The effects of gypsum and an additive (Mighty 150 from Kao Soap Co.) on the time dependence of the apparent viscosity ηa of cement paste were studied by the use of a coaxial cylinder rheometer. The apparent viscosity decreased first and then remarkably increased. The decrease in apparent viscosity is related to the breakdown of aggregation structure formed by hydration reaction, and hence its initial value varies with the rest period for cement paste without gypsum. In the presence of gypsum the time dependence of the apparent viscosity was not affected by the rest period, because gypsum retards the hydration of C3A(3CaO·Al2O3) and C4AF(4CaO·Al2O3·Fe2O3). Moreover, gypsum had a tendency to reduce the rheopectic characteristics. The addition of Mighty led to a significant decrease in apparent viscosity.
Hastelloy XR has been developed in Japan Atomic Energy Research Institute to improve the corrosion resistance of Hastelloy X in VHTR helium environment. To evaluate weldability of Hastelloy XR, bead welding tests by EB welding and Trans-Varestraint tests by TIG welding were carried out. The results obtained are as follows. (1) The weld defects formed in the bead welding test by EB welding were R-porosity and microcracks, both of which are the same as in Hastelloy X. R-porosity and microcracks can be prevented by using a proper ab value and by controlling the weld heat input, respectively. (2) In the Trans-Varestraint test by TIG welding, susceptibility to weld cracking was the same for both Hastelloy XR and Hastelloy X. To eliminate microcracking in TIG welding, it is effective to decrease the welding current.
The typical plastic flow that occurs under repeated passages of load by rolling contact, was reproduced by the use of a two-disk wear testing machine. As for the modes of cyclic loading, pure rolling contact, rolling contact driven by frictional force and rolling-sliding contact were applied. Residual stress was measured by means of X-ray diffraction with Dölle's analyzing method (8). The changes in micro-Vickers hardness and half-value width of X-ray diffraction pattern along the depth direction were measured, and the following results were obtained: The major residual stress components σx and σy were compressive, and others were relatively small, less than ±50MPa. The value of σx decreased from the surface layer to the depth direction showing maximum in the region where plastic flow is observed, while σy showed a maximum value at the surface and decreased along the depth direction. The distributions of these residual stress components along the depth direction were consistent with the analytical calculation by the method of Merwin et al. (6). In the case of rolling-sliding contact, the additional residual shear component τzx was observed. In the plastic flow zone caused by tangential force at the surface, marked work hardening and broadening of half-value width were observed. In the region of plastic flow caused by Hertzian stress, however little work hardening effect and minor broadening effect of half-value width were found.
In order to clarify the effect of the fiber/matrix interfacial bond strength, the fiber/matrix ratio and the condition of heat treatment were changed and the temperature dependence of the strength of roving glass cloth FRP in the flatwise direction was investigated. Although the roving glass cloth FRP has the fiber/matrix and the FRP/FRP interfaces, the weakest interface in the present study was the fiber/matrix interface irrespective of the fiber/matrix ratio and the condition of heat treatment. The results obtained are as follows. (1) The temperature dependence of the brittle and the ductile behaviors observed in the load-displacement curve was changed by the fiber/matrix ratio irrespective of the condition of heat treatment. The ductile behavior was also changed by the thermal properties of adhesive. (2) The temperature dependence of the location of the weakest interface was affected by the fiber/matrix ratio, but not by the condition of heat treatment. (3) The temperature dependence of the bond strength was changed by the fiber/matrix ratio as well as by the condition of heat treatment below the glass transition temperature (Tg) of adhesive, but the effect of the condition of heat treatment differed with the fiber/matrix ratio above Tg. (4) The bond strength for the samples with the fiber/matrix ratio of 100/0 and 0/100 showed a similar change irrespective of the condition of heat treatment, but the bond strength for that with the fiber/matrix ratio of 50/50 was affected by the condition of heat treatment.
