During the hardening process of concrete, many structural defects are produced at the boundary between cement paste and aggregate. It is considered that both the seepage water around the aggregate and the sedimentative phenomenon of aggregate influence remarkably the formation of defects. The present observations using a scanning electron microscope show that the defective boundary consists of many fragile calcium hydroxide crystals.

View full abstract

Structural changes during ageing of Cu-Ni-Al alloys containing various amounts of Ni and Al were investigated by means of transmission electron microscopy. The characteristic features of their deformations were also investigated on single crystals.
The results obtained are as follows.
(1) In Cu-5at%Ni-2.5at%Al (A), Cu-5at%Ni-5at%Al (B) and Cu-7.5at%Ni-2.5at%Al (E) spherical precipitates α₁ (Ni₃Al) nucleated within grains. Lump-like precipitates with the same crystal structure as α₁ existed at the grain boundaries of(A) and (E) alloys.
(2) In Cu-5at%Ni-10at%Al (C) and Cu-5at%Ni-15%Al (D), plate-like β₂ (NiAl) precipitated by ageing at high temperatures. At intermediate and low temperatures α₁+β₂ and α₁ precipitated, respectively.
(3) The relation between CRSS (critical resolved shear stress) and ageing time at 500°C was studied on each alloy. CRSS increased as ageing proceeded and had a peak at ageing of∼500hr.
(4) Lüders type deformation occurred in (A) (B) (C) and (E) alloys during ageing before CRSS had the peak. The characteristics in the slip mode and deformation structure indicated that dislocation would pass through the coherent precipitates α₁.
(5) The characteristics in the deformation of (D) and over-aged (C) alloys indicated that dislocation would by-pass or cross slip to avoid incoherent precipitates β₂.

View full abstract

Constitutive equations of elasto-plastic materials with saturating work-hardening are proposed. The equations consist of a stress rate-strain rate relation proposed by Tokuoka and two evolutional equations which govern the variations of a scalar and a tensor internal variable. They show that the isotropic and the translational work-hardening get saturated when the magnitude of work-hardening is considerably large. The stress-strain relations are formulated theoretically for the cases of uniaxial stress extension and of simple shear deformation. Seven material parameters including two elastic constants are determined to fit the stress-strain relations given by experiments. The obtained equations with determined parameters show a good agreement with the experimental data in the process of loading, unloading and reloading.

View full abstract

The constitutive equations of elasto-plastic materials with saturating work-hardening are incorporated into the finite element method. Two problems, that is, the complex loading with abruptly-changing strain paths and the metal extrusion are numerically solved. For the complex loading, the stress variations along strain paths are obtained which are consistent with the experimental results. The extrusion process is calculated until the steady state is attained and the stress fields in this state are obtained. The residual stress field seems to show the possibility of the surface or internal cracks. For these calculations, the Bauschinger effect and the saturating work-hardening are taken into consideration, which enables more accurate stress analysis by the finite element method.

View full abstract

In the previous studies a series of reloading stress relaxation tests⁸⁾¹⁰⁾ were conducted to clarify the effect of retightening on stress relaxation behavior of high temperature bolts used in steam turbines.
In the present paper, the effect of test temperature on reloading stress relaxation data with a specific reloading time interval was examined on several types of bolting steels; 1Cr-0.5Mo-0.25V, 12Cr-1Mo-1W-0.25V and 19Cr-9Ni-1.4Mo-1.4W-Nb steels. Discussions were made for the relation between the relaxed stress ratio {σ_Ri=(σ₀-σ_ri)/σ₀} and the number of loadings (N) at each temperature (T; 4∼5 levels), where σ₀ is the initial stress, σ_ri is the residual stress, and t_i is the specific reloading time, say 240h. Numerical analyses were made on the present data as well as on the previously obtained ones which concern the effect of time interval (t_i; 4 levels) on the reloading stress relaxation data at a constant test temperature.
These reloading stress relaxation data with various time interval and temperature conditions can be successfully incorporated into the following empirical equation.
σ_Ri=KN^at_i^{b₀+b₁logt_i}exp(c₀+c₁logN/T+c₂/T₂)
where k, a, b₀, b₁, c₀, c₁ and c₂ are constants.

