Magnesia powders having different particle sizes were prepared by a vapor phase oxidation process, and the effect of particle size on their properties was investigated. The particle size of magnesia powder increased with increasing the flame length of combusting magnesium vapor which was affected by the diffusion coefficient in the flame and the evaporation rate of magnesium. The magnesia powders obtained were highly pure, and especially the contents of U and Th were both below 1ppb. The properties of magnesia powders were strongly affected by the particle size: The green density increased with increasing particle size, whereas the lattice constant decreased with increasing particle size. There was an optimum particle size to achive the maximum sintered density.
A new preparation method of high-purity ultrafine nitride particles was demonstrated by applying the low temperature plasma induced by 2.45GHz microwaves to the gas evaporation technique. Ultrafine Al, Fe and Si particles produced by the evaporation of these metals in a low pressure N2 gas condition were nitrided by passing them through the nitrogen plasma region. From the electron diffraction, the following nitride phases were recognized as AlN, γ-Fe (solid solution of nitrogen), γ-Fe4N, ε-FexN (2<x≤3), ζ-Fe2N, α-Si3N4 and β-Si3N4. The formation of these phases depended mainly on three factors: N2 gas pressure, the gas velocity at the plasma region and the microwave power. The preferable conditions of these factors for preparing pure nitride particles were investigated. For example, the particles formed at the condition of 7 Torr in N2 pressure, 1m/sec in gas velocity and 1.2kW in microwave power consisted completely of nitrides of the starting material evaporated. In the cases of Fe-N and Si-N system, a mixture of those nitride phases was formed generally. Particles tended to grow in the plasma region. The particle size of nitrides ranged from 20 to 300nm, which was found to be larger by a factor of 5 to 10 than that of particles collected at the front of the plasma region.
The effect of atmosphere of heat treatment on the formation of tungsten borides was investigated by the solid state reaction between tungsten and amorphous boron, in which the formation ranges and synthetic conditions of WB and W2B5 solid solutions were determined. The low temperature activation treatment at 500°C in a hydrogen stream promoted the boride formation in the subsequent heat treatment. Since a coexisting phase tended to from in the case of WB synthesis, the pretreatment at about 1000°C in an argon stream prior to the final heat treatment, was required to obtain a homogeneous composition of specimen. WB solid solution had a narrow formation range of B/W=1.0-1.2, while W2B5 solid solution had a wide formation range of B/W=1.75-2.75. The formation amount of WB or W2B5 solid solution showed the maximum at the composition of mixed powder which contains 10at% boron in excess of the stoichiometric composition of each boride. The synthesized tungsten boride powder with uniform grain size of 0.5-2μm was expected to have a good sinterability.
Generally, a slip casting process is closely related to the flocculation state of particles. The present investigation was undertaken in order to understand the relationship between the flocculation state of particles and the packing structure on the surface of gypsum mold. The flocculation state of particles in Al2O3-slurry was discussed on the basis of rheological behavior as a function of dispersant concentration. As the result, it was found that the type of rheological properties was classified into three modes. For these slurries, the absorbing rate and pressure of water and the increase of packing layer thickness were measured in order to explain the packing structure on the surface of mold. From these results, it was shown that the most suitable condition for slip casting could be prepared by using a weakly flocculated slurry.
The surface properties of silicon carbide and its surface-treatment with cetyl alcohol were investigated through X-ray photoelectron spectroscopy, X-ray diffractometry, scanning electron microscopy, dispersive properties and pyrolysis in comparison with those of silicon dioxide and activated carbon. The following results were found: (1) the surface of silicon carbide was coated by silicon dioxide, and the deposited carbon and their surface functional groups were composed of ≡SiOH from SiO2 and _??_CO, ≡COH, -C=O-OH and≡CH from activated carbon, (2) The surface-treatment affected only the surface properties of silicon carbide, but not the properties of the substrate such as crystal structure, particle size and the agglomerated state of the particles, (3) The surface-treated silicon carbide with cetanol became hydrophobic, while the untreated one was hydrophilic. This fact indicates that the surface properties of silicon carbide can be effectively improved by the surface-treatment with cetanol, (4) The IR absorptions of CH2 and CH3 due to CH stretching vibrations were observed in FTIR spectra of the surface-treated silicon carbide, (5) The pyrolysis of the surface-treated silicon carbide proceeded at the temperature range from 250 to 500°C and the main pyrolysis product was cetene. This shows that the surface groups consist of alcoxy groups formed by the dehydration condensation of surface hydroxides and alcohol.
