The high temperature reduction properties of processed iron ores in packed beds were investigated as a function of applied load and reduction transition. The microstructures of cohesive packed beds quenched during the reduction test were also studied quantitatively by the use of an electron microprobe. The results obtained are summarized as follows; (1) At a given temperature, the softening viscosity of dolomite-fluxed pellets was found to have minimum value at a certain reduction degree, which tends to increase as the temperature decreases. This behavior may be explained due to the competition of the increase of porosity and the sintering of metallic iron formed by reduction. (2) The permeability resistance of the packed bed during reduction under load was found to be estimated accurately based on Uchida's model even after the irrigation by liquid flowed-out from wustite core. (3) The melting-down of cohesive bed was found to start from the formation of Fe-C liquid at the surface of iron largely grown during reduction of molten wustite with solid carbon. The fact that the surface carbon concentration of the coagulated iron increases with the reduction of the FeO content in the coexisting silicate melts shows the possibility that melting-down phenomena would be controlled by the reaction equilibrium among them.
The liquid phase formation process and its effect on softening and melting-down properties were investigated in a synthetic quinary MgO-Al2O3-SiO2-CaO-iron oxide (or metallic Fe) system. The increase in CaO/SiO2, MgO and also in T.Fe was found to be effective to improve the softening properties at the wustite reduction stage by increasing solidus temperatures and by reducing amount of liquid developed at solidus temperature, respectively. Even at reduction stage of metallic iron, where network of iron sustains the deformation stress, liquid formation in coexisting silicate phase was also found to affect the softening and melting-down behavior due to subsequent increase in mobility of iron atoms. Utilizing the topochemical nature of the reduction of iron ore pellet, softening and melting-down properties were found to be improved significantly by means of two-layered structure, where chemical compositions of the core and the shell parts are chosen to have as high solidus temperatures and also as little amount of liquid developed at the solidus temperature as possible at the reduction stage of wustite and metallic iron, respectively.
Mixing characteristics in a molten metal bath are analyzed for inert gas injection through a nozzle at the center of the vessel bottom. It is postulated that the bath consists of two zones: bubble plume zone where gas-liquid mixtures flow upward and annular zone where liquid flows downward. The analysis is made by setting up a steady-state energy balance for the liquid phase. The liquid velocity in the plume zone, the liquid circulating flow rate and the mixing time are calculated under various injecting conditions and correlated as simple functions of gas-flow rate, liquid depth and cross sections of both the plume zone and the vessel. It is found that the cross section of the plume zone has a significant influence on the circulating flow. Large cross section of the plume zone is desirable for mixing in the bath. The calculated results of circulating flow rate and mixing time agree with experimental results obtained previously.
Reoxidation behavior of molten steel during the several pouring processes is studied referring to the previous results of model experiments. It is found that gas volume entrained during teeming of molten steel increases with increasing length of stream and velocity at nozzle exit, and with decreasing nozzle diameter. It is, however, independent of physical properties of molten steel such as surface tension, viscosity and density, except under the condition of high teeming height and small nozzle diameter. As for the amount of oxygen absorption through surface of teeming stream, it is approximately 1/10 of that absorbed by the air entrainment at plunge region in metal pool, if the stream is practically laminar. However, difference between them decreases with decreasing teeming rate and teeming height. A nomograph for estimation of oxygen absorption by gas entrainment and also through surface of stream was proposed. Oxygen absorbed during teeming from ladle to tundish at continuous casting of molten steel decreases with increasing casting rate. Under a constant casting rate, it decreases with increasing casting time accompanying increase of stream radius and decrease of stream exit velocity. It is also suggested that all of oxygen in entrained gas bubbles dissolves into liquid steel.
Dissolved oxygen contents of Fe-Si-O alloy (0.1% Si addition) were determined at 1555°C during repeated cooling from 1600°C to 1555°C. The difference between the dissolved oxygen contents at 1600°C and 1555°C was significantly greater than the oxygen content corresponding to the larger sized silica particles formed during such cooling processes. This means that, besides the larger sized particles, a large number of very fine silica particles are present in the melt. It was thus proposed that very fine silica, presumably of several tens angstrom in size, precipitated easily in the melt with decreasing temperature and silica particles were formed by their successive adhesion to the possible nuclei suspended in the melt. From a striking resemblance of morphology between the double-leaf type inclusion and the precursor of spherulite, the morphology of silica particles was explained on the basis of the regular branching mechanism known well as the formation mechanism of spherulite.
