In order to clarify the anisotropy of fractography at fracture initiation site, monotonic tension, bending fatigue and tensile creep fracture tests were carried out at room and elevated temperatures, using three oriented specimens extracted from a Cr-Mo-V steel turbine rotor. In the case of monotonic tension, although the effects of temperature on tensile properties and fractography was prominent, the effects of specimen orientation on those were quite small at room and 839K temperatures. The fatigue limit of L-R oriented specimen was about 380MPa, and 25% higher than those of C-R and C-L oriented specimens. The scattering of fatigue life of C-R oriented specimen at the near fatigue limit stress condition was clearly larger than those of the other oriented specimens. Under bending fatigue, markedly different fractographic patterns were observed at the crack initiation site for all the three oriented specimens. It was found that the orientation of two kinds of inclusions, MnS and Al₂O₃ particles at the fatigue crack initiation site, influences the fractographic mode strongly. The creep result of L oriented specimen showed the longest life, as compared with those of the other oriented specimens, under nearly the same stress condition. The effect of specimen orientation on the fractographic patterns of creep was quite small.

View full abstract

The effects of hardness and prior austenite grain size on the impact property of hot work tool steel JIS-SKD6 at room temperature were investigated by the analyses of impact fracture surface and of load-deflection diagrams obtained from instrumented Charpy impact test. In order to vary hardness and the grain size, the specimens of SKD6 produced by ingot and powder metallurgy were quenched from different temperatures, i.e., 1303K and 1403K, and then tempered twice at 873K. It was found that most of impact energy was exhausted in crack initiation for every material and the impact fracture modes were divided into three groups. Although the impact energy tended to decrease with increasing hardness, fine-grained structure suppressed the reduction in impact energy due to the enhancement of the resistance to the crack initiation.

View full abstract

The effect of σ phase precipitation on the toughness of duplex stainless steel and its weld metal was investigated by means of fractography. The material used was SUS329J1 steel, and the samples were welded using SMAW SUS329J4L (DP3) electrode and SUS329J3L (DP8) electrode. The materials were reheated at 1023K and 1123K for various times. The toughness was evaluated by Charpy impact test at 273K.
The fine secondary austenite (γ^*) was formed from the ferrite and then σ phase was precipitated by reheating both base metal and weld metal. The σ phase precipitation in DP3 and DP8 weld metals was much faster than that in the base metal. The lowering of toughness by σ phase precipitation for the weld metals was much larger than that of the base metal. The fracture morphology showed that brittle cracks occurred in the σ phase and at the interface between the matrix and the σ phase in every sample.
In case of the base metal, the fracture morphology showed a dimple pattern at the region around the σ phase. The lowering of toughness was affected by the large volume fraction of σ phase, because of the wide region of ferrite.
In case of the weld metals, the primary austenite and the secondary austenite were fine and crowded and the ferrite region was fine. The fracture morphology showed brittle fracture with fairly tear ridge. Only a small amount of σ phase was enough to cause the lowering of toughness in the weld metals, because of fine σ phase precipitation by reheating.

View full abstract

Stress corrosion cracking (SCC) of austenitic stainless steels in a boiling 42% MgCl₂ solution at 416K was studied fractographically by scanning electron microscopy. The crystallographic orientations of transgranular SCC were determined with the aid of electron channeling patterns (ECPs) obtained from the fracture surfaces. By means of the two-surface trace analysis using the single crystals of SUS304, SUS316 and SUS310S steels, the respective crack orientations of {100} (partly {110}), {210} and {100} plane were found in relation to the crystallographic characteristics of cracking. The distinct ECPs were obtained after relief of internal strain by the heat treatment in a vacuum, though the ECPs obtained directly from the fracture surfaces were very indistinct. Thus the crack orientations determined by the ECPs coincide well with those by the above two-surface trace analysis. The technique with ECP is also applicable to the determination of crack orientation of polycrystalline, and is superior as a microscopic fractographic analysis of SCC.

