Toughening mechanisms hitherto proposed, such as microcracking, crack-path-deflection, crack-bowing, and transformation-induced-plasticity, are briefly introduced. The increase in crack resistance (rising R-curve) due to the process zone wake is discussed in relation to test methods and precracking techniques. Stable crack extension is apt to be accompanied by the increase in crack resistance due to the wake of transformation or frictional interlocking, whereas pop-in crack extension seems to be little attended by these phenomena. The loading rate dependence of KIC value is also discussed. The KIC value can be affected by the time dependent phenomena such as stress corrosion cracking even at room temperature and crack tip blunting, stable crack extension, etc., at elevated termperatures. Various test methods tentatively used to evaluate the critical stress intensity factor, KIC for ceramic materials are explained, and their advantages and disadvantages are discussed, partly referring to the results of the round robin tests organized by the Japan Fine Ceramics Association (J.F.C.A.). The bridge-indentation method, a new precracking technique, is also briefly explained.
Though the high efficient grinding technique for ceramics using a high rigid machining center with a durable diamond wheel has recently been developed, the effect of grinding flaws on material strength has not yet been clarified. So experimental investigation was conducted for evaluating the effect of machining on strength of silicon nitride, and the effects of grinding wheel, grit size and depth of cut were investigated. The strength data were analyzed by Weibull statistics. The depth of flaw in the surface, induced by high efficient grinding perpendicular to tensile stress, was about 40 μm. The strength of samples ground parallel to tensile stress was not decreased compared with the strength of lapped samples. The high efficient grinding generated compressive residual stress of 590 MPa in the surface.
Sinter+HIP which is the most common HIP method is composed of sintering and HIPing processes. Both are performed separately and continuatively in two furnaces. On the other hand, in Sinter/HIP, sintering and HIPing is performed in one furnace. Therefore, Sinter/HIP eliminates extra heating-up, cooling, and handling in Sinter+HIP. In the present work, at full dense Si3N4 sintered body which was added 6 wt% Y2O3 and 2 wt% Al2O3 as additives was tried to fabricate by Sinter/HIP, whose cycle was composed of the following stages; 1) hot evacuation; at 1 400°C, for 2 h, under 70 Pa in vacuum 2) gas pressure sintering; at 1 800°C, for 2 h, under 1 MPa of N2 3) HIPing; at 1 750°C, for 3.5 h, under 190 MPa of 0.5 vol% N2+Ar. At the same time, gas pressure sintering was attempted in which its heating pattern was just the same as that of Sinter/HIP, but its gas pressure which was kept constant was 1 MPa of N2. Three point bending strengths of Sinter/HIPed bodies measured at room temperature and 1 200°C were 1.09 GPa and 605 MPa, respectively. On the contrary, those of gas pressure sintered body were 814 MPa at room temperature and 420 MPa at 1 200°C. Porosity of Sinter/HIPed body was 0.15 % which was much smaller than 0.98% of porosity of gas pressure sintered body. Large pores of 310 μm were observed in gas pressure sintered body, whereas they were eliminated in Sinter/HIPed body. It was concluded that the increase in strength by Sinter/HIP was due to the decrease in porosity and elimination of large pores.
Sinter/HIP is a new and efficient way to fabricate a high strength sintered body, compared to Sinter+HIP. Appropriate Sinter/HIP conditions were determined on this study to fabricate a high strength Si3N4 to which 6 wt% Y2O3 and 2 wt% Al2O3 were added as additives. Sinter/HIP was composed of the following stages; (stage 1) hot evacuation, (stage 2) gas pressure sintering, and (stage 3) HIPing. Important factors among these three stages were (1) N2 gas pressure at stage 2 and (2) gas, pressure, and time at stage 3. N2 gas pressure at stage 2 was found to be the most influential factor on mechanical properties of Sinter/HIPed body, especially its strength. To fabricate a high strength Sinter/HIPed body, N2 gas pressure at stage 2 must be less than 1 MPa. Pure N2 gas is usually used as HIPing gas. However, the mixture gas of 0.5vol%N2+Ar could be used as HIPing gas. Because three point bending strength of Sinter/HIPed body using 0.5vol%N2+Ar was almost same as that of sinter/HIPed body using N2. HIPing time should be short on the condition that a sample could be densified fully, because high temperature (1 200°C) strength decreased slightly with increasing HIPing time. With regard to HIPing pressure, 120 MPa was sufficient to fabricate a high strength Si3N4.
The microstructure evolution and grain growth during high-temperature heat treatment in ZrO2–4 and 8mol%Y2O3 alloys were examined. ZrO2–8mol%Y2O3 alloy was of single cubic phase (c-ZrO2), while ZrO2–4mol%Y2O3 alloy consisted of mixed grains of c-ZrO2 and tetragonal phase (t-ZrO2) with the equilibrium yttria contents after a certain period of heating. This structure was different from the microstructure containing precipitates in grain interior, and was referred to be the dual-phase structure. The grain growth of dual-phase structure was much slower than that of single c-ZrO2. The grain growth kinetics of dual-phase structure under the equilibrium partitioning of yttria between c-ZrO2 and t-ZrO2 grains could be described by the rate equation which was derived by assuming the growth of individual grains limited by grain-boundary diffusion of yttrium ions.