Blow end control technique for carbon content and temperature in BOF has been remarkably improved by sublance. However it is necessary to develop technique for slag formation control for the purpose of obtaining over 90% simultaneous hitting ratio for carbon content and temperature and controlling of phosphorous content. In Mizushima Works, for continuous detection of slag formation in BOF, various experiments such as measurement of furnace vibration, calculation with waste gas analysis, and measurement of sound were performed. Thus, the best reliability in the continuous monitoring of slag foaming height was found by measuring lance vibration during blowing. This technique is based on the principle that degree of lance vibration is correlated with lance length immerzed into foaming slag and oxygen flow rate. When slag foaming height is continuously detected, slag formation control can be realized by the correction of lance height or oxygen flow rate by computer. Combining this slag formation control technique with program blowing and blow end control ones, fully automatic blowing system was established. The system improved the hitting ratio and yield of steel, reduced reblowing ratio and lengthened furnace life.
In order to produce large ingots of sound inner quality, important factors which influenced the loose structure formation were investigated by employing the ultrasonic testing method. The loose structure was found to differ both in its formation mechanism and solidifying condition depending on the location under investigation. The loose structure which appeared in the axial region varied depending on the ingot shape. The degree of looseness could be evaluated by the shrinkage pipe index, α ; the looseness decreased as the index α increased. The index is defined as a function of three parameters on ingot shape: α=ψ-35.3 [(L/D)/P]+7.7 where, ψ: taper of wide face (%), L/D: height to thickness ratio (-), P: hot top ratio (%). On the other hand, the loose structure which appeared in areas other than the axial region varied depending on the hydrogen content in molten steel. It was found to be essential to reduce the hydrogen content in order to decrease the degree of looseness. Its reduction was remarkably effective for the structure especially in the region where inverse V-segregation appeared.
Phosphatability and surface characteristics of silicon-manganese dual-phase steels have been investigated using bright annealed steel sheets of various contents of silicon and manganese. Silicon remarkably reduces the phosphatability because adsorption of titanium colloid, acting as a nucleating agent (chemical activator) for the phosphate crystal formation, is reduced with an increase in the silicon content. The reduction in the titanium adsorption is due to surface carbon enrichment, which is facilitated by the silicon in steel, during annealing. On the other hand, manganese increases the titanium adsorption because of prevention of the carbon enrichment, resulting in a good phosphatability. Therefore, the phosphatability of dual-phase steel is primarily controlled by a balance of silicon and manganese contents.
Dynamic recrystallization mechanisms have been studied in a fully austenitic Fe-33% Ni alloy, using the orientation analysis of individual recrystallized grains. The dynamic recrystallization has been found to start in the form of bulging of existing boundaries, similarly to the static recrystallization at low deformation strains, and multiple annealing twins have been formed at the growth front. Appart from the twin boundaries, twinning has resulted in the formation of new high angle boundaries, which controlled the growth of recrystallized grains. Evidence has been found suggesting that twins are formed by chance during the boundary migration, and the twinning probability is nearly constant except for some large grains having preferred orientations, where twinning is suppressed. Schmid factor analysis has been unsuccessful to prove possible mechanical effects of twinning based on the applied compressive stress.
The effect of aged structure on the susceptibility to hydrogen embrittlement (SHE) of a 13Ni-15Co-10Mo 2750 MPa grade maraging steel has been investigated for the specimens thermomechanically processed and then aged at various temperatures. In particular, the objective of this study is to elucidate whether the stage of aged structure is or is not severely susceptible to hydrogen embrittlement, when aged at 500°C to produce the maximum strength. The SHE was evaluated using the strain rate dependence of reduction of area in tensile tests in air. The SHE increased markedly when aged at temperatures below 475°C. The aged structure to produce the maximum strength at 500°C showed obviously the same of aged structure with the lower SHE. The higher SHE of this steel compared with the lower strength maraging steels, therefore, could not be attributable to the stage of the aged structure. The increase of SHE by aging at relatively low temperatures was discussed in terms of the both changes in slip mode and solubility and diffusivity of hydrogen due to the presence of the fine and coherent precipitates.