It was previously found that material removal in orthogonal cutting of ceramics was characterized by a trajectory of a crack: it unstably extended downwards by some distance from the region near the tool tip, after a pause re-extended, and finally turned upwards to emerge out at the free surface. The crack trajectory was revealed to be considerably affected by the position of crack initiation and the magnitude of the material fracture ahead of the tool rake face during cutting as well as property of the material. This paper deals with the effect of material plasticity on the position of crack initiation and the role of the material fracture to the trajectory of the crack extension. The finite element method was employed to derive the distributions of stress in models corresponding to those at the incipient stage of orthogonal cutting. The main results obtained are as follows. (1) A macro crack in an elasto-plastic material initiates near the elastic-plastic boundary in a region ahead of the tool tip and extends both backwards and forwards following a stress trajectory of the minimum principal stress. The backward extention terminates at a position above the tool tip, while the forward extention trends downwards so far as the material fracture does not occur. In the case that the material fracture takes place in the whole region above the extended crack, the crack turns upwards to emerge out at the free surface. If the material fracture is restricted in a region smaller than the above, however, the crack extends downwards after it does upwards for a short distance. (2) A macro crack in an elastic material initiates near a region ahead of and below the tool tip. Before the material fracture occurs, it extends in a manner similar to that in the elastoplastic material, the end of the backward extention being at a position below the tool tip. Even if the material fracture takes place, the crack continues to extend downwards except that it is arrested by removal of the whole region above the extending crack.
The fracture toughness tests of ductile cast iron (JIS FCD 60) were conducted at room temperature by using compact tension specimens with blunt-notches. The method to determine the point of crack extension was established and the effect of notch-tip radius on the fracture toughness Ki was investigated. The main results obtained are summarized as follows: (1) The DC electrical potential method can be applied to detect the point of crack extension in blunt-notched CT specimens of ductile cast iron, which is almost identical to that determined by the R-curve method. Acoustic emissions became severe before the crack extension point was detected by the electrical potential method. (2) The relation between Ki and notch-tip radius ρ can be divided into three regions: region I where ρ is smaller than the radius of spheroidal graphite ρG(30μm), region II where ρ is above ρG and below 2mm, and region III where ρ is larger than 2mm. Ki was constant in regions I and III, while it increased linearly with √ρ in region II. The plastic strain εf on the fracture surface estimated by X-rays showed a similar trend in its variation with ρ. In region II, εf became larger with increasing Ki/√ρσY.
The elastic-plastic fracture toughness JIc value and fracture resistance of A533B-1 steel were evaluated using compact type (1/2CT, 1CT and 2CT) specimens. The results obtained are summarized as follows: (1) According to the standard method of the Japan Society of Mechanical Engineers (JSME), the JIc value can be evaluated accurately by using the smaller size of specimens compared with the ASTM method. (2) The valid JIc value becomes constant regardless of stress states or yielding conditions, when J integral is evaluated from a load-load line displacement curve. (3) As long as the flat fracture preceeds in the midthickness of the specimen, the tearing modulus, TJ(Δamax), evaluated from the J-Δamax resistance curve, becomes a material constant for a wide range of Δamax. (4) Before the initiation of shear fracture, the following equation can stand; Δaave=0.38Δamax Within this region, TJ(Δaave) evaluated from the J-Δaave resiatance curve becomes constant in appearance. However, as the area of shear lips increases, TJ(Δaave) decreases rapidly.
The fracture toughness tests of ductile cast iron (JIS FCD 60) were conducted at ambient and low temperatures by using compact tension specimens and three-point bending specimens with fatigue pre-cracks. The distribution of the half-value breadth of X-ray diffraction profiles beneath the fracture surface was determined. Based on those results and electron fractographic observation, the mechanism of fracture in ductile cast iron was discussed as a function of test temperature. The main results are summarized as follows: (1) The size of the plastic zone ωy determined from the half-value breadth distribution is related to the fracture toughness Ki through ωy=0.13(Ki/σY)2 where σY is the yield stress. (2) The fracture toughness decreased with lowering temperature and the fracture mode of the matrix changed from dimples to cleavage. The unique relation was established between the fracture surface strain and Ki2/σY.
The elastic-plastic fracture toughness, JIc, and the stable crack growth characteristics of an aluminum alloy, 5038-O, were evaluated using center cracked tension specimens. To investigate growth and coalescence of voids, a fractographic approach was attempted. The results obtained are as follows: (1) The critical stretched zone width, SZWc, becomes constant regardless of the specimen thickness, B, and the initial crack length, a0, and it agrees with the SZWc obtained using compact (CT) specimens. So, the JIc value determined using a subcritical stretched zone width, SZW, versus J blunting line is identical to the JIc of CT specimens. (2) The relation between the net section stress, σnet, and the crack extension, Δa, is not influenced by B nor by a0. (3) Regardless of the macroscopic fracture modes (the flat fracture or the slant fracture), the stable crack growth preceeds in the midthickness of the specimen.