Thermogravimetric (TG) data on reoxidation behaviour of a particular type of sponge iron (coal based sponge iron) are mainly reported in this paper. While emphasis is on isothermal kinetics, some nonisothermal TG data are also presented. The data have been analysed using the reduced time plot and the activation energy values evaluated using both integral and differential approaches. The kinetic analysis developed has been used to establish the reaction mechanism which is found to be essentially of the first order. The paper also discusses the general importance of high temperature studies in the assessment of reoxidation behaviour of sponge iron and how the outcome of this investigation may be useful for devising means to protect sponge iron against reoxidation and spontaneous combustion.
A mathematical model for particle size distribution in bell-type charging at blast furnace top was developed, which is capable of estimating the radial average size distribution as well as the radial deposit distribution of each particle size for multiple size burden with a few fixed parameters. The following findings were considered in the construction of the model. The radial deposit distribution of particles in blast furnace was determined by both the sieving process and the deposit process on the slope. The elemental process in the sieving process is the percolation of small particles through voids between large particles acting as sieves. Principal factors affecting the percolation are the ratio of large particle size to small particle size and the particle velocity gradient in the flowing layer on the slope. The validity of the model was confirmed through actual filling tests at various blast furnaces with different throat diameters and charging conditions. The model was applied to the prediction of the change of particle size distribution caused by the decrease of large bell stroke at Wakayama #4 BF. The monitored results showed the decrease of gas flow resistance and the suppression of peripheral gas flow as were predicted by the model with help of a gas flow simulation model.
This paper describes the effect of graphite crystallization morphology on shrinkage cavity formation of hypereutectic spheroidal graphite iron cast in metal mold. Metal mold used can be considered as designed for riserless casting. In the melts containing 3.5–3.6%C, large spheroidal graphite nodules were observed with increasing silicon content beyond 3.0% (carbon equivalent beyond 4.5), and their counts increased with increasing silicon content (carbon equivalent). When the number of these large graphite nodules was more than 10 counts/mm2, shrinkage cavities were seen to be generated, whose area then increased with increasing large graphite nodule counts. In the melts with high carbon content of 3.8–3.9%, large graphite nodules of 25–30 counts/mm2 were observed in low silicon content range from 2.6 to 3.0%. In this range, however, the shrinkage cavity could not be found through detailed observation. In the melts with low carbon content about 3.3%, on the other hand, shrinkage cavities were produced in spite of no presence of large graphite nodules even in 3.7% Si content. Large graphite nodules were considered to be generated in spheroidizing treatment and to have been in existence before pouring. This caused the ratio of those graphite area which was crystallized in metal mold after pouring to decrease. It was followed by the decrease in expansion of cast iron, subsequently by the shrinkage cavity formation. It was quite clear, from the study on graphite morphology with image analysis, that sound cast irons were produced when graphite amounts corresponding to the carbon content more than 3.3% was crystallized during the solidification in metal mold.
Fretting fatigue tests of high strength steels having ultimate tensile strengths of 490, 690 and 880 MPa were carried out in sea water under free corrosion condition, and the effect of fretting damage on fatigue life was investigated. At high stress amplitudes, the higher the strength of steel, the longer the fretting fatigue life in sea water, and the fretting fatigue life in sea water was almost the same as that in air. At low stress amplitudes, the fretting fatigue life of all the steels in sea water decreased to 10-20% of the plain corrosion fatigue life and showed almost the same life irrespective of the strength of steel. The fretting fatigue life in sea water was much shorter than that in air, and the fretting fatigue strength at 107 cycles in sea water was lower than that in air. The saturation of fretting fatigue demage occurred beyond a certain number of fretting cycles. For 880 MPa grade steel, the smallest number of fretting cycles to cause the saturation of damage in sea water was less than 0.005% of the fretting fatigue life, but that in air was about 30%. It was shown that the fatigue life in sea water was reduced by the fretting damage produced by a small number of fretting cycles.
The α/γ equilibrium in the ternary Fe–Mn–Al system was experimentally investigated at 1 073, 1 273 and 1 373 K. The tie-lines in the (α+γ) two-phase field were determined at 1 273 K by measuring the alloy content of the individual phases of equilibrated two-phase specimens by means of X-ray microanalysis. These experimental results were compared with the calculated α/γ phase boundaries using a regular solution model for the ternary system which was extended from the constituent binary systems. It was found that the calculated results were in reasonable agreement with the experimental measurements in the low manganese contents, approximately less than 10 wt%.