The uniaxial creep rupture strength of Mod. 9Cr-1Mo steel welded joints is lower than that of the base metal, due to the existence of the heat affected zone (HAZ) softening region which is formed by welding heat. In this paper, the multiaxial creep damage of welded joints is evaluated analytically using the Finite Element Method (FEM), so as to study the width of HAZ softening region that affects creep damage.
To begin with, creep strain equations were established based on the result of uniaxial creep tests. The uniaxial creep damage was evaluated analytically using these creep strain equations, and the result agreed with the experimental result. It was found from the uniaxial creep damage analysis that the effect of HAZ softening region on creep rupture strength is negligible, in the case of relatively narrow HAZ softening region compared with the diameter of welded joints.
The analytical result of internal pressure creep tests using the FEM was correlated well with the experimental result. Therefore, the creep damage of welded joints including HAZ softening region can be evaluated analytically.

View full abstract

Life analysis for fatigue-creep situations of Cr-Mo-V rotor steel was studied using the results of fatigue tests conducted with strain hold and stress hold at 500°C. The following conclusions were obtained.
(1) Linear damage theory and the ductility exaustion rule are applicable when the results of high-temperature low-cycle fatigue tests and creep rupture tests are available. However, both methods are not appropriate for accurate life analysis; the former tends to yield an unsafe estimated life and the latter tends to yield too generous results.
(2) Priest and Ellison well estimated the experimental results obtained at 565°C using a combined deformation map-ductility exhaustion approach which defines creep damage as that caused by grain boundary sliding during hold time. Therefore, the same method was applied to the experimental results obtained by the authors. It was found that the method is excellent for accurate life analysis.
(3) When the combined deformation map-ductility exhaustion approach is applied, the necessary creep rupture ductility can be substituted by the product of minimum strain rate and rupture time, both of which are obtainable by relatively short time creep tests.

View full abstract

Creep-fatigue is a fatal failure mode of the high temperature structural materials of LMFBR. In this report, two important issues are discussed for the creep-fatigue evaluation of normalized and tempered Mod. 9Cr-1Mo steel which is a promising structural material for the steam generator of large scale LMFBR in Japan.
A new ductility exhaustion theory (NDET), a simplified conventional ductility exhaustion approach (DEA) and a time fraction approach (TFA) based on NDET are studied for the prediction of tension strain hold creep-fatigue damage of this material. Though Mod. 9Cr-1Mo (NT) steel has quite different basic material properties from those of austenitic stainless steel, NDET proposed for SUS304 is able to give a proper prediction for the creep-fatigue life of Mod. 9Cr-1Mo (NT) steel. The applicability of this evaluation method to Mod. 9Cr-1Mo (NT) steel is discussed based on micro-structural observations. It is considered that the creep-fatigue damage of this material under actual loading conditions is dominated by the creep-cavitation of grain boundaries in the same way as that of austenitic stainless steel. Finally, TFA based on NDET is proposed as the design creep-fatigue evaluation method of Mod. 9Cr-1Mo (NT) steel from the point of view of its appropriate conservatism for time-extrapolation and its simplicity.
The life reduction mechanism of low cycle fatigue of Mod. 9Cr-1Mo (NT) steel with strain hold at the compression side is discussed based on the data observed by a scanning type electron microscope. A new concept based on the location of oxidation on the test specimen surface can explain the reduction of low cycle fatigue life of Mod. 9Cr-1Mo (NT) steel.

View full abstract

For long-term creep-fatigue of Type 304SS, intergranular failure is dominant when significant life reduction occurs. This phenomenon has its origin in grain boundary sliding as same as that in cavity type creep failure. Accordingly a simplified procedure to estimate intergranular damages caused by grain boundary sliding is presented in connection with the secondary creep.
In the conventional ductility exhaustion method, fracture ductility includes plastic strain, and damage estimation is based on primary creep recoverable during strain cycling. Therefore, the accumulated creep strain becomes a very large value, and is quite different from grain boundary sliding strain. As a new concept on ductility exhaustion, the products of secondary creep rate and time to rupture (ε₂t_R) is applied to fracture ductility, and grain boundary sliding is estimated using the accumulated secondary creep strain. It was shown that the time fraction rule and the conventional ductility exhaustion method can be derived analytically from the new concept.
Furthermore an advanced method on cyclic stress relaxation was examined. If cyclic plastic strain hardening is softened thermally during strain hold, softening of the primary and secondary creep can occur. And the unrecoverable accumulated primary creep strain causes hardening of the primary creep. The reduction of deformation resistance to the secondary creep accelerates the grain boundary sliding rate, and increases creep damage.
The new concept ductility exhaustion method based on the above consideration gave good life prediction for the intergranular failure mode.