View full abstract

In order to clarify the effect of retightening on stress relaxation behavior of high temperature bolts used in steam turbines, a series of studies on the retightening (reloading) stress relaxation were conducted previously^1)∼4).
Examinations were made, in the present paper, for the effect of test temperature on reloading stress relaxation behavior when stress was relaxed to a specific value of residual stress for 1Cr-0.5 Mo-0.25V, 12Cr-1Mo-1W-0.25V and 19Cr-9Ni-1.4Mo-1.4W-Nb bolting steels.
The relation between the testing time (t_j) and the number of loading (N) at each test temperature (T; 4∼5 levels) was analyzed numerically for the case when the initial stress (σ₀) was relaxed to a specific value of residual stress (σ_rj=0.6σ₀), and an empirical equation was obtained to represent these data at various test temperature conditions. This equation was combined with our previously proposed equation²⁾³⁾ which concerns the data with various specific relaxed stress ratios {σ_Rj=(σ₀-σ_rj)/σ₀; 5∼9 levels} at a constant test temperature.
In order to describe the retightening stress relaxation behavior of high temperature bolts at various temperatures and specific relaxed stress ratios, the following constitutive equation is proposed:
t_j=KN^aσ_Rj^{b₀+b₁logσ_Rj}exp(c₀+c₁logN/T)
where K, a, b₀, b₁, c₀, and c₁ are constants.

View full abstract

Creep rupture properties of 0.12C-22Cr-15Ni series austenitic heat resisting steels which are promising for boiler tubes of a high temperature and high pressure steam power plant have been studied.
The 22Cr-15Ni steel bearing Mo, Nb and Ti has higher creep rupture strength than the simple 22Cr-15Ni steel for a shorter period, but at a higher temperature and for a longer period, creep rupture strength of the former steel has a tendency to decrease owing to the precipitation of σ phase.
The reason for the decrease in creep rupture strength is attributed to the decrease in solid solution hardening. In spite of the precipitation of σ phase, the steel bearing Mo, Nb and Ti is superior to the simple steel in creep rupture elongation, because Mo changes the precipitation morphology so that the initiation and the growth of crack were prevented.
The grain boundary carbide restrains grain boundary sliding, thus the initiation of crack, so that the creep rupture properties of the aged simple 22Cr-15Ni steel are improved.

View full abstract

The effects of strain rate and strain wave form on the low-cycle fatigue behavior of 316 stainless steel at FBR temperature were studied in order to clarify the controlling factor of fatigue strength and fracture mechanism.
The following major results are obtained.
(1) Under symmetrical and asymmetrical straining (slow-fast and fast-slow wave) with the strain rate between 10⁰%/sec to 10^-3%/sec, the fatigue life decreases with a decrease of strain rate in tension going period. The fatigue life is affected only by the strain rate in tension going period, and is not affected by the strain rate in compression going period. Slow-fast wave is most damaging, but the effect of saw-tooth wave is not significant.
(2) The dependence of fatigue life on the strain rate in tension going period may not be due to the creep effect but due to the dynamic strain aging effect proper to FBR temperature. (500°C to 600°C)
(3) The fracture mode changes from transgranular cracking to intergranular cracking with a decrease of strain rate in tension going period. Slow-fast wave enhances the intergranular cracking, whereas fast-slow wave enhances the transgranular cracking.
(4) Thermal aging increases the fatigue life under symmetrical and asymmetrical straining, and the life reduction with the strain rate reduction in tension going period is not so significant for the thermally aged condition. This may be due to the disappearance of the dynamic strain aging effect by the thermal aging. Thermal aging enhances the precipitation of second-phase particles (mainly carbides) at the grain boundary, so that the dissolved carbon in the matrix which causes the dynamic strain aging decreases.