The influence of microstructure of metal on the fatigue crack propagation has to be clarified to improve the fatigue strength of metal. A modulated pattern method to study the micro fatigue crack propagation behaviour was developed, by which every fatigue striation on the fracture surface can be numbered, and the roles of grain boundaries, inclusions and other microstructures were discussed using a precipitate hardening aluminum alloy. Grain boundaries and sub-grain boundaries often worked as resistance for the fatigue crack propagation. Inclusions etc. did accelerate or resist the fatigue crack propagation depending on their mechanical properties. No reverse direction crack growth caused by preceded cracking has been observed.
Rotating bending fatigue tests in a high cycle region have been performed on smooth cylindrical specimens of low-carbon steel (S15C) and austenitic stainless steel (SUS 304) which have different slip characteristics. Many surface cracks were successively observed, and their states of initiation and distribution as well as their behaviors of coalescence and propagation were investigated. The main results obtained are summarized as follows; (1) More surface cracks initiated as the stress amplitude became higher, and they grew individually with coalescence in each other and propagated to failure. The propagation of cracks by coalescences was remarkable in S15C, but rare in SUS 304. (2) In the moment of coalescence, the crack grew at a jump, and for a while its propagation rate reduced transiently after that. However, the rate approached that of the previous independent crack as progression of fatigue. (3) The limit condition of coalescence was expressed as h=C1l1(l1≤l1cr), h=hcr(l1>l1cr), where, l1 is the longer crack length in the adjacent two cracks and h the spacing between them, and the values of l1cr, hcr and C1 were 1.5mm, 260∼300μm and 0.18, respectively, for S15C. (4) The average aspect ratio of depth (b) to length (l), (b/l) of surface cracks were 0.36 and 0.33 in S15C (for l≤5.0mm) and SUS 304 (for l≤10.0mm), respectively. (5) The relation between the propagation rate of main cracks and the stress intensity factor parameter was shown as dl/dN=C(σa√πl)m.
Propagation of microscopic fatigue cracks by periodic overstressing was studied with a low carbon steel. Significant acceleration of crack propagation (more than one hundred times) occurred also in microscopic cracks similarly as in the case of macroscopic cracks previously reported when the understress values were equal to the threshold stresses. There was an appreciable effect of microstructure in microscopic cracks. When the understress values were below the threshold stresses, significant acceleration did not occur in microcracks smaller than 200μm whereas it did in cracks larger than this size. These results indicate that microcracks propagate faster under periodic overstressing than under steady alternating stresses once they started to propagate, and this would be a point that should be remembered in practice.
Fatigue crack growth characteristics in type SUS 304 stainless steel shielded-metal-arc weldments and A533 Grade B Class 1 steel submerged-arc weldments were studied using CT specimens at room temperature. The growth rates in SUS 304 stainless steel weldments were also examined at-100°C to evaluate the effect of low temperature on the growth rate of weldments. The low temperature was achieved by changing the flow quantity of liquid nitrogen. Crack closure of the weldments during fatigue crack growth was measured by a compliance method. An electron fractographic analysis was employed to determine the mechanism of crack growth and to measure striation spacings. The resistance to crack growth in the SUS 304 stainless steel weldments was considerably higher than that of the base metal due to the presence of high compressive residual stress at the crack tip in the CT specimen. The growth rates in the stress relieved weldments in A533B steel and the base metal fell within a relatively narrow band of scatter. Therefore, the growth rates of stress relieved weldments were found not to be sensitive to the changes of microstructure or strength of materials. The fatigue crack growth rates in the SUS 304 stainless steel weldments at -100°C were slightly reduced as compared with those of the weldments at room temperature. The crack growth mechanism in the weldments of both steels was controlled by striation formation, even at -100°C. The fatigue crack growth rates and striation spacings in the base metal and the weldments with residual stress at different temperatures were closely correlated with ΔKeff based on the crack closure measurement.
Retardation of fatigue crack propagation due to a single tensile overload and crack closure behavior were investigated on a high tensile strength steel, HT-80 and an aluminum alloy, A 5083-O by using the unloading elastic compliance method. Although the delayed retardation was observed at the specimen surface where the plane stress condition dominated, retardation of fatigue crack propagation was found to occur immediately after a peak overload at the interior of the specimen where the plane strain condition was developed. In the latter case, the crack tip blunting due to a overload, which might be the reason of the nodelayed retardation, was confirmed by fractography and also by the load-displacement hysteresis curve. Moreover it was found that the retardation behavior of fatigue cracks which initiated and propagated from the root of the blunted crack could not be explained only by the change of the macroscopically measured crack tip opening level.
