The growing market for industrial gas turbines has led to an increased demand for large, cost-effective units of high efficiency, A critical issue in the development of such units is the durability of hot section components, especially first stage blades and vanes to which thermal barrier coatings (TBCs) are applied. This paper introduces the current state of the art in the production of TBCs, and discusses their various degradation mechanisms. Special emphasis is placed on the potential for the development of a new generation of TBC systems through the control of microstructure and porosity, as well as through new processing techniques.
In order to examine the applicability of the effective-stress-based analysis method of perforated plates that is proposed by the authors, inelastic behavior and creep-fatigue life of a perforated cylinder subjected to cyclic thermal stress were predicted. Firstly, basic creep behavior of a perforated cylinder under axial tension at uniform temperature was examined by means of three-dimensional FEM creep analysis, and the results showed that macroscopic and local creep behavior of a perforated cylinder is predictable using the proposed method on perforated plates. Secondly, thermal fatigue testing was carried out on a perforated cylinder, and the elastic-plastic-creep behavior of the cylinder was numerically analyzed by modeling the perforated cylinder to the equivalent solid cylinder based on the effective stress concept. Predicted local stress-strain behavior around circular holes was used for creep-fatigue life prediction based on the linear damage rule. Comparison of numerical results and experimentally observed damage clarified that the difference in damage mode at the inner and outer surfaces of a perforated cylinder could be successfully reproduced.
A convenient method for predicting mechanical properties of solders was proposed in this study. Indentation method was used to obtain mechanical properties including elastic, plastic and creep deformations. Elastic-plastic finite element method (FEM) correlated Young's modulus, E, and yield stress, σys, with unloading portion of load-displacement curve and Vickers hardness, HV, respectively. The creep properties of solders were evaluated in terms of the Norton's law, ε=A·σn. Indentation creep and tensile creep deformations were correlated by elastic-plastic-creep FEM analysis, and the latter could be directly predicted by the former. The agreement of the prediction was confirmed for Sn-37Pb eutectic solder in the range between room temperature and 125°C. Using the proposed method, mechanical properties of lead-free solders such as Sn-3.5Ag and Sn-3Ag-0.5Cu-0-3Bi alloy systems were determined. The predicted results revealed that mechanical properties of Sn-3.5Ag and Sn-3Ag-0.5Cu alloys were similar, and that E, σys and activation energies for creep of Sn-3Ag-0.5Cu-0-3Bi lead-free solders tended to increase with increasing Bi content.
Type IV cracking in heat affected zone (HAZ) of weldment is a problem for advanced high Cr ferritic steels. The present paper investigates the creep properties and microstructures of W strengthened P122 steel weldment at 923K. From the investigation of creep properties of simulated HAZ, it is clarified that heating up to around Ac3 during welding minimized the grain size and creep strength. Most of the welded joint specimens were type IV fractured in fine-grained HAZ and resulted in shorter creep lives than those of the base metals. Electron beam welded joint with very narrow HAZ also showed the brittle type IV fracture due to the formation of creep voids and cracks. The growth of intergranular precipitates was faster for pine-grained HAZ. On the basis of experimental results, the FEM code that simulates type IV crack growth behavior has been developed. The vacancy diffusion under multi-axial stress condition in HAZ of weldment is analyzed.
Crack growth mechanisms taking into account the oxidation and mean stress effect in nickel- and cobaltbased superalloys were determined by crack growth tests using a strain waveform including compressive strain hold. Cracks were found to propagate on the dendrite boundary in both superalloys. Crack growth rates were increased by increasing the compressive hold time. Oxidation around the crack tip was detected by electron probe micro analysis (EPMA) and was found to accelerate the fatigue crack growth. The oxidation component in the normalized crack growth rate by oxidation is proportional to about the 1/3 power of loading frequency for Ni-based superalloy, whereas the power for Co-based superalloy is about 1/2. Furthermore, mean stresses in the nickel-based superalloy, which has a high proof strength even at high temperature, were increased by the compressive strain hold. It was also found that increasing the mean stress accelerates the crack growth. A crack growth model that considers effects of both oxidation at crack the tip and mean stress was discussed from these test results.
