test. The flank angle (θ) does not affect the fatigue limit in case of θ being greater than 60 degree.
The aim of this investigation was to clarify the fatigue behavior of steel coated with TiN by the physical vapour deposition (PVD) or chemical vapour deposition (CVD), in order to apply the ceramics coatings to machine components and structures. Cantilever-type rotating bending fatigue tests in air were performed by using the round specimens of 0.37% carbon steel normalized, JIS S35C, coated with TiN by PVD or CVD. From the experimental results, 16-23% increase of fatigue strength was observed for the TiN coated specimen, as compared with that of the base metal. Also the fatigue crack propagation rate observed on the surface of coated specimen was equivalent to that without coating. It is suggested that the fatigue crack may initiate at the surface of base metal under the coating film. The increase of fatigue strength of the TiN coated material is attributed to the retardation of crack initiation due to the restriction of dislocation movement in near surface of substrate by hard thin film.
Fatigue crack propagation behavior under single and multiple overloading conditions in polycarbonate was studied fractographically. The results obtained are summarized as follows: (1) The fracture surface of the crack extension accelerated by a single peak overload consists of two different regions, one contains many fine tearing lines oriented parallel to the crack propagation direction, and the other contains many voids. The former is similar to that found in the striation, and the latter is similar to that found in the discontinuous growth band. (2) The amount of crack extension accelerated by the overload is almost constant inside the specimen, but decreases markedly near the boundary between the interior fracture surface and the shear lips formed at the specimen surface layers. (3) The acceleration factor of crack propagation due to the overload is dominantly dependent on the ratio of peak to baseline stress, and is not affected appreciably by the crack propagation history. (4) The degree of retardation of crack propagation after a single peak overload shows the dependence on baseline stress, which can be explained from the viewpoint of crack propagation mechanisms. (5) The width of characteristic retardation region which is produced after the multiple overloads is about equal to the size of plastic zone at the crack tip produced by the overloads.
Fatigue crack propagation behavior in cold-rolled polycarbonate was studied under constant-amplitude loading and single peak overloading conditions. The results obtained in this experiment are summarized as follows: (1) Cold-rolling improves the resistance of fatigue crack propagation at low ΔK levels in polycarbonate, and restrains the formation of discontinuous growth bands which are easily formed at low ΔK levels in unrolled polycarbonate. (2) The acceleration factor of crack propagation due to a single peak overload is markedly lower in cold-rolled specimens than that in unrolled ones, and the fracture surface of acceleration region shows many fine tearing lines in cold-rolled ones in contrast to the granular markings in unrolled ones. This fact suggests that the cold-rolling restrains the craze formation at the crack tip, which causes lower acceleration. (3) Fatigue crack propagation after a single peak overload shows the so-called delayed retardation in cold-rolled specimens, which contrasts with an undelayed retardation in unrolled ones. (4) The maximum retardation rate after a single peak overload is lower in cold-rolled specimens than that in unrolled ones. This is considered to be due to the smaller scale of the overload-induced crack tip blunting or orientation hardening near the crack tip in cold-rolled ones.
Fatigue crack growth behavior under varying amplitude load conditions can not be estimated presisely from the fatigue crack growth data under constant amplitude load conditions, even when discussed from the view point of the effective stress intensity factor, because it depends upon other factors such as crack tip blunting. Hitherto, crack growth under repeated block load (i.e., two-step) conditions has been ordinarily discussed microscopically, and on the other hand, those after single and multiple peak overloads have been discussed macroscopically. It is preferable to relate the above two approaching methods with each other in order to evaluate the crack growth behavior under varying load conditions systematically. In order to clarify the transient phenomena of fatigue crack growth behaviors under two-step varying load conditions, a series of fatigue crack growth tests were carried out on aluminum alloy (A2017-T3) CT specimens. An analytical crack growth law during the transient period was established through the empirical simulation between crack growth length and stress cycles in this period. The procedure presented in this study is useful to predict fatigue crack growth under other types of varying load conditions.
