The X-ray diffraction method was used to observe the material substructure due to plastic deformation near the fatigue fracture surface of low-carbon steel. The fracture surface was made under a stress-intensity-factor controlled condition by using a computer-controlled servo-hydraulic fatigue testing machine. The distribution of the half-value breadth of the diffraction profile beneath the fracture surface was determined. The half-value breadth took the maximum value at the border of reversed plastic zone, while it approached to the initial (pre-fatigue) value near the boundary of the monotonic plastic zone. Microstructural parameters derived from computer image analysis of X-ray microbeam patterns were also usable to determine the sizes of the reversed and monotonic plastic zones. The distributions of X-ray parameters in the monotonic plastic zone were controlled by the maximum stress intensity factor, and those in the reversed plastic zone by the range of the stress intensity factor.
An X-ray fractographic study was made on the fatigue fracture surface of HT80 and S53C steels at elevated temperatures up to 400°C. Emphasis was put on examing the effects of oxide film and fracture surface roughness on the residual stress and half-value breadth of fracture surface at elevated temperatures. It was found that the oxide film induced on the fracture surface at elevated temperatures was thin enough and was not detrimental for the application of X-ray fractography to analyze the fracture surface of steels up to 400°C. However, it was also found that the increased fracture roughness beyond a certain value was detrimental. The value was estimated as about 7.5μm in the center line average roughness. For this case, it is difficult to apply X-ray fractography because both the residual stress and half-value breadth are controlled not only by Kmax or ΔK but also by the roughness of fracture surface.
Fatigue crack propagation tests of a low alloy steel (JIS SNCM 439) tempered at 200°C and 600°C were conducted both in air and in 3.5% NaCl solution. The residual stress near the fatigue fracture surface was measured by the X-ray diffraction method. The results obtained are summarized as follows: (1) The residual stress measured on the fracture surface was tension both in air-fatigue and corrosion fatigue. The tensile residual stress increased with the maximum stress intensity factor Kmax in the case of the material tempered at 200°C, while it had a maximum value at about Kmax=30MPa√m in the case of the material tempered at 600°C. When compared at the same Kmax value, the residual stress was lower for a lower stress ratio and in corrosion fatigue. (2) The distribution of the residual stress beneath the fatigue fracture surface was able to be decomposed into two components: the tensile residual stress in the vicinity of the fracture surface caused by monotonic tensile plastic deformation, and the compressive residual stress in the vicinity of the fracture surface caused by stress relief due to roughness and by compressive plastic deformation. (3) The maximum depth of the plastic zone was evaluated on the basis of the residual stress distribution. The depth ωy is related to Kmax by the following equation: ωy=α(Kmax/σY)2 where σY is the yield strength obtained in tension tests. α is 0.19 for air fatigue and 0.06 for corrosion fatigue. The small value of α in corrosion fatigue suggests the hardening of the material in the plastic zone due to the environmental effect.
The employment of aluminum alloy castings for automotive parts is increasing every year, and recently some of them require relatively high fatigue strength. Under these circumstances, the improvement of their fatigue strength is an important problem to be solved. One of the authors has found that it is very hard to improve the strength, especially the fatigue limit of aluminum alloy castings, either by T6 treatment or by changing their chemical compositions. Although numerous literatures show that shot peening increases the fatigue strength of many metallic materials, few studies deal with aluminum alloy castings. In this study, the effect of shot peening on the improvement of the fatigue strength of aluminum alloy castings was examined first. Secondly, the improving factor of the fatigue strength of the shot peened castings was identified through the other experiments with the upset specimens of the castings, and finally an evaluation method for the fatigue limit of shot peened aluminum alloy castings was proposed. The following conclusions were derived from the present study: (1) The fatigue strength of aluminum alloy castings was able to be improved with shot peening. (2) The improving factor is a compressive residual stress at the surface layer resulting from shot peening, and the fatigue limit is evaluated by dealing with the residual stress as the mean stress during cyclic loading. (3) Hardening at the surface layer resulting from shot peening does not contribute to the improvement of fatigue strength in aluminum alloy castings.
Fatigue damage was detected by X-ray, using a position-sensitive detector, and these data were compared with the data obtained by a usual X-ray stress measuring apparatus. It was noticed that the change of half value breadth of diffraction X-ray during the fatigue process depended on stress and materials. The values of change measured by a position-sensitive detector were several times as large as those by a usual X-ray stress measuring apparatus. It is found that this difference is due to the singularity of the change in distribution of diffraction X-ray intensities during the fatigue process.