A practical evaluation of crack growth lifetime using nonlinear fracture mechanics is needed for remaining-life assessment of high-temperature structural components after long-term service. In this study, creep-fatigue crack growth behavior under displacement-controlled conditions was examined using CT specimens on 2 1/4Cr-1Mo steel. The J-integral estimation method and the crack growth prediction method using the fully plastic solution were also examined. Creep-fatigue crack growth under displacement hold could be separated into fatigue crack growth in the cyclic portion and creep crack growth in the holding portion. These two mechanisms were competitive with each other, and crack growth behavior was determined by the dominant mechanism of fatigue and creep. The J-integral approach using the fully plastic solution was introduced for predicting low cycle fatigue crack growth and creep-fatigue crack growth with relaxation. The suitability of this solution was verified by comparing prediction with experimental data obtained via the slope-line-control method.
Thermal aging embrittlement of newly developed W-alloyed 9%Cr ferritic steel KA-STBA29/KA-STPA29 (ASME T92/P92) was investigated. In order to clarify the controlling factor of embrittlement, Charpy impact tests were carried out, and characterization of precipitates was determined through use of materials that had been aged under various conditions. Drops in the upper and lower shelf energies and an increase in the fracture appearance transition temperature (FATT) induced by pre-aging were found to be closely associated with an increase in the amount of precipitates, such as the M23C6 type carbide and the Laves phase. As the result, the drop in impact toughness was found to be uniquely correlated with an increase in the area fraction of the Laves phase. In order to develop a nondestructive procedure for detecting thermal aging embrittlement, the changes in electrochemical properties of KA-STBA29/KA STPA29 induced by pre-aging have been investigated, and the electrochemical polarization technique has been applied to the pre-aged materials. Experimental results on electrochemical polarization measurements revealed that the peak current density “Ip” which appears at a specific potential during potentiodynamic polarization curve measurements in 1N-KOH solution increases linearly with the degree of embrittlement as evaluated by impact absorbed energy at 0°C. This increase in Ip was correlated with an increase in selective dissolution volume of the Laves phase. Consequently, the Charpy impact absorbed energy, along with the degree of thermal aging embrittlement, can be nondestructively estimated by the electrochemical technique.
This paper studies creep, fatigue and creep-fatigue interaction properties of several ferritic steels such as conventional Mod. 9Cr-1Mo, 9-12Cr-2W steels, newly produced boron- or palladium-added 9Cr-3W steels, high strength oxide dispersion strengthened (ODS) ferritic steels, and a precipitation hardening carbon-free martensitic steel. From the tests, two conclusions were derived to improve the creep-fatigue interaction strength. One is to have high creep ductility, and second is to decrease plastic strain of cyclic deformation by ultra high strengthening. From a viewpoint of evaluation of the creep-fatigue life, the analysis by ductility normalized strainrange partitioning method explained well the effects of creep ductility and strength of the materials on the creep-fatigue interaction.
The microstructural change in a single crystal Ni-base superalloy, CMSX-4, was studied, in order to get basic understandings on the problems that might be serious on the process of repair and recoating of hot section components for advanced gas turbines. It was shown that a cellular γ/γ' microstructure was nucleated, when the material was subjected to damage associated with local plastic straining, followed by the re-heat treatments for damage recovery. Once the cellular microstructure was nucleated inside the material, the fatigue strength was remarkably reduced, hence the cellular γ/γ' microstructure might be a very serious hurdle to be solved. Special efforts were made to explore the method to reduce and prevent the unfavorable effect due to the cellular microstructure and to detect it nondestructively. The experiments indicated that the application of the surface modification technique, or protective metallic coating, on the superalloys, was found to be a possible method. It was also indicated indirectly that the control of parent alloy composition could be important to reduce the harmful effects.
The purpose of the present paper is to establish a damage constitutive model for predicting the elastic-brittle mechanical behavior of continuous fiber reinforced ceramic matrix composites (CFCC). In this paper, the anisotropic damage is applied to describe the matrix phase damage which reflects all types of damage that the matrix material undergoes such as matrix cracking and transverse cracking. The asymptotic expansion homogenization method is used to obtain the effective mechanical properties of the composites, and to derive homogenized damage elastic concentration factor of unidirectional and cross-ply laminate composite materials. Internal variables are introduced to describe the evolution of the damage state under uniaxial loading and as a subsequence the degradation of the material stiffness. Results obtained from the numerical simulations include damage evolution prediction and non-linear stress-strain analyses of macro-microstructure, and they are compared well with existing experimental data.