View full abstract

Type 316 stainless steel with low carbon and medium nitrogen composition gains large interest because of its selection as a main structural material of the demonstration fast breeder reactor plant. As a part of the efforts for collecting strength data for design use, the authors have been conducting various strength tests on two product forms of this steel manufactured by hot-rolling and forging process, respectively. Using the results of these tests, they also examined the applicability of several creep-fatigue life prediction methods to this steel. It was found that the modified ductility exhaustion method proposed by one of the authors gave satisfactory agreement with the test results in comparison with the time fraction rule and classical ductility exhaustion method which have been used in design or integrity assessment of high-temperature reactor plants.

View full abstract

In order to investigate the applicability of a ductility exhaustion rule to the creep-fatigue life assessment of FBR316 stainless steel, a feasibility study using the published data was conducted. The assessment method was proposed based on the linear damage summation rule. In the proposed method, fatigue damage was calculated by Minor's rule and creep damage was calculated by a ductility exhaustion rule. The creep-fatigue lives in the published data were predicted by the proposed method. The results obtained are as follows:
(1) All the data could be predicted within a factor of two on life by the proposed method.
(2) The creep-fatigue lives under 10 minute strain hold at 550°C were overestimated, while those under 60 minute strain hold at 550°C and 600°C were estimated adequately. From the above facts, the proposed method seemed to be effective for the prediction of creep-fatigue life in which the creep damage was dominant and also the intergranular cracking was remarkable.
(3) The creep damage was simultaneously calculated by the time fraction rule in order to compare with the ductility exhaustion rule. All the data could be also predicted within a factor of two on life by this rule, but it tended to overestimate the life.

View full abstract

In order to clarify the validity of the previously proposed creep-fatigue life prediction model based on the strain range partitioning concept, the applicability of the proposed model to type 316LC steel was examined. Two-step variable PP type straining (high-low and low-high) tests were conducted at 700°C, and the experimental results were compared with those predicted by the proposed model. In both high-low and low-high tests, 1.0% was chosen as the higher strain range, and 0.8%, 0.4% and 0.3% were chosen as the lower strain ranges. Furthermore, PP type small crack growth behavior in smooth bar specimens was observed at 700°C under constant strain range by the surface replica technique and compared with that predicted by the proposed model.
The high-low and low-high test results could be predicted by the linear damage rule (LDR) when the lower strain range was 0.8%. On the other hand, when the lower strain range was 0.4% or 0.3%, the deviation of experimental data from the LDR prediction was found. That is, the sum of the life ratio of the first and second step straining was smaller than unity in the high-low tests, whereas it was larger than unity in the low-high tests.
These experimental results were well explained by the proposed fatigue life prediction model, in which it was assumed that the crack initiation life is not negligible when the inelastic strain range is smaller than the critical strain range, (Δε_pp)_cr, and it was found that the proposed model was successfully applied to type 316LC steel. The value of (Δε_pp)_cr was 0.23%. The experimentally obtained small crack growth curves were well predicted by the proposed model using the correction factor, with which the macro through crack growth rate in the CCT specimen was converted into the small surface crack growth rate in the smooth bar specimen.

View full abstract

Creep-fatigue damage occurs gradually in high temperature components and structures. It is necessary to clarify the micro-damage evolution mechanism under creep-fatigue conditions for the quantitative evaluation of creep-fatigue damage in local portions of the components. In order to examine the relation between damage evolution and change of microstructure, such as creep cavities and surface microcracks, strain controlled creep-fatigue tests were performed and interrupted at several damage levels. The creep-fatigue tests were also performed under a low stress level by the high temperature fatigue machine combined with a scanning electron microscope (SEM), and the microcrack initiation and growth behavior were continuously observed by the SEM during the tests.
It was found that even though many cavities were initiated and grew at the internal grain boundaries of the specimens during the strain controlled creep-fatigue tests, the failure life was governed by the propagation of surface cracks. On the other hand, microcracks of about one grain size order initiated mainly on the grain boundaries normal to the loading axis under low stress creep-fatigue, and the crack propagation rate of the microcracks was slow and at random by the influence of microstructure. The microcracks gradually opened to the loading direction with increasing number of cycles and coalescening occurred to grow. From the experiments, the creep-fatigue damage extension mechanism was divided into the Fc type, where the main cracks on the surface propagated through damaged grain boundaries, and Cf type, where damage evolution was characterized by opening and coalescening of the microcracks.