View full abstract

The smooth specimens of 3% silicon iron polycrystals were fatigued under completely reversed bending. The early propagation of a small fatigue crack was studied based on the etch-pit observation of slip deformation and compared with the propagation behavior of a long fatigue crack. Most of small cracks began to decrease the crack propagation rate, departing from the relation between the rate and the effective stress intensity range for long cracks, either when the growth path changed from the grain boundary or slip band to the transitional or non-crystallographic path. The crack growth rate was found to be uniquely correlated to the highly strained region near the crack revealed by the etch pit method. It is concluded that the crack closure is mainly responsible for the anomalous behavior in the relation between the crack propagation rate and the stress intensity range for small cracks.

View full abstract

A study has been made on the extension of an elastic-plastic crack under the condition of anti-plane strain and small scale yielding. Two idealized slip planes, inclined to the crack at angles ±βπ, are assumed to emanate from the crack tip. The crack extension is represented by a branch emanating from the tip in the direction of the original crack. The process of crack propagation consists of two stages. Initially, in the first stage, the length of the slip lines increases with an increase in crack length, since further loading occurs on the slip lines. In the second stage, unloading takes place so that the slip line length remains constant during the propagation. The relevant stress functions were obtained for both stages using the complex variable method. The stress intensity factors were then analyzed. It was found that the fracture load is almost the same as that of a Griffith crack and that for both stages crack extension is unstable, i. e., fracture load decreases as the crack extends.

View full abstract

Fracture toughness tests were carried out with blunt notched compact tension specimens of a high-strength low alloy steel, JIS SNCM 8 (equivalent to AISI 4340), tempered at several temperatures. The stress intensity factor K_ρ at the initiation of crack extension detected by the DC potential method was related to the initial notch-tip radius ρ by
K_ρ=K₀(1+ρ/ρ₀)^1/2
where K₀ and ρ₀ are material constants. This was derived through the concept of the fictitious notch-tip radius proposed by Tanaka and Hirose. The values of material constants K₀ and ρ₀ took the lowest values for the material tempered at 300°C.
The residual stress near the fracture surface measured by the X-ray diffraction technique was tension. The depth of the plastic zone ω_y was measured by the distribution of residual stress beneath the fracture surface. It was related to the stress intensity factor at crack initiation Ki by the following fracture mechanics equation:
ω_y=α(K_i/σ_Y)²
where σ_Y is the yield stress and α is from 0.14 to 0.12. The values of residual stresses very close to the fracture surface tended to decrease with increasing K_i, and were found to be a single-valued function of the fracture strain at the notch-tip determined from K₀ and ρ₀ values based on the fictitious notch-tip radius concept.

View full abstract

3-points bending fatigue tests and microfractographical observations were carried out on spheroidal graphite cast irons in the as-cast (FCD55), annealed (FCD45), or normalized (FCD70) condition. The main results obtained are summarized as follows.
(1) The fatigue strength of as-cast specimens increases with increasing pearlitic matrix and decreasing graphite particles.
(2) In the case of high loading repeated stress, the fatigue strength increases in order of annealed, as-cast, and normalized specimens. The S-N curves cross each other at about 6×10⁵ cycles, and the order of the fatigue strength is reversed at the fatigue limit. It seems that these phenomena are due to the tensile residual stress originated from surface grind finishing of specimens. Their fatigue limits are in the range of 118N/mm²-157N/mm².
(3) The macrostructure observations of fatigue fracture show that ductile fracture occurs in the case of high loading repeated stress, and brittle fracture in low loading stress.
(4) The initial point of crack mostly occurs from a graphite edge or non-metallic inclusion on the loading side of the maximum tensile stress. This result is the same as those other conventional ratary bending fatigue tests.