The adsorption of dyes (direct, acidic, basic, reactive and disperse) on Itaya zeolite consisting mainly of clinoptilolite and mordenite was investigated. The adsorption isotherms were found to be approximately expressed by the Freundlich equation. The order of adsorption selectivity was basic dye>acidic dye>other dyes. The Itaya zeolite surface-treated by the disperse dye and binder has less adsorbing power to nitrogen than the untreated zeolite, but has about the same hydrogen and ammonium ion exchange capacity. Therefore, it is useful for soil for gardening, an ornament rock for cleaning of water in an aquarium and for building materials (for example, breathing walls).
Kneading torque and apparent viscosity were investigated on several series of alumina-thermoplastic resin systems selected from six kinds of alumina and three kinds of resin. The curves of kneading torque vs time at 453K were classified into four patterns which were dependent on resin type and alumina grade. The change of the pattern was also brought about by the change of solid concentration in the same alumina-resin system. The torque after 1h showed a good correlation with the apparent viscosity measured at the same temperature by a capillary type viscosimeter. The apparent viscosity of alumina-polypropylene and -polystyrene systems at high solid concentrations of 50% and 60% was influenced by the tap or compaction density of powders much more than by their surface area or size. The relative compaction density was an excellent parameter to estimate the flow and kneading properties at high solid concentrations. The alumina-polyethylene system, however, behaved somewhat curiously, because the type of degradation of polyethylene caused by the mechanical and thermal mixing was different from that of polypropylene or polystyrene.
Homogeneous powder packing into a powder cell is one of the fundamental and also important problems which should be urgently elucidated in powder technology. For example, to get a sufficient mechanical strength of sintered powder chips in fine ceramics or electronics industry, a good homogeneous powder packing method should be developed. However, it has not been sufficiently established up to now. In this paper, from such a point of view, the initial powder packing states in different powder cells with one to five powder feeding points were systematically radiographed by X-ray, and their initial powder porosity (or powder bulk density) distributions were quantitatively analyzed by applying the analysis method presented by the authors. As the result, the following conclusions were obtained. (1) In every type of powder packing, the minimum porosity column was generated just under the powder feeding points. (2) The interference effect between two neighbouring powder feeding points changed remarkably depending upon test powder. That is, the porosity at the central region between two feeding points changed from the minimum value of the porosity distribution to the maximum one with the change of test powder.
In the previous papers, a strict method of calculating electric resistance of a binary particles system was described. In that method, the total electric resistance was calculated by applying Kirchhoff's law. In the present paper, in order to simplify the calculation of total electric resistance (R) of a binary system, an approximation (series and parallel) method was proposed whereby R could be calculated rapidly without any trouble of overloading a computer system. The symbol R (total electric resistance) presented in the previous papers was now rewritten as RK in order to draw a distinction from the value of approximation (RS). This approximation method is considered particularly applicable to a binary particles system containing a large number of particles. Furthermore, its application is effective when the difference in electric resistance between two groups of particles is not so much. In such a case, RS can be calculated rapidly even by a computer which has not so large capacity.
Solid state welding of Ti/cermet and Ti/TiC was conducted under phase transformation of Ti in air. They were joined well under the load of 0.6MPa and the number of thermal cycle more than 5 (the welding time was 5min), but not joined at the constant temperature of 1233K for 30min. There existed two layers at the welding interface of Ti/cermet, but no layer at the welding interface of Ti/TiC. The strength of welds of Ti/cermet increased up to 102MPa with increasing number of thermal cycle, while the strength of welds of Ti/TiC had no relation with the number of thermal cycle, and the maximum strength was 24.5MPa. The reflectivity of ultrasonic wave at the welding interface increased with the strength of welds.