View full abstract

Successive process from fatigue small crack initiation to propagation in two kinds of aluminum base composite materials was studied in comparison with long crack growth behavior at high temperature: One is an AIN-particulate reinforced composite and the other is a SiC-whisker (SiCw) reinforced 6061 alloy composite. The results obtained were also compared with that of a traditional heat-resisting aluminum alloy, AC8A. It was shown that the fatigue crack in the both composite materials easily initiated from the region where the reinforcement phase, AIN particulate or SiCw, gathered with high density. Like to general polycrystalline metallic materials, some noteworthy phenomena, which indicate the lack of similitude between small and long cracks, were seen in the small crack propagation process in the both composites: the small cracks grew even at the lower stress intensity factor range than the long crack threshold, and they exhibited higher growth rates than the long cracks. It was also shown that with respect to the effect of reinforcement the small cracks sometimes behaved with conflicting manner to the long cracks. These experimental evidences indicate the importance of studies on small fatigue crack propagation.

View full abstract

To develop a method for evaluating the fatigue life of large structural components by using the data obtained with bar specimens 10mm in diameter, a bending fatigue test was performed on a large butt-welded joint plate of a 50mm thick hot-rolled SUS304 stainless steel made by hot-wire TIG welding. This test confirmed the evaluation method proposed and showed the followings.
(1) The peak strain range in the welded metal becomes 1.8 times as large as the average strain range of welds because of the metallurgical discontinuity in the welded metal and the different deformation characteristics of the base and weld metals. As a result, fatigue cracks initiate where the peak strain occurs.
(2) The most part of fatigue life of large structural components is crack propagation process, so that it is reasonable to evaluate fatigue life based on crack length.
(3) Scale factors between the large structural component and the small bar specimens can be calculated using the fatigue failure probability of the small bar specimens, by taking the ratio of two test region areas into consideration.
(4) If the crack initiation life corresponding to a surface crack length less than 2mm is used, the life evaluation method based either on the base metal data or on the welded joint data obtained using small bar specimens is reasonable.

View full abstract

In order to clarify the validity of the previously proposed creep-fatigue damage rule, in which small crack growth curves are predicted by the crack growth rate equations based on the strain range partitioning concept and used as the damage accumulation curves, the correspondence between the predicted small crack growth behavior and the experimentally obtained behavior was examined. The small crack growth behavior of smooth bar specimens in PP, CP, PC and CC tests was observed for Mod. 9Cr-1Mo steel at 600°C by the surface replica technique, and compared with that predicted by the proposed damage rule. The tests were conducted under constant strain ranges, where 1.0% was chosen as the high strain range and 0.3%-0.5% were chosen as the low strain ranges.
The predicted small crack growth curves well corresponded with the experimentally obtained relationship between the maximum crack length and the life ratio except for the CP, PC and CC type low strain ranges. The deviation between the prediction and experiment for the CP, PC and CC type low strain range was found to be decreased by modifying the initial crack length in the proposed damage rule and by taking the crack initiation life into consideration like the case for the PP type low strain ranges.

View full abstract

Creep-fatigue tests of a Type 304 stainless steel are conducted and cracking behavior is observed. The results obtained are summarized as follows; Intergranular fracture under creep-fatigue conditions at high temperatures is classified into inner cracking type and surface cracking one. The crack initiation inside of a specimen in the former type is caused by the growth and coalescence of grain boundary diffusive cavities and the fracture is brougth about by the coalescence of distributed small cracks at the final stage of life. On the other hand, the cracks at the surface in the latter type is attributed to the grain boundary sliding and the fracture is caused by the growth of surface small cracks. The inner cracks tend to appear at lower tensile strain rate, at higher compressive strain rate, and at higher temperature. A fracture mechanism map is proposed on the basis of the experimental observation.

View full abstract

The most dangerous fracture of creep-fatigue is caused by multiple small cracks which initiate and grow inside of materials. In this study, a numerical simulation method is proposed on the initiation and growth of inner small cracks on the basis of the experimental observation of a Type 304 stainless steel during creep-fatigue in order to elucidate the evolution process of damage. The model proposed in this study is similar to that for creep-fatigue intergranular fracture of surface cracking type which has already been reported by the authors. It is based on a discrete model of grain boundaries having random shape and sizes with random fracture resistances and a Monte Carlo simulation combined with a damage mechanics concept. As a result, the spatial and temporal distributions of inner small cracks are successfully evaluated by the simulation.