Stress corrosion tests were conducted on the notched compact tension specimens of quenched-tempered SNCM 439 steel in three kinds of environmental solution: distilled water, 3.5%NaCl, and 0.1N H2SO4. The crack nucleation behavior was analysed from a view point of fracture mechanics. The results obtained are summarized as follows; (1) Time to crack nucleation tn of the notched specimens with various root notch radii was determined by a parameter of Kρ/(ρ+ρ0)1/2, where Kρ is the apparent stress intensity factor, ρ is the notch root radius and ρ0 is the intrinsic notch root radius. The life got shorter as the hydrogen content became larger in the order of distilled water, 3.5%NaCl, and 0.1N H2SO4. (2) Stress corrosion cracks were formed from the notch on three different planes: the original notch plane and two inclined planes to the notch. The crack plane changed depending on environments, notch-root radii and stress levels. Two inclined planes of crack initiation from the notch agreed with the plane of the maximum shear stress.
In order to investigate the effect of titanium nitride (TiN) coating on corrosion fatigue behavior of metal, cantilever-type rotating bending corrosion fatigue tests were carried out in 3.0% saline solution by using the round bar specimens of 0.37%C carbon steel, JIS S35C, coated with TiN by the physical vapour deposition (PVD) method or the chemical vapour deposition (CVD) method. From the experimental results, obvious improvement of corrosion fatigue strength was observed in TiN coated specimens, as compared with those without coating. The corrosion fatigue crack propagation rate observed on the surface of coated specimens was equivalent to that without coating. Fracture surface observations with a scanning-electron microscope revealed that the corrosion fatigue crack propagated from small corrosion pits formed at the base metal under the coating layer due to aqueous solution transmitted throughout the small pinhole or flaw of the coating. The causes of the improvement in the corrosion fatigue resistance of TiN coated material were elucidated that the TiN layer does not peel off from the substrate surface owing to its enough adhesion and corrosion resistance of the layer itself is more excellent than that of the substrate metal.
Modified 9Cr-1Mo steel is now being used for high-temperature structural components on a world wide scale because it has an extremely high allowable stress. In the design of high-temperature equipments, the prevention against creep-fatigue failure is an essential problem. However, the method of creep-fatigue life prediction has not been clarified for modified 9Cr-1Mo steel. In the creep-fatigue design of structural components, an inelastic analysis is often used when their service conditions are severe. And in such case, the estimated life is affected not only by the accuracy of a life prediction method but also by the accuracy of an inelastic analysis. Therefore, both of these accuracy should be taken into consideration to establish a rational creep-fatigue design method. In this paper, the method of life prediction by the linear damage rule for modified 9Cr-1Mo steel was investigated in the viewpoint of both the inelastic analysis and the damage evaluation. The results show that the effect of loading history should be considered both in stress-strain properties used for inelastic analysis and in creep-rupture properties used for creep damage evaluation in order to make a rational prediction.
Two-step PP/CP and CP/PP variable straining tests were conducted for Mod. 9Cr-1Mo Steel at 600°C in air to clarify the small crack growth curve in CP test, where the strain range was kept constant and the strain waveform varied from PP to CP (PP/CP tests) or from CP to PP (CP/PP tests). It was found that in PP/CP tests the sum of life ratios was smaller than unity when the primary cycle ratio n1/N1 was large. On the other hand, in CP/PP tests the sum of life ratios was larger than unity in the large n1/N1 regime. The smaller the straining level, the larger the discrepancy between the experimental result and the linear damage rule prediction. The small crack growth curve in CP test was determined so that the fatigue properties under two-step variable straining tests could be well explained by the proposed crack growth model based on the strain range partitioning concept. It is suggested that the initial crack length in CP test (a0)pp+cp is equal to the CP type initial crack length (a0)cp when Δεcp/Δεpp≥1 and that (a0)pp+cp is equal to the PP type initial crack length (a0)pp when Δεcp/Δεpp<1.