Surface strains of Al2O3/NiCr thermally sprayed SUS304 steel specimens during the fatigue test (σmax=173MPa, R=0, T=873K) were measured using an electronic speckle pattern interferometry (ESPI) system. The relationships between surface strain and crack initiation/delamination behavior are discussed. The strain values obtained from the ESPI system were confirmed to be almost the same as those from strain gauges on un-sprayed specimens when tensile stresses were loaded at 293K. Thermal expansion deformation and stress deformation at high temperatures were easily measured with the ESPI system. The presence of cracks and delamination on the sprayed coatings can be nondestructively detected by analyzing the strain concentration or decrease. The surface strains of sprayed specimens were almost the same as those of un-sprayed specimens at 873K, indicating that the deformation of the sprayed coatings are always associated with that of the substrate surfaces at high temperature. The maximum surface strain after 1×105 cycles test was a little lower than after 2 cycles test. Surface cracks occurred but stopped at the inner NiCr layer after 2 cycles test at 873K. Many surface cracks and delamination along the interface between the NiCr layer and substrate inte face were confirmed after 1×105 cycles test.
Residual stress was generated in several structural ceramics by sintering and grinding, then estimated using the Vickers indentation method. First, we sought the most appropriate pretreatment for measuring fracture toughness (basis value, KC), while preventing any influence from residual stress. This is important in estimating residual stress using the Vickers indentation method. Based on that value, the residual stress in Al2O3 and Si3N4 ceramics was estimated by the Vickers indentation method. Next, several problems in using the indentation method on ZrO2 ceramics were discussed, focusing on phase transformation. Residual stress in Al2O3 and Si3N4 was nearly eliminated by annealing the specimen after hand grinding. Consequently, this treatment method is considered effective for determining the basis value KC. The estimated residual stress value in Al2O3 and Si3N4 obtained by the Vickers indentation method at 98N corresponded closely to the values obtained by the X-ray method. The value for ZrO2, however, was larger than that found by the X-ray method. This difference is assumed to derive from phase transformation.
The thermal residual strain in the three layered materials, [WC-10mass%Ni]-[Ni]-[WC-10mass%Ni], [WC-10mass%Ni]-[SUS304(plate)]-[WC-10mass%Ni] and [WC-10mass%Ni]-[Ni(plate)]-[WC-10mass%Ni], has been investigated by neutron diffraction measurement. They were made by self-propagating high-temperature synthesis. Original materials of composite [WC-10mass%Ni] are powders of W, C and Ni, and those of the middle layers are the powders of Ni, a plate of SUS304 and a plate of Ni. The samples were 40mm in diameter and about 4mm in thickness of each layer. The result of the measurement shows that the middle layer made from powders of Ni has no strain, which suggests that it shrinks from high temperature freely from the existence of both side layers. On the other hand, the middle layer made from a plate material experiences complex stress according to each position and to each direction of the sample, which suggests that it shrinks from high temperature in a state of tight binding on both side layers. The Ni region in the composite material, [WC-10mass%Ni], has a large tensile strain of 0.6 to 1.0%, whereas WC region has a negligibly small compressive strain. In this case, the Ni region shrinks under the hard connection with the WC region, of which the thermal expansion coefficient is relatively small.
Pin on disk friction tests were carried out for three types of metals, 316L S. Steel, Co-28Cr-6Mo, and Ti-6Al-4V alloys in PBS(-) to estimate the range of loads that can be applied to the bearing surface of a joint prosthesis, based on the degree of damage to the passivation film. The results were as follows: (1) When two metals of the same type were rubbed against each other, the passivation film was almost completely destroyed immediately after the onset of friction, even when a very low normal load was applied. (2) In the metal to UHMWPE friction test, the passivation film was severely damaged immediately after the onset of friction. However, the passivation film recovered until the polarization potential reached a certain level, which was specific to each metal. (3) The allowable load on the bearing head was calculated, assuming that the degree of damage to the passivation film was 20% or 40%. When a metal acetabulum socket was used, damage to the passivation film on the head was severe, regardless of the type of metal used for both the acetabulum socket and the head. When a UHMWPE acetabulum socket was used, the allowable load ranged from 8.16×102N to 1.31×105N for 316L S. Steel, from 7.83×103N to 1.31×105N for Co-28Cr-6Mo alloy, and from 43N to 6.08×102N for Ti-6Al-4Valloy.