The effect of grain size on the high temperature low-cycle fatigue behavior and other material strength properties of Inconel 617 was studied at 1273 K in air. The strain controlled low-cycle fatigue tests were conducted with a symmetrical (FF type) and an asymmetrical (SF type) strain wave forms. The latter wave form was used for the evaluation of creep-fatigue interaction. The main results obtained in this study are as follows: 1) The tensile strength slightly increased with the increase of the grain diameter. On the other hand, the tensile ductility remarkabley decreased with the increase of the grain diameter. 2) The creep rupture life remarkabley increased with the increase of the grain diameter, especially at the lower stress levels. The effect of grain size on creep ductility has not detailed. 3) The low-cycle fatigue life remarkably decreased with the increase of the grain diameter, especially at the lower strain ranges. 4) The creep-fatigue life was less sensitive to the grain diameter than the fatigue life, because the grain size effects on creep and on fatigue were contrary. It is seemed that the creep-fatigue life is determined by the proportion of the creep and fatigue contribution. 5) The fatigue and creep-fatigue test results have good relations with the tensile and creep ductilities at the test temperature.
In order to investigate the effect of impure helium environments of HTGR (high temperature gas-cooled reactor) on the creep rupture properties of heat resistant alloys for intermediate heat exchanger tubes, a creep rupture test has been carried out on four nickel-base superalloys in two impure helium gases, He-2 and PNP-He, at 1000°C. All the alloys were decarburized in some extent in PNP-He, while some of them were carburized in He-2 during the creep rupture testing, indicating that carbon potential of the gases was lower in PNP-He than in He-2 at 1000°C. The alloy hardened by γ' (Ni3(Al, Ti)) precipitation, R4286, was suffered from the environmental effects resulting in the formations of γ' depleted zone, Al internal oxides and grain boundary carbide depleted zone beneath the alloy surface. The creep rupture life of this alloy in PNP-He was found to be one-third of that in He-2, which was mainly caused by the formation of carbide depleted zone along grain boundaries due to decarburization in PNP-He. The alloy hardened by α-W precipitation, KSN and 113MA, was suffered from the environmental effects resulting in the formations of α-W and grain boundary carbide depleted zones. The creep rupture life of KSN containing highest concentration of tungsten, 26%W, were found to be almost the same in the two environments, although this alloy was remarkably decarburized in PNP-He. Because KSN was mainly hardened by very fine precipitates of α-W, it was considered that the creep rupture life was little affected by decarburization. The alloy SZ was hardly decarburized in PNP-He, showing a good resistance to corrosion.
Long term creep rupture properties and microstructure of 12% Cr heat resisting steels have been investigated. It has been found that solid solution hardening by addition of Mo is very effective on long term creep rupture strength, but precipitation or dispersion hardening by individual addition of V or Nb is less effective. In the case of Nb, the presence of excessive amounts of undissolved NbC leads to coarsening of MX precipitates, and a marked drop in long term creep rupture strength. The combined addition of V+Nb with Mo and B leads to the superior 30000 h creep rupture strength. The parabolalike decrease in creep rupture curves at 650°C is well concerned with the microstructural changes, in particular, subgrain growth. Microstructural differences between long term crept specimens at 600°C and short term crept specimens at 650°C were discussed by considering the parameter, T(C+logt). Long term specimens at 600°C exhibit higher dislocation density substructure with intensive precipitation reactions which include both fresh precipitation in laths and agglomeration at boundaries, and exhibit intergranular fracture probably caused by stress concentration around the agglomerates at boundaries, whereas short term specimens at 650°C exhibit transgranular fracture. It is considered that the deviation of long term creep rupture time from extrapolated values is caused by differences in precipitation hardening level and fracture mode.