The low cycle fatigue tests were conducted on Type 304 stainless steel at 500°C and 550°C, where dynamic strain aging easily occurs, and the effect of strain waveform on the low cycle fatigue life was studied. As the strain waveforms, the triangular, slow-fast and fast-slow waveforms under the total strain range Δεt=1.0 or 0.5% with a tension going strain rate of 1 to 0.001%/s and compression going strain rate of 1 to 0.001%/s, were used. The results of the present study are summarized as follows; (1) The negative strain rate dependence due to dynamic strain aging was observed at the strain rate more than about 0.01%/s to 0.005%/s under both strain range Δεt=1% and 0.5% at 550°C. A similar tendency was also observed for the thermal aged material (650, 700°C×500-3000h, Δεt=1%), although hardening due to dynamic aging was low. (2) The low cycle fatigue life was approximately dominated by the tension going strain rate. The increasing of intergranular fracture with decreasing the strain rate was caused by the creep effect in tension going period rather than the dynamic strain aging. (3) The thermal aged material (700°C×3000h) had a longer fatigue life than the solution heat treated material, because the stress in tension was reduced due to the decreased dynamic strain aging.
Creep-fatigue tests were carried out on notched plates under cyclic bending loads out of plane at 550°C, and the local strain at the notch-root and micro crack propagation behavior were measured. Then, inelastic analysis was performed for the experiment by using three kinds of constitutive models, such as kinematic hardening, ORNL and Ohno models. From the comparison of the experiment with the results of analysis, the following conclusions were obtained. (1) Creep strain caused at the notch-root during load holding was negligibly small compared with plastic strain, so that the neighborhood of the notch-root is subjected to constrained strain type damage. (2) The strain range at the notch-root can be calculated from the results of elastic-plastic analysis for monotonic loading independent of the constitutive models used, where the cyclic stress-strain relationship was used as the material monotonic deformation property. (3) The mean strain calculated was consistent with the experimental value in case of kinematic hardening or ORNL model, while not in case of Ohno model. (4) A method for predicting the crack initiation life of a notched plate has been proposed on the basis of micro-crack propagation behavior obtained by a fundamental creep-fatigue test.
Recent thermoplastic composites have the potential to offer improved environmental tolerance, damage tolerance and decreased manufacturing cost compared with thermoset composites. However, it is difficult to impregnate the melted thermoplastic resin into reinforcement strands due to high viscosity of resin. Therefore, the laminate composite materials in which fiberous thermoplastic resin is coupled to the reinforcement strands have been developed. It is expected that the adhesive condition in micro interface of coupling laminate will be better, because there is thermoplastic resin in the orthogonal interface of reinforcement stands before the resin melt. The purpose of this study was to estimate the adhesive condition of micro interface of the coupling laminate through immersion testing of the film stacked laminate under hot water environment. After immersion tests of water absorption, the mechanical properties, flexural modulus retention and flexural strength retention were determined. The degradation behaviors of coupling laminate and film stacked one were simulated by the finite element method with consideration for the degradation of constituent materials. It was found that the material constants depended on water absorption under hot water environment by material testing and they were predicted quantitatively from classical Fick's diffusion equations. The numerical results were in good agreement with those of experiments under flexural conditions.
Itaya natural zeolite and its cation-treated zeolite were characterized by X-ray diffraction, thermo-gravimetric analysis, adsorption and chemical analysis. No change of X-ray diffraction pattern was found for Itaya zeolites treated with 1mol·dm-3 hydrochloric acid but a slight change of SiO2/Al2O3 ratio was found. The surface of Itaya zeolite seem to contain both meso- or macropores, but its heterogeneous surface became homogeneous with hydrochloric acid treatment, being predominantly composed of micropores. The monolayer adsorption capacity of zeolite for nitrogen gas decreased as the amount of cations in the Itaya zeolite increased. The cation (Na+, K+, NH+4, Ca2+) exchanged Itaya zeolite samples had less nitrogen adsorbing power than the hydrochloric acid-treated zeolite. The gas separation behaviour of Itaya zeolite was compared with that of the cation-treated zeolite. Separation of hydrogen, oxygen, nitrogen, methane and carbon monoxide from the surface of zeolites was investigated at 30-100°C by using Itaya zeolite and its cation-treated zeolite as column packings. Hydrogen, oxygen, nitrogen, carbon monoxide and methane were found to be separatable completely from the mixture gases by using Itaya zeolite, hydrochloric acid treated zeolite and its K+ exchange zeolites. A methane-oxygen mixture could not be separated by Na+, NH+4 and Ca2+ exchanged Itaya zeolites.