View full abstract

A large number of fossil power plants have already been operated beyond the design life of 100000 hours in Japan. It is therefore very important to predict residual life of materials used in boiler and to prolong the service life of boiler plant. In order to predict the residual life of materials used in boiler, several non-destructive life assessing techniques based on metallographical examination using replica were developed and applied to the life assessment of actual boiler plants.
For the purpose of improving the accuracy of the present life assessing techniques, a new non-destructive evaluation technique was developed based on a quantitative analysis of microstructure. The results obtained are following.
(1) The creep rupture strength is reduced by thermal aging. In order to evaluate damage by thermal aging, a quantitative analysis of microstructure using an image analyzer was performed and an area fraction of dark parts of microstructural image was selected as an appropriate parameter for aging evaluation.
(2) This evaluation technique based on a quantitative analysis of microstructure can be applied to the synthetic evaluation which was developed in our previous study. Also, using this technique, there is a possibility that creep damage is estimated directly from microstructure.

View full abstract

Scratching tests were carried out on the (111) face of Cu-14.7 at.%Al single crystals and bicrystals with the use of pyramidal indentor. In the scratching on the single crystals three kinds of scratching directions, [112], [112] and [110] were chosen. In the case of bicrystals, the [110] and [110] directions were adopted. After scratching the distribution pattern of the slipped traces and the surface profiles across the scratched track were examined. In addition, the dislocation distributions inside the crystals were revealed by successively removing thin layers and developing etch pits on the exposed surface.
The conclusions obtained are as follows:
(1) The slipped traces on either side of the scratched track are produced more extensively than those around the indented terminal.
(2) The swells of the material were caused in front of the scratched track by slips on the {111} faces which diverge downwards. So the width of the scratched track appears narrower than that in the indented terminal.
(3) In the scratching which acrosses the grain boundary, the deformation in the surface layer of the crystal is obstructed due to the existence of grain boundary. But the deformation inside the crystal is little influenced by the grain boundary.

View full abstract

Based on “metallo-thermo-mechanics” proposed by the authors, a CAE system “HEARTS (Heat Treatment Simulation System)” has been developed to simulate the heat treatment processes associated with phase transformation, such as quenching and tempering by utilizing the finite element method. In the first part of the paper, the summary of the governing theory is introduced; the fundamental equations of metallurgical change due to phase transformation, heat conduction and inelastic stress/strain as well as the diffusion of carbon in the course of carburization, and the effect of coupling among these three fields are discussed. The developing strategy, methodology, structure of the program and the direction for use of thus developed system “HEARTS” are stated in the following sections. Some examples of simulated results on carbon content, temperature, structural change and stress/strain during the heat treatment process of engineering components in two- and three-dimensional shape are illustrated by use of the system combined with an available pre/post processor, and the validity of the system is evaluated by comparing with the experimental data.

View full abstract

It is generally recognized that one of the reasons for improvement on fatigue durability by shot peening is surface compressive residual stress. And a shot peening condition decides a residual stress profile on surface layer. But, at this moment the method to confirm this profile is only measurement by X-ray difractometer. In this study, a simulation technique was applied to predict the residual stress distribution caused by shot peening. Firstly, the residual stress on shot peened plate specimens was measured by X-ray difractometer and the effect of shot peening condition on profile was confirmed. The equation to simulate the profile was obtained by the superposition method of simple three stresses, based on the equation that Y.F. Al-Obaid suggested. It was confirmed that the predicted stress profile shows good correspondence with the experimental value every shot peening condition of this experiment.

View full abstract

In the present study, the strength of adhesively bonded Al-FRP cylindrical joints under torsional moment was investigated. In the experiments, Al-GFRP and Al-CFRP joint specimens were tested. And for the Al-CFRP joints, three kinds of specimens having different bonded areas were prepared. These specimens were used in order to investigate the effect of the variation of bonded area on the torsional strength. For the stress analysis of the joints, an elastic-plastic FEM program was made. Accuracy of this FEM program was confirmed by comparing the results calculated by the program with the experimental data on the distribution of strain at the surface of the bonded part. And the distribution of stress in the adhesive layer at the breaking load was examined. The followings were made clear from the results of this study.
(1) Both Al-GFRP and Al-CFRP joints had the same torsional strength.
(2) The fracture torsional moment of the joints increased with the bonded area and the mean shear strength at the adhesive layer decreased with increasing bonded area.
(3) The adhesive yielded mainly at one side of the edge of the adhesive layer at the breaking load. The area of high stress in the adhesive layer was one-sided. And the area of low stress in the adhesive layer expanded with increasing bonded area.
(4) Maximum equivalent plastic strain and plastic area at the edge of the adhesive layer increased linearly with the bonded area.