The WC-Co alloys coated with TiN by chemical vapour deposition process are substantially improved in their wear resistive properties, but are deteriorated in the failure resistance in comparison with those of the same alloys coated by physical vapour deposition process. When the residual stress of WC-Co alloys coated with TiN by CVD or PVD process was measured by X-ray method, the residual stress of the CVD coated alloys was tensile, while that of the PVD coated alloys was compressive. To estimate the effects of these residual stresses on the strength of the TiN-coated alloys, transverse-rupture strength was measured, and the crack sensitivity was also examined from the crack length around a Rockwell hardness indent. Owing to the residual stress of tension in the CVD coated alloys, the crack sensitivity increased and the fracture strength decreased. When the PVD coated alloy was annealed at a temperature above 1173K, the residual stress of compression changed to tensile stress and the crack sensitivity increased. To make clear the cause of the change of these residual stresses in TiN-coated alloys, X-ray diffraction patterns were measured by use of Co Kα radiation monochromated by a graphite monochrometer, and furthermore, the crystalline states of the coated layer, the substrate and the interfacial phases were investigated by EPMA.
The X-ray diffraction method was used to measure the residual stress in the surface layer of sintered alumina finished under various conditions. X-ray elastic constants were determined for the diffraction lines of (1.0.10) plane and (220) plane obtained by using CrKα radiation. Those values were nearly identical for the as-fired surface, ground surface and lapped surface. The measured residual stresses were all compressive for various surfaces. The residual stress was the maximum compression of about -140MPa for the surface ground with #300-diamond wheel. The surfaces finished by #600-diamond grinding and lapping were about -90MPa. The residual stress measured on the fracture surface was small compression. Both the full width at the half maximum of diffraction profiles and the feature of X-ray microbeam diffraction patterns changed depending on the amount of plastic deformation associated with surface finishing and fracture processes.
In the field of X-ray stress measurement of polycrystalline materials, a diffraction plane at higher Bragg angle has to be selected in order to obtain the precise value of stress. However, the stress measurement on an optional (hkl) plane desired is not always possible because the X-ray beam exited from a metal target has a dispersive wave length. Recently, we have been able to use the synchrotron radiation source (SR) as an excellent X-ray source. In Japan, the facility of synchrotron radiation (Photon Factory, PF) was constructed in the National Laboratory for High Energy Physics (KEK) at Tsukuba academic city. The use of this SR enables the stress measurements on many (hkl) planes with high accuracy in the higher Bragg angle region by providing an X-ray beam having an optional wave length. We have started the X-ray stress analysis by use of the synchrotron radiation source. This paper reports the system of measurement and some results of preliminaly experiments. Since a monochromatic X-ray beam is required for the stress measurement, we used a beam line which consists of a double crystal monochrometer and a focusing mirror. X-rays between 4KeV (λ=0.31nm) and 10KeV (λ=0.12nm) are available with this optical system. We adopted a constant Bragg angle of 2θ=154° for all the diffraction planes. A PSPC having a carbon fiber anode is made and used as a detector with the use of a fast digital signal processor. We could observe the diffraction profiles from (200), (211), (220), (310) and (321) crystal plane of alpha iron, respectively, and the residual stresses in these planes except the (200) plane were measured with high accuracy in a short time. Such feature especially suits the stress analysis of the material which has preferred orientation or stress gradient.
Strengthening due to the thermo-mechanical treatment (TMT) was examined in the polycrystalline Cu-7.5at%Ni-2.5at%Al alloy. Specimens were solution treated, aged, cold rolled and finally aged. It was found that their mechanical properties were much improved not only at room temperature but at high temperatures up to 500°C, comparing to those simply aged and mechanicothermal treated (MTT: solution treated, cold rolled and finally aged). Precipitated particles with Ll2 type ordered structure grown during preageing, were sheared by cold rolling. High density of dislocations was introduced. However, sheared particles played an important role to strengthen the alloy. The first notable effect of TMT is to retard recovery of the cold rolled structure extremely in final ageing. In consequence, the deformed structure was kept after final ageing. The second is the growth of precipitated particles with smaller sized and higher density in final ageing, which were formed from sheared particles and nucleated in the matrix. Excess vacancies introduced by cold rolling would accelerate the growth of particles. Strengthening in the TMT specimens is considered to be attributed to the random dispersion of a large number of small sized particles and the existence of high density of dislocations.
