Experiments were conducted to extract neodymium in bismuth by melting Fe-Nd-B magnets and bismuth together in a graphite crucible at 1200°C. Molten iron and bismuth separated into two phases, and neodymium in the magnet dissolved in the bismuth phase. Neodymium dissolved in bismuth was considered to form BiNd as a result of XRD. The neodymium concentration in iron phase after the dissolution treatment was 0.05 at.% or less,and the recovery rate was 99% and more. Compared to other metals(Mg, Ag, Cu) that separate into two phases with respect to molten iron, bismuth can be said to be a metal extraction material with a low melting point, safety and low energy cost.

Ceramic roller can be used in the heating furnace conveniently because of its high temperature resistance. The roller consists of ceramic sleeve and steel shaft connected only under a small shrink fitting ratio because of the brittleness. However, the coming out of the shaft sometimes happens from the ceramic sleeve under repeated load. Therefore, it is important to find out the driving out force generation mechanism to prevent the coming out failure. Based on the previous studies, a two-dimensional shrink fitted structure is considered by replacing the shaft with the inner plate and by replacing the sleeve with the outer plate. Then, this research focuses on calculating the driving out force generated on the inner plate by introducing a newly designed stopper on the outer plate. The finite element simulation shows that the coming out phenomenon can be prevented due to the contact between the inner plate and the stopper installed on the outer plate. Then the mechanism of driving out force generation is clarified. Finally, the process of coming out in terms of the residual displacement is illuminated.

Observation of chemical state of solid-solution carbon in a low-carbon steel was tried by C-K near-edge x-ray absorption fine structure spectra measurement. In addition, the wavelength dependence of the photoelectron spectrum on the surface of the bulk steel was evaluated, and the contamination and oxidation layer of 3 nm in thickness on the surface of the steel was found. As a result, it was possible to observe the chemical state change of carbon existing in the bulk iron located deeper than the oxidation and contamination layer, by evaluating the difference spectra between the sample and the reference. Furthermore, by evaluating the shape change of the difference spectra depending on the heat treatment time, this study suggested that the chemical state of carbon in bulk iron changes with heat treatment.

Copper sulfide (Cu_xS) has been frequently observed in steel samples, prepared using selective potentiostatic etching by electrolytic dissolution (SPEED). It is often the case that Cu_xS is detected unexpectedly from the precipitates extracted from steel samples by selective potentiostatic etching, although such Cu_xS formation during the heat treatment conducted is not anticipated by the thermodynamic equilibrium calculations.

In this study, we observed such artificial Cu_xS along with manganese sulfide (MnS) precipitates, which were extracted from steel materials by SPEED, using secondary electron microscope (SEM) equipped with energy dispersive X-ray spectroscopy (EDX) and Auger electron spectroscopy. These Cu_xS-MnS sulfide complex would be formed by the following mechanism: as the solubility of Cu_xS is far bigger (i.e. 10 times or more) than that of MnS, Cu²⁺ ion dissolved from steel matrix would be exchanged with Mn²⁺ ion on the MnS surface during the etching process, leading to a formation of Cu_xS-MnS sulfide complex.

In order to suppress the formation of such Cu_xS, we propose the use of following electrolyte: a non-aqaueous solutions of 4% methyl salicylate + 1% salicylic acid + 1% tetramethylammonium chloride (TMAC) + 5% Triethylenetetramine (TET) in volume fraction, in methyl alcohol (Cu ion selective hold etching by electrolytic dissolution, abridged as CUSH electrolyte). Then, this electrolyte was applied to precipitates in steel samples. It was effective to prevent the formation of sulfides in electrolyte, with the effect of metallic (Cu²⁺, Ag⁺, Pb⁺, etc.) chelating ability of TET.

In hot rolling of ferritic stainless steel, prevention of sticking is very important from the viewpoint of productivity. However, the formation mechanism of sticking has not been clarified sufficiently. Therefore, in this work, rolling experiments were carried out using a tribo-simulator. The results clarified the following points: Sticking occurs more easily on ferritic stainless steel than on high strength steel. On ferritic stainless steel, the work roll sticks with the hot-rolled sheet at the entrance of roll-bite, and the work roll then moves forward on the hot-rolled sheet. Therefore, it is thought that the surface layer of the hot-rolled sheet is fractured by large shear strain, and the work roll stuck with the fractured layer advances further on the sheet, forming a defect with an accumulated fractured surface layer on the hot-rolled sheet. Applying a lubricant oil is effective for prevention of sticking between the work roll and hot-rolled sheet.