In the preceding paper, epoxy resins (EPIKOTE828 and 154) were cured with 1, 3, 5-trichloro-1, 3, 5-tridimethylamino-cyclotriphosphazene (C13). These epoxy resins exhibited high Young's modulus (30°C-Tg) and high elongation (around Tg) in tensile tests, and high weight remaining (500-700°C) in thermo-gravimetric analysis. In this paper, the chemical resistance of the epoxy resin cured with C13 has been investigated. As for the dimensional stability, the cured EPIKOTE828 was severely damaged by the immersion in H2O 60°C and 36% aqueous solution of HCl 25°C. For the cured EPIKOTE154, the change of dimension was 2.9-7.4% after the immersion in H2O 60°C and 0.2-1.2% in 36% HCl 25°C. Both of the cured EPIKOTE828 and 154 were rather stable in 38% aqueous solution of NaOH 25°C and 60°C, and the dimensional change was below 1%. In the vibron measurement of the cured EPIKOTE828, E' decreased remarkably by the immersion in H2O 60°C and the peak temperatures of E" were observed at 90°C and 140°C. The degradation by 38% NaOH was larger for EPIKOTE828 than for EPIKOTE154, and was larger at 60°C than at 25°C. In the torsion pendulum measurement, the cured EPIKOTE154 immersed in H2O 60°C showed almost the same rigidity as before immersion at the temperature range above 150°C. In conclusion, the epoxy resin cured with C13 is degradated by acid and is resistant to alkali.
It has been found that properties of fresh cement paste are remarkably changed by devided addition and subsequent devided mixing of water for the same mix proportion (this method has been referred as double mixing.). It is strongly suggested that some surface chemical interaction between cement and water is related to this phenomenon. But this has not been elucidated completely yet. In this paper, mechanism of these double mixing effects and influences of double mixing method on bleeding and rheological property of cement paste were experimentally investigated. From these experiments it was suggested that the double mixing effect is not only associated with physico chemical interaction between water and cement powder but also with chemical reaction of cement particles with water and electrostatic charge (zeta potential) of cement particles that might be related to the early stage chemical reaction of cement with water.
This paper deals with a kinetic analysis of the bleeding of mortar and concrete mixtures by applying Darcy's law for the capillary kinetics. Mortar and concrete mixtures with desired mix proportions were mixed at room temperature, and the exuded volume of the bleeding water from the mixtures and its absorbed volume were continuously measured during hardening by using an apparatus developed by the author. In conclusion, several linear relationships between the exuded and absorbed volume of the bleeding water and elapsed time after mixing of the mortar and concrete were obtained. In general, the exudation process of the bleeding water from the mortar and concrete mixtures consists of four and three stages, respectively. The absorption behavior of the bleeding water into the mixtures may be divided into two stages.
When constructing large caverns such as the one used for an underground powerhouse, it is of extreme importance to assure the stability of the cavern. As one of the techniques for stability monitoring, uniaxial compression tests were performed in various rocks and AE activity was measured. From the examinations of AE activity and the maximum amplitude of a single event, two types of AE occurence patterns and three types of m-value variations were found. When predicting failure modes by means of AE, it is thought that the total AE counts and the m-value can be useful for non-homogeneous rocks. On rocks of schistose or very homogeneous structures, the total AE counts provide useful information on failure prediction but the m-value does not. Furthermore, from the results of spectral analysis of AE signal waves measured, it is found that the dominant frequency increases with increasing load.