Failure in the Brazilian test (diametral compression of a disc) has been numerically analyzed on the basis of the Griffith fracture criterion. The analysis indicates that the angle subtended by the loaded arc strongly affects the position and the load of failure. Failure occurs away from the center of the disc for small angles, but it initiates at the center for the angle over a certain degree. The load of failure increases with increasing angle, and the derivative of the load against angle changes abruptly at that certain angle. These results agree with the experimental studies performed by others. It is erroneous to calculate the splitting tensile strength as P/πRL without respect to the angle subtended by the loaded arc, where P is the load of failure, R is the radius of the disc, and L is the thickness of the disc. A definition of the splitting tensile strength is suggested. For practical purposes, the suitable length of the loaded arc and an approximation of the splitting tensile strength are suggested.
Prestrain has been pointed out to be one of the most significant factors to degrade plastic deformation capacity and fracture toughness of structural members. In this paper, resting on this viewpoint, fracture toughness tests and finite element analysis were carried out on notched plates of structural steels subjected to prestrain in order to clarify their mechanical properties. The analytical and experimental results were compared with regard to the J-integral and deformation characteristics around crack tip. The conclusions drawn through these studies are summerized as follows. (1) The non-dimensional J-integral can be expressed as a function of the overall strain, ε, regardless of the prestrain, and hence the interrelation between J and ε can be evaluated through the rigid-perfectly plastic analysis. (2) The plastic constraint influences greatly the correlation between J and the crack tip opening displacement, CTOD, or the contraction ahead of crack tip in the thickness direction, δz. (3) The critical values of J, CTOD and δz at the initiation of stable crack growth, Jin, δin and δzc, decrease as the prestrain increases. The critical J-value, Jin, is represented as a function of δin or δzc. Moreover, Jin correlates closely to the uniform elongation of material.
The surface small crack growth behavior of SUS304 stainless steel under fast-fast and slow-fast strain controlled low-cycle fatigue conditions was investigated at 873K with the specimens having different grain sizes. It was shown that the cumulative probability of crack length distribution in both fast-fast and slow-fast cyclings could be represented by a Weibull distribution as a function of crack length in the early stage of fatigue life. In the stage after the half fatigue life, the cumulative probability of crack length distribution showed a knee, and a small number of main cracks grew predominantly. The knee occurred when the crack length was comparable to one grain size. It was also shown that the surface small crack growth rate showed the minimum value when the cracks arrived at the grain boudaries. The above behavior resulted from that the grain boundaries temporarily impeded the small crack growth. Furthermore, the small crack growth rate was compared with the macroscopic crack growth rate. The law for the macroscopic crack growth was applicable to the small crack growth in slow-fast cycling, up to the crack length being one grain size. On the other hand, the small crack growth rate in fast-fast cycling was about ten times as large as the macroscopic crack growth rate in the region where the crack length was smaller than or comparable to the grain size.
Low-cycle creep-fatigue tests of the dissimilar metal electron beam welded joint between A387 Gr. 22 steel and SUS405 steel were carried out under the strain-controlled cycling over the welded joint at 873K. It was shown that the creep-fatigue life of the dissimilar metal welded joint was significantly shorter than those of the base metals. This resulted from the strain concentration on the SUS405 side with lower deformation resistance. It was also found that the hardness distribution was one of the important measures which reflect the strain distribution. Furthermore, the predicting method of creep-fatigue life for dissimilar metal welded joints was proposed by applying the strain range partitioning approach. The predicted lives were in good agreement with the experimental results.
This paper describes the effect of the welding residual stress on the fatigue strength of welded joint. The fatigue strength was measured on Aluminum alloy, mild and stainless steels under the condition that the factors affecting the fatigue strength of welded joint were not changed except the residual stress produced by electron beam welding. When the residual stress at the initiation area of fatigue was compressive and did not vanish during the fatigue test, the fatigue strength was improved. When the tensile residual stress existed at the fatigue initiation area, the following two cases appeared: The fatigue strength of a non-quench hardening material as well as that of a quench hardening material with high ductility and low yield ratio were not influenced by the tensile residual stress. However, the fatigue strength of a quench hardening material with poor ductility and high yield ratio was lowered by the tensile residual stress.
