Key Engineering Materials Vols. 345-346

Abstract: A roller mill is composed of several rollers, rotational table liners, hydraulic cylinders and raw materials are ground between the rollers and the table with the assistance with a shearing and compressive force. It has been reported that an unexpected fatigue failure occurred in a table liner in the course of grinding portland cement. It is subjected to the cyclic bending stress by the rollers load and the centrifugal force by the table rotation and fractures at the edge of grinding path of outer roller. It demands design life 4×107 cycles but has fatigue life 4×106 ~ 8×106 cycles. The purpose of this study is to reveal the failure mechanism of the table liner occurring in the grinding operation of roller mill and to show the structural design guidelines for reducing the fatigue failure.

Abstract: The present work is a further investigation into the effects of the carbon (C), nitrogen (N) and niobium (Nb) contents on then fracture properties of the Type 347 stainless steels at 316oC. 9 heats of systematically designed alloys were examined. Through SEM-EDS, TEM and XRD analyses, two kinds of precipitates, Nb(C,N), CrNbN were identified in the Type 347 steels with a high ratio of wt% N to wt% C, on the other hand only Nb(C,N)s were found in the Type 347 steels with a low ratio of wt% N to wt% C. The tearing moduli were decreased in the range of 52~60% as the carbon content increased from 0.03wt% to 0.05wt%. The tearing moduli were lowered by 52~59% in the alloys with a high nitrogen. It was deduced from the microstructure analysis results that the coarse Nb-rich precipitates control the fracture resistance of the Type 347 as they act as the potential sites for the nucleation of micro-voids.

Abstract: P92 steels as well as P91 are widely used as boiler tube materials of ultra super critical (USC) power plants these days. And thus embrittlement is very important for structural integrity of the USC plants. The embrittlement was observed when P92 (Modified 9Cr-1.8W-0.5Mo-V-Nb) steels were quenched to and held at the temperature of 320 to 350°C, which were the temperatures intermediate between Ms and Mf, and then air-cooled. Nearly same kind of the embrittlement had been observed with the T/P91 steels and a theory had been proposed to explain the mechanism of the embrittlement by us. From the theory, the embrittlement might be caused by the brittle martensite which is freshly formed during air-cooling. We tried to apply the theory for the embrittlement of the T/P92 steel. The behaviors of the embrittled T/P92 steel were explained well by the theory.

Abstract: We developed the 3-D local hybrid method to evaluate the 3-D stress field inside the specimen from displacement data on the free surface obtained from the 2-D intelligent hybrid method. When a uniform load was applied to the structure with a surface crack, high accuracy was already acquired in stress analyses. The 3-D local hybrid method was anew applied to a structure with a surface crack which is subjected to bending load. It is expected that the accuracy depends on local model size. In this study, the width, the thickness and the height of the local model were changed widely, and analyses were carried out. Then the size of the local model necessary for the analyses was examined. Assessment of analyses was performed by comparing J integral value of a whole model and a local model.

Abstract: The transverse crack, which is the most dangerous damage among the various types of rail defects, is developed from shelling near the rail running face, and grows perpendicular to the rail surface. Therefore, to assure the safety of rolling stocks, it is necessary to investigate the fatigue crack growth behavior of a gas pressure welded part of rail steel under mode I and mixed mode loadings. In this study, the fatigue crack growth behavior under the mixed mode was discussed by using comparative stress intensity factor ranges proposed by Richard. In addition, to evaluate the effect of the crack closure quantitatively, the crack opening load was determined a using clip-on gauge and computerized image processing system. As a result of this, a mixed mode crack growth rate regardless of the stress ratio could be correlated with the effective comparative stress intensity factor ranges derived from the crack opening loading.

Abstract: Stress and displacement fields for a transient crack tip propagating along gradient in functionally graded materials (FGMs) with an exponential variation of shear modulus and density under a constant Poisson's ratio are developed. The equations of transient motion in nonhomogeneous materials are developed using displacement potentials and the solution to the displacement fields and the stress fields for a transient crack propagating at nonuniform speed though an asymptotic analysis.

Abstract: Semiconductor devices are usually formed on a single silicon wafer during a batch processing method. Individual devices are separated from the wafer during the wafer sawing or dicing step. Subsequent packaging processes are then performed on the individual devices, whose edge portions are very susceptible to mechanical damage from the sawing process. Defects formed along device edges due to the dicing saw blade often provide potential sites for serious reliability problems. If the scribing area is reduced, the number of the separated devices from a single wafer increases, which results in productivity improvement. However, the closer the scribing position of the saw blade comes to the active device pattern, the greater possibility of sawing-induced damage to the active pattern is. Thus, this work shows methods to reduce the negative impact of the saw blade while maintaining close proximity of the scribe lines to the IC devices. In particular, this work suggests that a decrease in the size of the diamond particles embedded in the saw blade and in the rotation speed of the saw blade might contribute to the prevention of sawing-induced damage to device patterns.

Abstract: For explaining the SCG behavior of polyethylene, the crack layer theory is applied based on the description of two driving forces: crack and PZ. The relations between the speed of SCG, crack length and elapsed time are the most important characteristics of polyethylene resistance to crack propagation, or long-term brittle fracture. The crack layer model of slow crack growth in polyethylene is designed in such a way that it qualitatively reproduces the main features of the process indicated above and makes it possible to quantitatively match any pattern of step-wise crack growth. In this paper, the behavior of SCG of polyethylene is developed for numerical simulation based on the crack layer theory. Some parametric study and applications are addressed based on the developed simulation program.

Abstract: The process zone (PZ) that surrounds and precedes a crack is a common feature of fracture in engineering polymers. Depending on the material, the specimen geometry, the temperature, and the loading conditions various types of microdefects such as crazes, shearbands, microcracks, micro-voids, etc, constitute the process zone. The microdefects are formed in response to stress concentration, and shield the crack tip from high stress level. There is a complex crack – damage interaction, which is briefly addressed by means of a semi-empirical method. On a continuum mechanics level, the PZ appears as a domain with effective elastic properties different from that of the original material. The crack and PZ evolve as one system with multiple degrees of freedom. It is regarded as a Crack Layer (CL) in contrast with the conventional image of crack as an ideal cut. There are thermodynamic forces responsible for CL growth, which are defined as derivative of Gibbs free energy with respect to the corresponding CL “coordinates”. The thermodynamic forces can be expressed as integrals of the Energy Momentum Tensor of elasticity. Onsager type relations between CL growth rates and corresponding CL forces constitute a system of constitutive equations for CL propagation. Examples of solution of these equations, and comparison with experimental data as well as with conventional models are presented in accompanying paper.