View full abstract

The effects of σ phase and the volume fraction of austenite (γ) on stress corrosion cracking (SCC) were investigated in high temperature water and in boiling MgCl₂ solution for the sensitized duplex stainless steel by using transmission electron microscopy and fractography. The samples were solution-treated at various peak temperatures and then sensitized at 923K for 72ks. The volume fraction of γ phase decreased with increasing peak temperature. The slow strain rate testing (SSRT) at 4.17×10^-6s^-1 was mainly carried out to evaluate the SCC.
In case of high temperature water at 562K, the fracture strain depended on the volume fraction of γ phase for the as solution-treated sample. For the sensitized sample, the fracture strain was independent of the volume fraction of γ phase, because the secondary austenite γ^* phase near γ phase was preferentially corroded.
In case of boiling MgCl₂ solution, the fracture strain depended on the volume fraction of γ phase and the strain rate for the solution-treated sample. However, the fracture strain was independent of the volume fraction of γ phase for the sensitized sample, because the SCC showed fan-shaped pattern with secondary cracks related to σ phase.

View full abstract

The fracture toughness and fatigue crack growth properties of carbon steel (S35C) were investigated in 1.1MPa argon gas and in high pressure hydrogen gas up to 9.9MPa at room temperature. The effect of hydrogen on the crack growth was discussed. The results obtained are as follows;
(1) J_IH value in hydrogen was smaller than J_IC value in argon and decreased with increasing hydrogen pressure. The quasi-cleavage fracture was observed on the surface fractured in hydrogen, but no stretched zone was observed.
(2) The fatigue crack growth rate in hydrogen was larger than that in argon and increased with increasing hydrogen pressure. The mixture of quasi-cleavage fracture and intergranular fracture was mainly observed in hydrogen.
(3) The critical plastic deformation for the crack growth in hydrogen was smaller than that in argon. The cracks in hydrogen were propagated in forming the quasi-cleavage or intergranular fracture instead of the stretched zone or striations in argon.

View full abstract

In order to clarify the influence of H₂S and CO₂ gas environments on the fatigue strength of high strength steel welded joint, corrosion fatigue tests of 17-4PH T type welded joint were carried out in {H₂S gas (5000ppm)+humid air (353K, 90% relative humidity)} and 99.9% CO₂ gas (353K, 90% relative humidity) environments by use of a corrosion fatigue testing apparatus in gaseous environment.
The fatigue strength of 17-4PH T type welded joint at 2.0×10⁷ cycles in {H₂S gas (5000ppm)+humid air (353K, 90% relative humidity)} and in 99.9% CO₂ gas (353K, 90% relative humidity) was 50 and 46 percent lower than that in air, respectively. The influence of H₂S gas content on fatigue life was remarkable in H₂S gas content ranging 50 to 5000ppm. In the humid H₂S and CO₂ gas environments corrosion pits were observed at the crack initiation area of the fracture surface. While striation was observed at the crack propagation area of the fracture surface. Thus, it is concluded that the initiation of corrosion pits at the crack initiation area is the major cause of the reduction in fatigue strength of 17-4PH T type welded joint in humid H₂S and CO₂ gas environment.

View full abstract

Recently the discrete dislocation dynamics (DDD) methods have been applied to the problems of mesoscopic phenomena. In the DDD simulations, it is very important to know how a dislocation moves in the stress field. The relationship between the velocity of dislocation and external shear stress is often given artificially by referring to the experimental information or experience of researchers. Most of the relationships obtained are, however, qualitative and involve a lot of complex factors which depend on the experimental condition; temperature, grain boundary effect and so on. In this paper, the dynamic properties of edge dislocation whose Burgers' vector b is [111]/ 2 on 112 plane in bcc iron are studied by molecular dynamics (MD) using N-body potential proposed by Finnis and Sinclair. Then, the relationship between the velocity v of dislocation and external shear stress τ is investigated systematically. It is shown in our simulation that the relationship between v and τ is linear in the low stress region and v becomes close to sound velocity asymptotically in the high stress region. These results agree with the experimental results qualitatively. An experimental expression of v-τ relation for the DDD simulation is also proposed.

View full abstract

The atomic scale deformation near the crack-tip field in bcc crystal under anti-twining in-plane shear (Mode II) loading is analyzed by means of molecular dynamics (MD). The appearance and movement of dislocations on (112)-plane whose Burgers'vector b=[111]/2 are observed. The critical stress intensity factor (K^cr_II) obtained from the analysis of the edge dislocation observed at the crack tip does not depend on the condition of analysis, and agrees well with the Rice's standard value of K^cr_II. The experimental expression for the relationship between the velocity of dislocation v_d and τ, which was proposed in our previous report, is applied to the numerical calculation of the motion of dislocation in stress field which is evaluated by considering the interaction between crack and dislocation with linear elasticity. The results of the numerical calculation agree with those of MD simulation.