The cyclic fatigue behavior of different kinds of ceramics was investigated at room temperature. Flexural specimens, with an indentation induced flaw at the center of the reduced gage section, were tested under a cyclic load applied by four-point bending. All ceramics tested were shown to be susceptible to cyclic fatigue failure. The cyclic fatigue degradation compared to its flexural strength increased in this order: hot-pressed AlN, sintered SiC, hot-pressed Si3N4, sintered Si3N4, and sintered Al2O3. The apparent fatigue limit was about 80% of the average flexural strength for AlN, and about 60% for sintered Al2O3. There existed a correlation between the scatter in fatigue and flexural strength. The scatter in fatigue strength was smaller than the scatter in flexural strength. SEM fractography revealed a dominant intergranular fracture in Al2O3 and Si3N4, which suggests a possible stress corrosion cracking mechanism in the grain boundary glassy phase. On the other hand, hot-pressed AlN showed a mixed mode fracture of intergranular and transgranular cracking, and SiC exhibited transgranular cracking only.
It was already reported that many bubbles were observed gushing from the tip of cracks in fatigue tests in liquid environments. These bubbles occur due to the cavitation accompaning with the opening and closing action of the crack. In this study, the influence of cavitation bubbles on the crack growth behavior was investigated at various load frequencies in synthetic seawater and in ion exchanged water. The following conclusions were drawn. (1) The ratio of crack growth rate in a liquid environment to that in laboratory air at a constant ΔKeff was obtained as a function of load frequency. This curve showed a constant in low frequency range (<0.2Hz) and in high frequency range (>5Hz), respectively, and a transition range was observed between these constant ranges. (2) It may be considered that the corrosion fatigue crack growth at high frequencies in liquid environments was reduced to the almost same value as in the laboratory air, because the cavitation bubbles were gushing at the tip of the crack.
The statistical fatigue tests were carried out on S38C steel specimens in two environments by using two sets of rotating bending machines, each capable of testing ten specimens at the same time. One environment was an uncontrolled room atmosphere and the other was a dry atmosphere where the relative humidity at 21°C was reduced to about one percent. Fifty specimens were equally allocated to each stress level, covering the failure probabilities from zero to 100 percent. From the experimental results, Weibull three parameters were evaluated by maximizing the correlation coefficient, and their stress dependence was investigated. Moreover, the beta distribution was used to approximate the distribution of endurance limit and good agreement was obtained throughout the failure probabilities from zero to unity for both environments. The effect of the reduction of humidity was distinct: the fatigue life as well as the endurance limit were improved in a dry atmosphere.
A method of insuring a reasonable reliability for the reinforced concrete member was proposed by using the probabilistic method and considering the variations in strength of both reinforcing bars and concrete. This method was applied to obtain the reduction factors for the design axial load strength of compression member and the balanced and minimum steel ratio of flexural member, corresponding to the specified reliability, and to decide the applicable quality of materials for design moment strength of flexural member with a specified strength reduction factor. The results were compared with the values provided in the ACI Code.
Extrusion characteristics are intimately associated with the product properties of building materials. In this paper, both extruding pressure and extruding velocity were measured to investigate the extrusion characteristics of mortars. The mortars were prepared from fine silica sand aggregates of various sizes (6 kinds of silica sand and standard sand) and had different water contents. The water content ratio which resulted in good surface appearance of moulding was little higher for the fine aggregates than for the coarse ones. The extruding pressure dropped abruptly with an increase of water content ratio. The extruding velocity increased with increasing water content ratio. The upward trend of extruding velocity declined at high water content ratio. The water content ratio had much more influence on extruding pressure and velocity than the grading of aggregates had. In the cases of organic thermoplastics and clay like ceramics, the extruding pressure and velocity have been known to be prescribed by the extruder operating point determined by a cross point of screw characteristics and die characteristics. It seems to be possible to apply this operating point concept to the extrusion of mortar.