This paper first presents a new type of path independent integral expressing the energy release rate for the crack growth in nonlinear materials. It is path independent even for the non-straight crack and gives the energy release rate without any limiting process; the J-integral is path independent only for the straight crack. It also generalizes the well-known expression derived by Rice, which may be evaluated from the load versus load-point-displacement curves for slightly different crack sizes. Then, as an application of the new expression, a simplified formula under multiaxial loadings, which may be evaluated only from a single loads-displacements curve, is derived in the similar way as Rice et al. developed under uniaxial loadings. Finally, the difference between the J-integral and the new expression, when they are applied to elastic-plastic materials, is examined in some detail.
It is well known that the fatigue crack propagation rate is affected by stress history and thus crack propagation retardation or acceleration will occur. Consequently, many attempts to develop a suitable fatigue crack propagation model under varying loads have been made by various investigators. However, it seems that models developed until now have some problems. In the previous report, the present authors proposed a new fatigue crack propagation model based on the results of crack propagation tests under programmed or random load by using CT type specimens. However, these results were obtained for the case of stress ratio (R) equals 0.1. So, it is posibble to observe such a different crack propagation behavior that retardation is relieved or acceleration is increased by the compressive part of stress range when R equals -1. Accordingly the authors have investigated about this behavior by executing fatigue crack propagation tests under random loading in which stress ratio is fixed to R=0 and R=-1 using CCT specimen. Then a more reasonably improved model than that proposed in the previous report is proposed in this paper. As a result it is made clear that the fatigue crack propagation life under stationally random loading can be estimated accurately enough by the linear model when R=-1, and that for the case of R=0 it can be estimated most accurately when the new proposed model is employed.
A usual unloading elastic compliance method for determining the crack opening point is based on an unloading curve whose uppermost part looks like a straight line. By this method there is the possibility of obtaining a wrong conclusion, since the crack closure behavior is always accompanied with plastic deformation. In this paper the tension-compression fatigue tests were carried out (stress ratio were taken as 0 and -1), and the opening and closure point is successfully measured by the S shaped unloading curve. The crack closure point measured from the S shaped unloading curve does not coincide with the opening point. The crack growth rate can be expressed unifyingly by the effective stress intensity range, ΔKeffop=Kmax-Kop, or ΔKeffcl=Kmax-Kcl, independently of the stress ratio and the loading history. The inclination of the line of log(dl/dN) vs. log ΔKeffcl plot is approximately 2. This suggests that the proposed method for determining the crack closure point is rational. It is shown that the values of effective stress range ratio uop(=ΔKeffop/Kmax) and ucl(=ΔKeffcl/Kmax) plotted against Kmax respectively get together on one narrow band (when R is constant), if the cracks were propagated into a steady state. The effective stress range ratios uop and ucl increase linearly with Kmax in the range of Kmax=7.5MPa√m to 12.5MPa√m. Especially, the ucl based on Kcl is nearly in direct proportion to Kmax. This is closely related to the background of the fact that the Paris crack growth low dl/dN=CΔK4 holds good in the present material.
In this paper the model of crack closure is developed to analyse the experimental results of the previous report. The plasticity induced crack closure is simulated by leaving residual stretch in the wake of the advancing crack tip, and the roughness induced or the oxide induced crack closure is modeled by the debris whose thickness is assumed to be constant along the crack surface. The method of analysis used here is the extended body force method. It is shown by the present model that the effective stress range ratio Uop(=1-σop/σmax) or Ucl(=1-σcl/σmax) of fatigue crack in the stabilized state is mainly controlled by the Kmax (when stress ratio R is constant) and that the reversed crack tip opening displacement can be expressed unifyingly by the effective stress intensity range ΔKeffop(=Kmax-Koq) or ΔKeffcl(=Kmax-Kcl), independently of the loading history and the stress ratio. These are in good agreement with the experimental results of the previous report. Moreover, the transitional behavior of the crack closure observed in the experiment of the previous report, when load is changed, can be illustrated as well by this model.
An experiment was performed on galvanic corrosion of cast iron-stainless steel couple in 0.016% NaCl solution, and the distribution of the potential in the electrolyte and the current density across the galvanic couple were measured. The distribution of potential within the electrolyte was described by the Laplace's equation with nonlinear boundary conditions, which were enforced based on experimentally determined polarization curves. This boundary value problem was solved by the boudary element method using the Newton-Raphson iterative numerical procedure. It was shown that the computational results agreed well with the experimental data.