View full abstract

The present study focuses on the derivation of effective elastic properties for two-dimensional orthotropic elastic continuum containing several different distributions of cracks. The emphasis of the study is placed on the crackinduced anisotropy and comparison of the results derived using the class of mean field theory.

View full abstract

The internal damage evolution of thin tubular specimens of CFRP was studied by means of AE (acoustic emission) measurement. The AE signals were detected first in the quasi static tests on the CFRP tubular specimens of 0°, 90° and ±45° fiber orientations, and three damage modes induced in the tests, i.e., matrix crack, delamination and fiber breakage were identified by the frequency analysis of AE signals. Then, the fatigue tests on the specimens of ±45° were performed under the particular stress ratios R=0 and R=-0.25 by detecting the AE signals for the whole process of fatigue. By use of the results of the quasi static tests, the behavior of damage development in every stage of fatigue was analyzed from the frequency analysis of AE signals. The AE signal related with matrix crack was observed in the middle of the fatigue process, and the AE signal due to fiber breakage was observed in the later stage of the fatigue. The AE signal due to delamination was mainly observed in the later period of the fatigue under stress ratio R=-0.25. The damage state in some stages of fatigue was also observed by a scanning acoustic microscope by interrupting the fatigue test, and the validity of the identification of damage modes by the frequency analysis was confirmed by comparing the results of AE analysis with those of the internal observation. These results agreed also with the results of other measurement of the macroscopic variation of elastic modulus, as well as with those of damage observation on the sliced section of the specimens.

View full abstract

Compression tests on CF/PEEK tubes were performed using a MTS servo-hydraulic material testing machine and a crush testing machine to clarify the mechanism of the initial failure process and the characteristic of the energy absorption. It was found that all the specimens were fractured in a progressive crushing mode and the specific energy absorption indicated the dependence of testing speed in the range of our study. Morphological observation was conducted to examine the influence of testing speed on the specific energy absorption. The mechanism in the steady state stage which takes place after the maximum load is characterized by the interlaminar cracks and lamina breakage at the crush zone. In the initial failure process, the formation mechanism of the debris wedge was examined through the step-wise morphology analysis. The angle and the dimension of the debris wedge were the essential factors in the specific energy absorption.

View full abstract

The effect of microstructure on the high-cycle fatigue strength of an oxide dispersion strengthened Ni-base superalloy, MA758, was studied at temperatures ranging from room temperature to 950°C. Three kinds of materials with different microstructures were produced by different thermo-mechanical treatments: with a fine and equiaxed grain structure, with a coarse and highly elongated one, and with an intermediate and elongated one. The microstructure was found to have a pronounced effect on the fatigue strength in this alloy, which did not obey a simple correlation with the tensile strength, as can be seen in usual metallic materials. The optimum microstructure also behaved in a complicated manner depending on test temperature. These features were studied on the basis of the resistance to fatigue crack initiation, as well as that to crack propagation. The positive role of oxidation film in the fatigue crack initiation, to which it was worthy to pay special attention, was also identified and discussed.

View full abstract

Fatigue crack propagation (FCP) behaviour of titanium aluminide TiAl with a nearly fully lamellar microstructure has been studied in ambient temperature on two different FCP directions relative to the lamellar orientation, i.e. parallel (Type A specimen) and perpendicular (Type B specimen) to the lamellar orientation. It was found that the FCP resistance of the former was considerably lower than that of the latter. Close examination on crack morphology revealed significant differences between two FCP directions. In Type A specimens several cracks along lamellar were seen on both surfaces and sections of the specimen, thus uncracked ligaments were formed in the wake of the crack tip. On the other hand, such ligaments were scarcely produced in Type B specimens because only main crack could propagate without remarkable deflection and branching. The FCP rate of Type A specimens decreased with crack extension under constant ΔK tests, suggesting the role of crack bridging by uncracked ligaments. FEM analysis indicated considerably reduced ΔK experienced at the crack tip, thus the difference in FCP resistance between two FCP directions based on the actual ΔK after allowing for crack bridging became much larger than that based on the nominal or applied ΔK.

