The importance of energy saving in the ironmaking process is widely recognized. Many energy saving efforts related to ironmaking have already been carried out, and further energy savings by conventional methods are hardly to be expected. The oxygen blast furnace is considered to be a promising process in terms of flexibility of energy use and advantages related to CO2 mitigation. Focusing on energy saving, in this study, the optimum configuration of the ironmaking process based on the oxygen blast furnace was investigated by numerical approaches and case studies.
First, because productivity can be greatly improved in the oxygen blast furnace, blast furnace inner volume can be reduced while maintaining the same production rate. Because the downsized oxygen blast furnace makes it possible to relax burden material strength requirements, energy consumption for agglomeration in the coke oven and sintering machine can also be reduced. Therefore, a DEM simulation was carried out to confirm the effect of the burden load reduction in the downsized condition. It was found that the compressive stress in the downsized oxygen blast furnace was 20-30% less than that in the conventional blast furnace. The energy flow in the ironmaking process was also investigated by using a material and energy balance model, considering the functions of an integrated steel works. It was found that the energy consumption of the ironmaking process based on the energy saving oxygen blast furnace could be reduced by 5.3% while maintaining the same energy supply to downstream processes.
The shaft furnace, which is a scrap melting furnace, plays an important part as an energy supplier in a steelworks, because the shaft furnace produces high calorie gas. The shaft furnace is required either to reduce coke rate or to increase exhaust gas calorie, corresponding to energy balance in the steelworks. Coke rate and exhaust gas calorie are determined by the coke gasification reactivity, so the control technique of the reactivity is very important. In this study, coke surface was coated by CaCO3, Fe2O3 and SiO2 to control the coke reactivity, and the gasification rates were measured at 1573-1773 K. As the results, the gasification rates were accelerated by CaCO3 and Fe2O3, and decelerated by SiO2. The acquired gasification rates applied to a one-dimensional mathematical model of the shaft furnace. The shaft furnace operations controlling the coke reactivity were simulated and the effects of the coke reactivity on coke rate and exhaust gas calorie were estimated.
In order to investigate the peritectic reaction, Ag-18 mass%Sn alloy has been used in this study. Approximately 10 g of this alloy was melted in the alumina crucible. Using a fine K-type thermocouple, thermal history during cooling was recorded. The peritectic reaction was observed at about 725 ºC, which corresponded to the equilibrium phase diagram. The metallographic SEM observation of the specimen, which was quenched at 725 ºC, revealed that the secondary phase, ζ phase, completely covered the primary (Ag) phase. These indicated that ζ phase easily nucleated on the primary (Ag) phase and the peritectic reaction rapidly finished. EBSD analysis showed that plural crystals of ζ phase covered on the single crystal of (Ag) phase. This also indicated that ζ phase formed easily on the (Ag) phase. The analysis of relation between temperature and phase fraction indicated that thick ζ phase rapidly formed along the (Ag) phase / liquid interface and peritectic reaction plays a major part of transformation in this alloy system formed in an isothermal cooling.
The age-hardening behavior at around 750°C and microstructure analysis of experimentally prepared heat resisting steels based on SUH38 containing phosphorus in the range of 0 to 0.3 mass% have been investigated, in order to identify the formation mechanism of the homogeneous distribution of fine carbide within grain interiors (MDP: matrix dot precipitation) by the addition of phosphorus. The steel does not show age-hardening until the phosphorus content in the steel beyond 0.15 mass%, and the hardness increases from 200 HV to 350 HV with increasing phosphorus content. The high age-hardening was caused by the high density of fine M23C6 carbides with cube on cube orientation relationship between γ and M23C6. Phosphorus was found to exist in γ-matrix as substitutional manner after solution and aging treatments. The fine carbide formation mechanisms within grain interiors are attributed to an interaction between phosphorus and carbon in solution.
Plastic deformation and dislocation accumulation in dispersion hardening alloys are numerically analyzed by a crystal plasticity finite element technique and work hardening characteristics are discussed. The critical resolved shear stress for slip system is given by the extended expression of the Bailey-Hirsch type model which include the Orowan stress as size effect of microstructure. Work hardening of slip system is estimated by statistically stored dislocation (SSD) density. Increment of the SSD density is evaluated by slip strain and the mean free path of dislocations (the Kocks-Mecking model). The mean free path depends on the average spacing of dispersed particles, which is also used to estimate the Orowan stress. The average spacing of dispersed particles is calculated from the volume fraction and average diameter of dispersed particles. As a result, flow stress level at the initial stage of deformation agreed very well with experimental result but work hardening rate was higher than that of experiment. From this fact, it is considered that the mean free path and the average spacing of dispersed particles are different spacing factors. When we assume that the mean free path is two to three times larger than the average spacing of dispersed particles, numerical result of the strain hardening agrees very well with experimental one.
The effect of stress-induced martensite on the work hardening behavior of ferrite-austenite duplex stainless steels has been investigated. A precise grid marker method was utilized to observe the plastic strain distribution in each phase during the tensile tests. Two kinds of duplex stainless steels containing 22Cr-5Mn-0.34N and 20Cr-5Mn-0.25N were employed. Stress-induced martensitic transformation did not occur in 22Cr-5Mn-0.34N steel while it occurred in 20Cr-5Mn-0.25N steel. The former uniform elongation is 0.33, but the latter is 0.61, indicating that the transformation-induced plasticity (TRIP) is apparent in the latter steel with metastable austenite due to the low nitrogen content. The precise grid marker method revealed the inhomogeneous plastic deformation in both steels. Particularly, a large difference of plastic strain was observed between the areas of newly formed martensite and the austenite adjacent to the martensite, although the plastic strain due to the martensitic transformation itself was not clearly detected. Vickers hardness tests were performed, and it was clarified not only that austenite phase is harder than ferrite phase but also that stress-induced martensite is harder than residual austenite. Based on these results, the contribution of the stress-induced martensite to the work hardening behavior in the duplex stainless steel was discussed from the view point of load transfer from the soft phase to the hard phase, i.e., the internal stress induced between the two phases.
Predictions of gigacycle fatigue strength in high-strength steel were derived by using previously proposed method and past fatigue test results. The predictions were proposed for 5 grades of high-strength steel mainly underR= –1. SUP7 then had 2 heat treatment conditions and predictions for SCM440 were not only under R= –1 but also under R= 0. Accuracy of the predictions was mostly good, while the predictions for S40C, SUJ2 and SCM440 under R= –1 showed a little bit inferior accuracy to others. Although the accuracy for S40C was the lowest, this was perhaps attributable to large scattering of the fatigue test results caused by poor hardenability. In these analysis, existence of fatigue limits was suggested in case of the internal fracture. The new fatigue limits could probably be confirmed by conducting 1011 cycles fatigue tests in future. Temporary predictions of the fatigue limits were derived in this report. Predicted S-N curves showed large difference among the steel grades in a short life region, while the difference was small in a gigacycle region. Although the predicted gigacycle fatigue strength were reduced according to increase of the inclusion size, the reduction became gentle for large inclusions. Accordingly, terribly low fatigue strengths were not predicted even for huge inclusions. Mean stress effects showed good agreements with modified Goodman’s rule. However, general predictions regardless of the steel grades were difficult to derive in this study, so analogy or additional fatigue tests were necessary to predict the gigacycle fatigue strength of unlisted steels.