The Fe-Zn alloying reaction and the selective oxidation behavior of 0.7 mass% Si - 1.15 mass% Mn added hot-rolled steel annealed at 600-800°C were investigated by comparison with those of cold-rolled steel. The Fe-Zn reactivity of the hot-rolled steel improved from 600°C to 700°C, while, it deteriorated from 700°C to 800°C. Above 700°C, the amount of Fe-Si-Mn oxide on the steel surface increased with increasing temperature and deteriorated Fe-Zn reactivity. Below 700°C, a thin layer of Fe oxide which existed on the steel surface deteriorated Fe-Zn reactivity. This thin oxide layer was reduced by Si and Mn diffused from the steel substrate. As the temperature increased from 600°C to 700°C, Fe-Zn reactivity improved due to the formation of reduced iron on the steel surface. In the case of the cold-rolled steel, the same selective oxidation behavior and reduction mechanism of the Fe oxide were also confirmed. As a result, the Fe-Zn reactivity of the cold-rolled steel showed the same behavior as that of the hot-rolled steel. However, the Fe-Zn reactivity of the cold-rolled steel improved at a lower temperature than that of the hot-rolled steel. This can be explained by the faster diffusion rates of Si and Mn in the cold-rolled steel than in the hot-rolled steel. At the surface of the cold-rolled steel, reduction of the Fe oxide layer was promoted and the Fe-Zn reactivity of the cold-rolled steel was improved at a lower temperature.

The effects of hydrogen charge on the formation of vacancy-type defects in low alloy martensitic steel were investigated by using positron lifetime spectroscopy. During slow strain rate tensile tests (SSRT), dislocation and vacancy were introduced and increased with increasing tensile strains. Vacancy clustering was significantly promoted by hydrogen charge, although there was no remarkable difference in the dislocation density and mono-vacancy equivalent vacancy density between hydrogen-charged and uncharged samples under the same tensile strains. Increase in hydrogen concentration in the steel promoted vacancy clustering. Preferential conditions for the vacancy clustering showed a good agreement with the conditions for significant elongation loss by SSRT, implying the vacancy clustering has an big effect on hydrogen embrittlement. In the case of constant load tests under applied stress in elastic region, no increase in vacancy and dislocation density was observed. This implies plastic deformation is necessary for the formation of vacancy-type defects even under hydrogen charging.

The effect of prior cold deformation on the dissolution speed of carbide in SUJ2 bearing steel subjected to induction heating austenitization at 920°C was examined. Microscopic observations using scanning electron microscopy and transmission electron microscopy, in addition to crystallinity measurements using powder X-ray diffraction, found no trace of the cold deformation effect in the carbide and few crystal defects in the spheroidal carbide particles of an as-received specimen. However, it is important to note that the cold deformation resulted in a refinement of austenite grains during austenitization, which accelerated the dissolution of chromium from the carbide compared to an unprocessed specimen. This led to enhancement of the resultant carbon diffusion in a matrix.

The volume fraction of carbide present at grain boundaries was estimated assuming a simple geometrical model. The relationship between the holding time and the volume fraction of undissolved carbide was derived based on a bulk diffusion model of chromium into an austenite matrix. Experimental results on the dissolution speed of carbide were qualitatively explained by considering the combination of grain boundary diffusion and bulk diffusion. The austenitization behavior of SUJ2 steel is remarkably influenced by the microstructure, in particular for short austenitization times.

Strengthening effect by microalloyed vanadium (V) on eutectoid pearlite steel has been investigated from the perspective of nano-precipitation and lattice strain. 0.2% proof stress of specimens, isothermally transformed at 873 K, increases around 160-170 MPa with 0.1% V addition. However, interphase precipitation of vanadium carbide (VC), regarded as the principal strengthening factor, has not been detected by transmission electron microscopy or 3D atom probe microscopy (3D-AP). On the other hand, lattice strain in lamellar ferrite, analyzed by broadening of X-ray diffraction peak, has clear correlation with proof stress. The lattice strain data of 0.1% V added pearlite specimens are plotted on the same correlation line as of V free ones. In addition, elemental map by 3D-AP shows that vanadium atoms concentrate in lamellar cementite rather than ferrite, which could change cementite lattice parameters and gain ferrite/cementite misfit causing lattice strain increment. These results reveal that microalloyed V influences not only precipitation of VC in lamellar ferrite, but also the lattice strain increment in pearlite lamellar. As far as pearlite steels containing at most 0.1% V, lattice strain is considered to be the major factor of their yield behaviors. Furthermore, 0.1% V addition has not enhanced work-hardening behavior as notably as estimated by Ashby’s work-hardening theory of dispersion-hardened crystals. Therefore, VC precipitation should not necessary for V strengthening effect on pearlite steel.