View full abstract

The strengthening of Al-Zn-Mg-Cu system alloy results from the interaction between dislocation and solute atoms or precipitates. In this study, the tensile properties of 7475 aluminum alloys having various average grain sizes were evaluated. The 0.2% proof stress and elongation were discussed based on the theory of dislocation mechanism and the observation of microstructures. The main results obtained were as follows:
(1) The 0.2% proof stress, tensile strength and elongation of the materials increased with decreasing average grain size.
(2) In accordance with the Hall-Petch's relation, the 0.2% proof stress was plotted against the reciprocal square root of the average grain size of the aluminum matrix. A good correlation was obtained, indicating that the 0.2% proof stress depends on the average grain size of the aluminum matrix.
(3) The contributions of the precipitation strengthening, solid strengthening and grain refining effect to the 0.2% proof stress of the present 7475 aluminum alloy were estimated to be 258, 153 and 5MPa, respectively.

View full abstract

The fatigue behavior of ceramics has been discussed on the basis of the relation between stress intensity facter (K_I) and crack velocity (V). In this paper, the effect of environment on the relation between K_I and V was studied on machinable ceramics (mica glass ceramics) and two kinds of glass ceramics with different grain sizes. The double torsion (DT) technique was used for the determination of the K_I-V characteristics under different environments of air and ion-exchanged water. The characteristics of acoustic emission (AE) during stress corrosion cracking of mica glass ceramics was also examined. In water environment, the region II in the K_I-V curve, in which crack velocity varies slowly with K_I, disappeared. From this experimental fact, it was considered that at high K_I, the crack velocity is encouraged by diffusion of the corrosive species to the crack and thus depended on the amount of water. SEM fractography revealed that mica single crystals in the material caused crack arrest and deflection to occur. It is also found that AE event rate is quantitatively related to the crack velocity. AE measurement can be used in studying the crack propagation behavior of mica glass ceramics.

View full abstract

Ota and co-workers determined a set of equilibrium constants which govern the number density of the structural units in alkali borate glass. Based on the results, the molar volume of the system was computed. However, this technique can not be applied to the Na₂O-B₂O₃-GeO₂ system where the state analyses on Ge ions in the Na₂O-B₂O₃-GeO₂ system have not been fully conducted. In the present study a new method to calculate the molar volume of multi-component alkali borate system is presented. Raman spectra were used to elucidate that Na₂O affinity for B₂O₃ is stronger than that for GeO₂. Assuming the relative magnitude of activity coefficient γ of Na₂O for B₂O₃ against for GeO₂ tentatively 1:2, the distribution of Na₂O among B₂O₃ and GeO₂ was determined. The molar volume of the Na₂O-B₂O₃-GeO₂ system was calculated from the partial molar volumes of the sub-systems Na₂O-B₂O₃ and Na₂O-GeO₂ in good agreement with the experimental value.

View full abstract

A computer-aided interactive system for coating desing has been developed, which enables to analyze coveniently the reaction diffusion of bonded materials. The object of this study is the overlay coatings of MCrAlY alloy sprayed by a low-pressure-plasma spray (LPPS) process for protection against high-temperature corrosion and oxidation in the field of gas turbine components. However, the reaction diffusion behavior at the interface between the MCrAlY coating and the substrate, which has an important effect on coating degradation, has not been fully clarified. Four kinds of low-pressure-plasma sprayed MCrAlY alloys, namely CoCrAlY, NiCrAlY, CoNiCrAlY and NiCoCrAlY, and single-crystal CMSX-2 were selected for the experiments.
The experimental results showed that the reaction diffusion layers consisted of aluminum compound layer and low aluminum layer. In case of the NiCoCrAlY and CoNiCrAlY coatings, the aluminum compound layer could not be observed clearly. It was also indicated that each diffusion thickness could be expressed by the parabolic time dependence. The order of reaction diffusion rate was NiCrAlY>CoCrAlY>CoNiCrAlY>NiCoCrAlY. It was also clarified by the simulation analysis that the diffusion distance during heating process can not be ignored in comparison with the total diffusion distance, and the estimation of long time diffusion was made.

View full abstract

An electric field analysis of a dilute NaCl solution for Scanning Vibrating Electrode Technique (SVET) is performed by using a model which is based on a mathematical description of the physical mechanisms including ionic migration and diffusion. A cylindrical problem is solved using a finite difference method and Runge-Kutta method. The computational results show that the el ectric potential gradient decreases significantly with time. This phenomenon cannot be explained by the conductance change due to ion distribution. It is found that the ionic diffusion has an important role in it.
The model is applied to a two-dimensional problem, in which the specimen has two parallel electrode plates in a resin. It is also found that the potential gradient reduces to about one-fifth of initial value or the result predicted by the conventional model with uniform electrolyte solution. A comparison between the results obtained by SVET and the computational results shows good agreement.

View full abstract