Effects of Si content in steel and the condition of partially- or non-oxidizing treatment followed by reduction treatment on dynamic wetting characteristics of Si-containing steels immersed in a molten zinc bath were studied by means of a partially- or non-oxidizing furnace simulator and a measureing instrument of gas reduction type meniscograph. 1) Steels containing 0.83% Si exhibited a good wettability when they were oxidized partially at an air/fuel ratio of 1.05 or more, which formed an oxide film of visible thickness and then reduced for 30 to 40s at 700°C. 2) The zinc wettability of Si-containing steels well corresponded to the degree of formation of Fe containing oxygen, namely Fe(O) and also the degree of Si enrichment at the steel surface layer which was composed of (FeO)2 SiO2. 3) The surface topography of steels changed in accordance with the air/fuel ratio during the partially-oxidizing treatment, Si content in steels and reduction treatment conditions. The rougher and more porous the surface morphology, the greater the initial wetting rate of the steels.
The effect of C content of less than 0.04% and Nb addition on the mechanical properties of base metal and the toughness of weld bond was investigated on Al-killed steels for low temperature service. The main results obtained are as follows: (1) The toughness of base metal was the highest when C content was between 0.01 and 0.02%. (2) The strength of Nb steel was not lowered significantly by decreasing the C content when it was less than 0.02%. This was attributed to the increase in hardenability due to solute Nb and the precipitation of Nb carbonitrides. The strength of Nb free steel, on the other hand, decreased with the decrease in C content. (3) The toughness of weld bond of Nb steel was the highest when C content was between 0.01 and 0.02%. In the case of Nb free steel, on the other hand, the decrease in C content below 0.04% did not improve the toughness of weld bond. This was due to the fact that the grain grew as C content decreased in the case of Nb free steel, while the grain growth was restrained by solute Nb in the case of Nb steel. (4) From these facts, C content between 0.01 and 0.02% and Nb addition were concluded to be the most appropriate for Al-killed steels for low temperature service.
Thermal analysis of Fe-Mo alloys was carried out by the use of Knudsen cell mass spectrometry. The composition range of the alloys and the temperature range studied were 25 to 67 at %Mo and 1300 to 1650°C, respectively. Temperatures for invariant transformation reactions obtained are as follows: α=L+δ; 1454±2°C δ=L+σ; 15031521°C σ=L+(Mo); 1610±1°C The temperatures for the decomposition of α and δ phases agree well with the previously reported values. As to the temperature of the decomposition of σ phase, present result is about 70°C higher than the previous value. On cooling of samples, linear relation was observed between the degree of super cooling and the temperature at which the transformation took place. In this technique, the ion intensity is used as the measure of the temperature of sample. The ion intensity is related to the effusate in equilibrium with the surface of sample. Further, the ion intensity is roughly exponential function of the temperature. Thus, this technique is featured with noncontactive, very low time constant and highly sensitive measurement of the temperature of sample.
The emission intensity of background spectrum occupies a large part of measured intensity, in determining micro amounts of elements in iron and steel by inductively coupled plasma emission spectrometry. Therefore, it is necessary to eliminate the background part in order to obtain the correct intensity of an-alytical spectrum. Two methods have been proposed to eliminate the background: 1) The sample is analyzed with calibration solutions containing same amounts of matrix elements as in the sample solution. 2) The intensity of background spectrum is measured with the neighbouring line and subtracted from that of analytical line. The former is valid for the analysis of samples with known chemical composition, the latter is applicable to all unknown samples. The difficulty, however, still remains for it to find the neighbouring line with which emission spectra of iron and other elements don't overlap. Both methods have been applied to the determination of trace amounts of cobalt and zinc in a steel. The background has been fully corrected by both methods, and the results obtained are well agreed with those obtained by conventional method of chemical analysis.
Optical emission spectrometry is superior to other analytical methods in its rapidity. Therefore, it plays important role in the field of elemental analysis of the steel making plant. However, further speedup and labor saving of the analysis by automation are required for progress of continuous steel making process. An apparatus has been developed to meet this, and the following results were obtained. (1) Analytical procedure was fully automated, from the reception of samples being analyzed at the laboratory to the transmission of analytical results. The time required for the analysis of a sample was 50 seconds. (2) Automatic standardization and automatic selection of working curves were also provided. (3) Better precision was obtained than conventional integration method in the analysis of routine samples from a converter by employing the pulse height distribution analysis method.