In order to investigate the effect of corrosive environment on fatigue strength, rotating bending fatigue tests were carried out on shrink-fitted specimens. The materials of axles tested were SUS 304 stainless steel, SCM430 alloy steel and S45C carbon steel, and those of bosses were SUS304 and SF55 for SUS304 axle, SUS420J1 for SCM430 axle, S45C for S45C axle. The tests were carried out in air, city water and 3% salt water. The fatigue strength reduction factor of shrink-fitted specimens in 3% salt water was approximately 2 for SF55+S30C, 3 for S45C+S45C, and 4 for SCM 430+SUS420. It was considered that the corrosion fatigue strength (CF) of shrink-fitted specimens depended on CF of these axles because CF of shrink-fitted specimens was affected by the sensitivity to corrosion of axle. CF of SUS304+SUS304 specimen increased more than the fatigue strength in air. In SCM430+SUS420 specimen fatigue fracture occurred at the plane part of specimen in city water and CF in salt water reduced to about 1/4 of that in air. It was concluded that the reduction of CF was due to galvanic corrosion, stress concentration and fretting near the shrink-fitted end.
Josephson-Junction device is expected to be applied as a highly efficient computer switching device. Such a device is usually subjected to thermal cycles from room temperature in manufacturing and repairing stages to absolute 0K in actual use. Due to these thermal cycles hillock can be initiated on the base electrode, and micro bridge may be formed between the upper and lower electrodes. In this study, many typical hillock formations in Pb-In alloy films were observed by a scanning electron microscope. An equation for the hillock growth rate, dSh/dN, was derived experimentally as a function of film thickness, h, and temperature change amplitude, ΔT, as follows. dSh/dN=0.05(h-0.3)0.7+0.0013h0.8ΔT where Sh is the hillock area, and N is thermal cycles. When the value of h was larger than 0.6μm, hillock was initiated at the first thermal cycle independent of ΔT. When the value of h was less than 0.6μm, the critical thermal cycles of hillock initiation increased with decreasing film thickness, h, and temperature change amplitude, ΔT. There may be the threshold region of hillock initiation under certain critical values of film thickness and temperature change amplitude. On the other hand, hillock was not initiated easily for Au-(Pb-In) alloy film.
It is significant to study the difference between thermal fatigue behavior and high temperature low cycle fatigue, in order to understand the fracture mechanism of metals. In this study, thermal fatigue tests (the upper temperature 800°C, the lower temperature 400°C) and low cycle fatigue tests at 800°C were carried out for super alloy IN738LC, and the initiation of a short crack and its propagation behavior were observed. The results of these fatigue tests showed that the difference in fatigue life between them were caused by the differences in crack initiation life and crack propagation rate during the time of the short crack (2a≤2.0mm). On the other hand, it was observed that the fatigue crack propagation behavior was affected largely by the fracture resistance of grain boundary and cell boundary in crystal grains at high temperatures.
Changes of thermal property and flexure strength of CFRP, which was fatigued with heat cycle by using a newly-made heat cycle testing equipment, have been investigated systematically by varying the condition of degradation. In CFRP, three peaks were observed in the damping curve when subjected to fatigue with heat cycle. The changes of damping curves and fracture surfaces suggest that the matrix region of CFRP may be classified into three groups, i.e. the regular matrix region, the surface region of fillers and the stratified region which surrounds fillers. The degradation of CFRP with heat cycle fatigue seemed to occur from the degradation of the regular matrix region. The degradation of CFRP causes the decrease of the first peak of damping curve and the lowering of flexure strength. The damping curves of other two regions did not change with fatigue. It is concluded that the existence of other two regions, i.e. the surface region of fillers and the stratified region which surrounds fillers, plays a major role against the durability for fatigue and the retention of strength of CFRP.
There have been many studies on the impact properties of carbon-fiber reinforced plastics composite materials (CFRP). Most of them are concerned with the evaluation and improvement of impact properties of CFRP, because the fracture strain of carbon-fiber (CF) is very small and its brittle nature is a matter of inquiry. In this paper, high speed impact tests using foreign projectile were performed on CF/GF hybrid composites and CFRP materials, which are composed of laminate plates, and the impact properties and fracture patterns of these materials were investigated. The results obtained showed the hybrid effectiveness of CF/GF composites, that is, the increase of impact fracture energy and the change of fracture pattern of these materials. Especially the mixing of GF materials into CF laminated plates was found to cause the enlargement of whitening area damaged by foreign projectile and thus the increase of impact fracture energy of these materials.