The effects of tensile prestrain on corrosion fatigue life and crack growth rate in 0.16%C carbon steel were studied under completely reversed plane bending stress in salt water (3.0%NaCl). The conclusions obtained are summarized as follows. (1) The corrosion fatigue strength of the prestrained specimen in the region of high cyclic stress amplitude increased in comparison with that of the annealed specimen and it increased monotonously with increasing prestrain and yield strength of the specimen. On the other hand, the effect of tensile prestrain on corrosion fatigue life disappeared in the region of low cyclic stress amplitude. (2) It was observed by an optical microscope that 80% of cracks were initiated from corrosion pits on the specimen surface under low cyclic stress amplitude. They were independent of the degree of tensile prestrain. The number of cycles to crack initiation, Ni, was not affected by tensile prestrain under the corrosion fatigue of low stress amplitude. In the region of high cyclic stress amplitude, it was found that Ni and the number of cycles during crack propagation increased with increasing tensile prestrain. (3) Micro-cracks on the specimen surface subjected to corrosion fatigue of low cyclic stress amplitude were initiated in the early stage of fatigue life and after that the surface crack density did not change with stress cycling. The growth rate of each surface crack on the specimen depended on the stress intensity factor in the early stage of fatigue life. But in the late stage of life, the growth rate of surface crack per cycle was constant. This constant value varied for each crack and followed normal distribution because of interaction and connection of distributed cracks.
The limiting temperature and time for diffusion welding of 18-8 stainless steel to mild steel were investigated under the conditions of no insert metal, 1kgf/mm2 pressure, 2.0∼5.9×10-5torr vacuum degree and the buff polishing surface, in order to obtain a perfectly welded joint. The effects of the vacuum degree and the surface roughness were also examined under the conditions shown above. And the fractographic features of imperfect welded joints were observed by means of a scanning electron microscope. The results obtained are summarized as follows: (1) Under the above conditions, the limiting temperature and time for perfect welding were approximately 800°C for 7 minutes. (2) The weldability of the joint was reduced as the vacuum degree became lower, and the rate of perfect welding in 0.6∼2.4×10-4torr was about 40per cent. In the case of 10-3 order, perfect welding was impossible. (3) When the surface roughness was over Rmax 1.2μm for SUS 304 and 2.4μm for mild steel, perfect welding was difficult. (4) The fractographic features of fractured bonded interface of imperfect welding were as follows: (a) Mild steel-mild steel welding showed transgranular ductile dimples and cleavage fractures. (b) SUS 304-SUS 304 welding showed only ductile dimples. (c) In mild steel-SUS 304 welding, mild steel was similar to mild steel-mild steel welding, and SUS 304 showed small ductile dimples with precipitated Cr carbide particles, and besides intergranular fracture with carbide precipitation occurred in 1000°C welding.
Thermal expansion of several rocks down to 80K was measured with a dilatometer under dry and wet conditions at atmospheric pressure. According to the results of thermal expansion measurements, the rocks could be classified into two groups: granitic and non-granitic. For the non-granitic rocks in this study, i.e., tuff, sandstone, andesite, and basalt, the thermal expension coefficient increased in proportion to porosity but in inverse proportion to density, Young's modulus, and the strength. On the other hand, for granitic rocks (two granites and quartz-diorite), the thermal expansion coefficient increased in proportion to density and Young's modulus, but decreased with an increase in porosity. These differences are attributable to the pore and crack shape within the rocks. Low aspect ratio cracks are the main cause of granitic rock porosity, and they cancel the thermal expansion of the whole rock because they can provide space for the expansions of the surrounding grains. On the contrary, high aspect ratio pores, mainly within non-granitic rocks, only negligibly affect the thermal expansion of the whole rock. When the rock was saturated with water, the thermal expansion coefficient became higher than that under a dry condition. As the thermal expansion coefficient of ice is several times higher than those of rocks, the ice in pores and cracks contributes to the increase in the thermal expansion coefficient. And in the case of granitic rocks, this is partly because the ice existing in the crack increases its net aspect ratio and reduces the space available for the expansion of the mineral grains. The large and complicated hysteresis patterns of wet rocks seen upon thermal cycling are due to temperature differences between the freezing and melting of water in the rock pores and cracks.