In order to elucidate the relation between hydrogen embrittlement and surface roughness, slow strain rate tension tests were carried out under various hydrogen evolution potentials, -0.1--2.0V (vs. S.C.E.), on ferritic stainless steel SUS430, surface roughness of which is varied in there sizes, 0.75, 3.5 and 5.0μm. The results obtained are summarized as follows: (1) The reduction in area and the time to fracture, which are the indicators of the degree of embrittlement of specimen, decreased at low hydrogen evolution potentials. The similar behavior was also observed when the specimen was held for longer time at each hydrogen evolution potential. (2) When the specimen surface roughness was large, the reduction in area and the time to fracture decreased at all setting potentials. The influence of specimen surface roughness on the embrittlement phenomena is summarized in the order of specimen A (0.75μm)<specimen B (3.5μm)<specimen C (5.0μm) because the reduction in area becomes smaller and the time to fracture becomes shorter in this order.
This study was intended to elucidate the behavior of minute creep strain in ferritic stainless steel SUS430 specimens which were subjected to various states of hydrogen evolution environment. Creep tests were carried out under various hydrogen evolution potentials and the results were compared and examined. The results obtained are summarized as follows. (1) From the comparison of creep curves obtained under various setting potentials with those in atmosphere, the creep strain was found to become smaller with lowering setting potential and the increase of hardness in the surface vicinity layer was observed. The creep curves obtained under different setting potentials were aligned in the order of atmosphere, -1.0V, -1.5V and -2.0V(vs. S.C.E.). (2) The anomalous phenomenon in which minute creep strain increases with a wave motion appeared after testing for 104sec. in all setting potentials. When the applied load exceeded 160kgf of the yield point, this anomaly appeared in large numbers and the value of wave motion became 0.4% or more in the minute creep strain. The minute creep strain increased by repeating this behavior. Therefore, even if the applied load is a yield point load at which the specimens don't fracture immediately, it may lead to fracture in a short period.
An ultrasonic experiment has been carried out to evaluate the accumulative creep damage which occurred in polycrystalline pure copper during high-temperature tensile loading. The shear wave velocities exhibited birefringence which arose from the macroscopic anisotropy due to intergranular creeping process controlled by grain-boundary cavitation and subsequent microcracking perpendicular to the stress axis. The velocity anisotropy, which is independent of specimen thickness, decreased very slowly and linearly up to approximately 80% of the time to fracture, and then decreased at increasing rate until the eventual failure. The first period was expended by the nucleation and growth of cavities and the second corresponded to the cavity coalescence and the cracking. Measurements of porosity and metallography supported the ultrasonic observation. The results gave basic information about nondestructive characterization of accumulative creep damage in high-temperature components.
An ultrasonic techique for the evaluation of creep cavities in steel for high temperature use is proposed. By using samples taken from crept or ruptured specimens, longitudinal wave velocities in the parallel and normal propagation directions to the stress axis (VLP and VLN) were measured. Then the shear wave velocities both in the parallel propagation direction to the stress axis (VSP) and in the normal propagation direction for parallel and normal polarizations to the stress axis (VSNP and VSNN, respectively) were measured. These results showed that all of the ultrasonic wave velocities decrease linearly with increasing amount of creep cavities. Wave velocities were also sensitive to the directions of the propagation and the polarization. This sensitivity suggests the usefulness of VLP/VSP, VLN/VSNN and VSNP/VSNN as new parameters. The values of these three parameters were found to decrease linearly with an increase of the amount of creep cavities, and the comparison among these parameters showed that the ratio VLP/VSP was most sensitive to the amount of creep cavities. The evaluation method of the amount of creep cavities using the parameters of ultrasonic wave velocity may be useful for practical purposes because it does not require the propagation path length.
A theoretical study was made on creep damage detection by ultrasonic wave measurement. The evaluation method was developed in order to predict the ultrasonic attenuation and velocity changes with increasing creep damage. The method was obtained from the results combining the following theories and experiments: (1) Ultrasonic attenuation of a void-containing medium, (2) the elastic constants of a solid containing spherical holes. (3) damage rate approach, and (4) creep damage assessment by density change measurement. By the developed method, it became possible to describe approximately the attenuation and velocity changes with increasing creep damage.