View full abstract

Tightening bolts in plastic regions, which can generate stable and high tightening force, has come to be used for joints of various machine parts. In this study, the fatigue strength of bolts tightened in plastic regions was investigated in its relation to the turning angle of the bolt or nut. The bolts used in fatigue tests were (A) shank-elongation bolts with a shank diameter (SD) of 7.2mm, (B) shank-elongation bolts (SD: 7.5mm) and (C) thread-elongation bolts (SD: 8.3mm). The fatigue tests were carried out for bolted joints and bolts alone. The former was tightened in elastic or plastic regions by varying the turning angle of the bolt or nut. The latter was chucked not only under no-turning but also under turning while loading axially till the initial mean load for simulating the actual tightening.
The results obtained are as follows:
(1) The endurance limit of the external load of the bolted joints in plastic regions was higher for shank-elongation bolts than for thread-elongation bolts, and that of the shank-elongation bolts with smaller shank diameter was higher. The reason is considered that the force generated in lower-rigidity bolts is smaller than that in high-rigidity bolts even if the external load is equal.
(2) The endurance limit load of the bolts was higher for the bolts tightened to the turning angle (TA) of 90° in elastic regions than for bolts alone (no turning), and also higher for the bolts tightened (TA<180°) in plastic regions than in elastic regions. It is considered that these are caused by the increase in the uniformity rate of the load on the circumference of the thread ridge because of more uniform contact with scrape of the thread surface during tightening. These phenomena were similar to the results of the simulation tests for bolts alone.
(3) However, the endurance limit loads of the bolted joint tightened to the turning angles of 270° and 360° in plastic regions were lower than that in elastic regions. In the simulation tests of the bolts alone, the endurance limit load reduced rapidly when the turning angle was over 180°, and it was almost constant between 180° and 1080°. The results of the bolted joints corresponded well with the above relation. These are thought to be because the load distribution of the first thread ridge increases due to the screw micro-pitch-error induced by high load.

View full abstract

Static and cyclic bending fatigue tests were carried out on two types of silicon nitride ceramics at high temperatures in air. The effect of the viscosity of grain-boundary phase on fatigue strength was investigated.
The experimental results show that the effect of stress cycling on fatigue strength varies depending upon the following three temperature levels in relation to the viscosity of grain-boundary phase.
(1) In the temperature region below the softening point of grain-boundary phase, the cyclic fatigue strength was lower than the static fatigue strength. This is attributed to the acceleration of crack propagation rate by stress cycling.
(2) At temperatures just below the softening point of grain-boundary phase, the cyclic fatigue strength was higher than the static fatigue strength, and almost all the specimens of cyclic fatigue did not fracture at 10⁷ cycles. The average flexural strength of these survival specimens was over 20% higher than the average static strength of virgin specimens at the same temperature. This may be due to the healing of defects such as machining-induced damages.
(3) In the temperature region where the grain-boundary phase is softened, the cyclic fatigue strength was lower than the static fatigue strength. The cyclic fatigue life, which was predicted from the static fatigue strength with an assumption that the fatigue fracture depends upon only time, was shorter than that in the experimental data. The longer experimental life in cyclic fatigue is attributed to high loading rate which generates smaller strain.

View full abstract

In a batch system with an electron beam irradiation, experiments have been carried out for accelerating chemical gas reactions. Carbon dioxide (CO₂) is a low free energy compound, and either a high energy input of electron beams or a use of other high free energy reactants is needed for radiation chemical reactions. In this paper, the effects of content of added gas elements and an exposure dose of the electron beam irradiation on the CO₂ gas reactions were investigated by monitoring the variations of gas temperature and gas pressure. As a result, there was a tendency that the CO₂ gas reactions with an added acetylene gas (C₂H₂) increased with increasing the content of C₂H₂, the exposure does of electron beams and the gas temperature. Also, in this experiment, it was confirmed that carbon monoxide, methanol and some solid polymer could be produced under the condition of room temperature and atomospheric pressure.

View full abstract