View full abstract

With broadening of application of adhesive joints for structural elements, many studies have been carried out on fatigue behaviors of various types of adhesive joints. However, investigations of impact fatigue have been made only in a relatively low stress cycles range.
In the present study, a series of impact fatigue tests were conducted on the butt joint specimens bonded with epoxy-polyamide adhesive, where the run-out number of cycles was over 10⁶. In addition, the ordinary reversed push-pull fatigue tests were carried out to obtain the fundamental fatigue data. From the strength behavior plotted on an S-N diagram for these two stress conditions, it was revealed that the impact fatigue strength was higher than the ordinary fatigue one in a relatively low stress cycles range but it decreased rapidly at a certain transitional number of stress cycles despite that the ordinary fatigue strength did not show such a strength trend in the range of this experiment.
Furthermore, the variations of total and inelastic strain ranges of the adhesive layer were determined under these two sorts of fatigue loading patterns to obtain a reasonable material information to interpret the characteristic impact fatigue behavior as mentioned above. It was found that the variations of these strain ranges in the impact fatigue were somewhat different from those in the ordinary fatigue. A detailed discussion was made on these trends. It was confirmed that the cumulative hysteresis energy could play a role of the fatigue failure criterion in the case of adhesive joints used in this experiment.

View full abstract

In order to evaluate the effect of uniaxial orientation on environmental stress cracking (ESC) in crystalline polymers, drawn samples were prepared from low-density polyethylene. The features of ESC in these samples were investigated under methanol environment. From the experiment, the following results were obtained: (1) The fracture angle increased with increasing the draw ratio and the change was distinct at low draw ratio. (2) In the experiment under a constant load, the crack length markedly increased to a critical value with increasing time. (3) The fracture surface was markedly changed with increasing the draw ratio. (4) The resistance to fracture increased with increasing the draw ratio and then decreased.
These experimental data were analyzed by considering the change of the stress distribution around a crack tip with drawing. Then, the following conclusions were reached: (1) The fracture angle is determined by the direction of largest negative hydrostatic pressure. (2) The resistance to cracking in uniaxially oriented samples can be qualitatively predicted from the calculation of the stress distribution around a crack tip.

View full abstract

The mechanochemical reactions of three kinds of FRP (Fiber Reinforced Plastic) were investigated at room temperature up to 50kbar static pressure by using a self-made high pressure apparatus capable of simultaneous shear deformation. All shearing curves showed such a characteristic variation that the shear-strength smoothly decreased after the gradual increase of it, and the higher the static pressure was applied, the clearer the tendency was observed. Although FRP is one of the most stable organic composite materials under these conditions, such a variation of the shear-strength is probably the demerit for practical use. The results of ESR studies together with the above shearing experiment suggest that the mechanochemical reaction of FRP proceeds mainly in the matrix area, and the fracture of the filament fiber itself scarcely occurs.

View full abstract

It is well known that the curved and splitted 2θ vs. sin²ψ distributions are often observed in the surface layers of the uni-directionally machined metallic materials in the case of the X-ray stress measurement. In order to solve this problem, a new X-ray method which provides more informations was proposed previously by Dölle and Hauk.
The present experiments were carried out to determine the influence of cutting speed on the residual strain state, half value width of X-ray diffraction lines, micro-vickers hardness and surface roughness in the cut layer of carbon steel S50C.
The experimental results showed that the changes in residual strain tensor components ε₁₁, ε₂₂ had similar tendency to those in half value width, micro-vickers hardness and centerline average height roughness in the machining direction.
It was also clarified that with an increase in cutting speed, i.e. with a decrease in cutting volume, the residual strain tensor components ε₁₁ and ε₂₂ increased. In some cases, however, the strain increased discontinuously, which was explained by the formation of built-up edge.
These results show that the present method of measuring strain tensor is one of the powerful methods to investigate the machining mechanism and surface layer state.

View full abstract

The expedient estimation method based on photoviscoelastic analysis for residual stress and residual strain generated by quenching both upper and lower surfaces of polyurethane-resin beams has been discussed.
The results obtained are summarized as follows;
(1) The fundamental formulae for residual stress and residual strain, which could be estimated expediently from the residual fringe order observed, were derived by introducing the 4-element photoviscoelastic model.
(2) The residual stress and residual strain calculated by substituting the residual fringe order obtained from simulated calculation, based on linear-photoviscoelastic theory, into the derived formulae corresponded remarkably with those values obtained directly from simulation.
The results obtained here might be useful for the establishment of expedient estimation procedure of residual stress and residual strain in general thermosetting resin moldings.

View full abstract