The mechanism of decrease in the corrosion resistance of Cr-plated parts when they are subjected to a heating process and measures for preventing the degradation are studied. When postfinishing is performed after the Cr-plating process, cracks in the Cr layer are closed as a result of plastic flow on the topmost surface layer. When the residual stress in the Cr layer becomes compressive due to the postfinishing, the cracks are completely closed, resulting in a high corrosion resistance. The residual stress in the Cr layer changes from compressive to tensile following heat treatment. When the residual stress becomes tensile, the cracks in the Cr layer which were closed during postfinishing open once again, resulting in a decrease in the corrosion resistance. We speculated that the change of residual stress from compressive to tensile is caused by the shrinkage of the Cr layer due to heat treatment and the difference in the coefficients of thermal expansion between the substrate (steel) and Cr layer, as well as the release of residual stress generated during processing. In order to prevent the decrease of corrosion resistance under heat treatment, it is effective to apply compressive stress which exceeds the level of change in residual stress due to heat treatment to the Cr layer during postfinishing, after the Cr plating process.
Artificial aggregates made from the natural rocks are usually used as the concrete material, since the natural aggregates along the rivers cannot be used due to the strict restriction to maintain the human society. The shape of these aggregates distinctly varies depending on the step of the producing processes such as blasting, crushing and screening. In this study, a quantitative analysis for the shape irregularity of such aggregates was attempted by applying the concept of fractal proposed by B. B. Mandelbrot. The Richardson effect was clearly confirmed between the total surface S and the measuring unit length ε. This fact indicates that the shape irregularity of the aggregates has the fractal aspect and the present method is applicable to the shape irregularity analysis. Thus it was finally found that the shape irregularity of the aggregates was successfully evaluated by combining the fractal dimension D and the body-shape irregularity index D*.
Through the phase-separation system composed of phenols and concentrated acid, the phenol components are selectively hybridized into C1-positions of phenylpropane units of native lignin to give 1, 1-bis(aryl)propane-type linear polymers (lignophenols) quantitatively. Lignophenols have unique properties such as high phenolic activity and obvious phase transition. Furthermore, the functionality of lignophenols was able to be controlled using the intramolecular switching function: Under alkaline condition, C1-phenols in lignophenols nucleophilically attack adjacent C2 as switching devices, followed by the cleavage of aryl ether linkages. In ligno-p-cresol and ligno-2, 4-dimethylphenol with similar switching functionality and structural characteristics, ligno-p-cresol was highly hydroxymethylated (HM) on the cresolic nucleus to give network-living pre-polymer. On the other hand, ligno-2, 4-dimethylphenol having reactive sites mainly on the terminal unit gave linear-living pre-polymers. The structures of polymerized lignophenols were controlled by mixing ratio of HM-ligno-p-cresol and HM-ligno-2, 4-dimethylphenol or by the hybridization ratio of p-cresol and 2, 4-dimethylphenol in lignophenols followed by hydoxymethylation. The resulting polymer chains were cleaved at the switching points to give low molecular weight subunits by the switching function. Recyclable composites were prepared by the combination of cellulose and HM-lignophenol (the mixing ratio of HM-ligno-p-cresol to HM-ligno-2, 4-dimethylphenol was 1:0, 1:1, 0:1, respectively). The composites had glossy surfaces and looked like wood. The dimensional stability of composites was improved significantly with increasing mixing ratio of HM-ligno-p-cresol to HM-ligno-2, 4-dimethylphenol. Using the switching functionality of lignophenol, the composites were re-separated into lignophenols with low molecular weight and cellulose.
The crystal phases, microstructural features, and local atomic coordination of CuO-Al2O3 catalyst for NOx removal were studied by some experimental techniques. A spinel-like phase, which exhibits short-range structure, appears at around 900°C. Catalytic activity can be maintained at <900°C, but degenerated at >1100°C. The activity of catalyst is controlled by the Cu-containing crystal phase of surface spinel. At low heating temperatures, the diffusion of copper into alumina will be important to surface modification to create active sites. The ESR and XANES results suggest that atomic structures in the interfacial area appear to play an important role in the lean-NOx removal activity.