A low carbon ferritic steel (Fe-0.0056%C) was used for the estimation of dislocation density. Annealed steel sheet with the ferrite grain size of 50 μm was subjected to cold rolling up to 80% thickness reduction and then provided to the X-ray diffraction analysis. Dislocation density was evaluated by the Modified Williamson-Hall / Warren-Averbach method. It was confirmed in cold rolled specimens that the equation; σ_y [GPa] = 0.05+18 ρ /10⁹ stands up in the relation between yield stress σ_y and dislocation density ρ [m^–2]. On the other hand, the values of parameter φ in the Modified Williamson-Hall equation were accurately estimated in each cold rolled specimen and it was found that the equation ρ[m^–2] = 4.5×10¹⁹φ² stands up between φ and ρ. In addition, it is suggested that the screw dislocation component decreases with cold rolling and, in specimens with cold rolling more than 40%, dislocations are of almost full edge dislocation.

Fatigue tests were performed to obtain co-planar and branch crack growth rates on rail and wheel steel under non-proportional mixed mode I/III cycles. In the experiments, sequential and overlapping mode I and mode III cycles that simulated the load experienced by the rolling contact fatigue cracks were applied to the crack in cylindrical specimens made of rail and wheel steel. Experiments showed that a long co-planar crack could be produced under certain loading conditions. Based on the fracture surface observations by SEM and the results of FEA, the long co-planar crack growth is thought to be driven mainly by mode III loading and the role of mode I is an assistant, keeping the crack face opened. It was observed that the cracks were apt to branch when the strength of the material increased. It was also observed that the crack branched when the degree of overlap between the mode I and mode III cycles increased. We proposed the equivalent stress intensity factor range for branch crack that can consider the crack face contact and successfully regressed the crack growth rate data. Comparing the fracture surfaces and the co-planar crack growth rates data under non-proportional mixed mode I/III loading with that under I/II loading, it is found that the mechanism of shear mode crack growth is essentially the same regardless of whether the main driving force is in-plane shear or out-of-plane shear.

The applicability of the finite element method coupled with electromagnetic field analysis and heat conduction analysis to optimization of induction heating of a bearing raceway was assessed. In order to obtain temperature uniformity in a heated raceway, precise design of the coil arrangement and heating schedule are required. It is important to know how the internal temperature changes with time for the development stage.

The one-dimensional skin current model predicts that the temperature difference between the outer and inner surfaces decreases and that the temperature becomes uniform as the heating region moves inward when the outer surface temperature nears the Curie point. This phenomenon was reproduced in simulations and experiments for a ring sample, but it depended on the balance between the thickness of the ring and the penetration depth of the electromagnetic field. It is necessary to set an appropriate rate of temperature increase and frequency for the power source according to the thickness of ring.

High accuracy analysis was possible by using temperature-dependent B-H characteristics when a heated ring was in the ferromagnetic state. However, in the paramagnetic state, the agreement between simulation and experiment became worse. This may be due to the high rate of temperature increase used in this study, which suggests that the shift in the transformation point to high temperature must be taken into account.

A continuous blast furnace slag solidification process was developed to promote the use of air-cooled slag coarse aggregate for concrete. In this process, molten slag can be solidified in only 120 seconds, and the thickness of the slag is about 25 mm. After crushing the slag, the water absorption ratio is much lower than that achieved in the past because gas generation is suppressed. With this apparatus, most of the slag is crystalline, but part of the slag has a glassy surface. Therefore, EPMA and XRD were used to study the glass transition phenomenon. It found that the thickness of the glass layer is about 2 mm. To discuss the glass transition and crystallization phenomena, the thermal history was simulated by heat transfer analysis. The results clarified the fact that all the slag on the mold has a glassy surface layer of about 2 mm, and good agreement between the calculation and experimental data concerning the layer was obtained. It was also shown that most of the slag crystallizes in the slag pit because the temperature inside the piled slags rises to more than 1173 K. The measured slag temperature and calculated temperature were also in good agreement.

In Modified Williamson-Hall / Warren-Averbach (MWH/WA) method, the estimation of parameter α is of vital importance for obtaining other parameters accurately. Almost all researchers use the procedure for determining parameter α proposed by Ungár et al. (route 1), which includes the rough approximation. On the other hand, Takebayashi et al. suggested an improved procedure (route 2) which could give more precise results. By comparing above two routes, it is found that route 1 can result in overestimated (about twice) α values and significantly impacts further analysis in MWH/WA method due to ignored terms. Since the difference of α value affects the crystallite size D and parameter φ, the application of route 2 is recommended in MWH/WA method.