Experimental resultt by present author's research group at Tohoku University are reviewed. High purity iron which is is considered to show the inherent mechanical properties of iron was prepared and used to investigate the low temperature brittleness, temperature dependence of the yield stress, solution softening and hardening, and effects of small addition of solutes on these properties. Specimens were wires and tensile tested at various temperatures above 4.2 K.
Iron of 99.999% purity and higher shows transgranular fracture (TGF) below 50 K if the microstructure is unfavorable to the intergranular fracture (IGF), and shows IGF below 120 K if the microstructure is favorable. The ductile-to-brittle transition temperature of IGF strongly depends on the microstructure. Less pure iron, 99.99%, shows IGF at 77 K even though the microstructure is unfavorable to IGF. Oxygen segregated at grain boundaries promotes IGF, but the effect is small. Carbon segregated at grain boundaries prevents IGF by increasing the cohesion at the boundaries. The strong temperature dependence of the yield stress is the inherent nature of iron, although impurities promote the dependence. High purity iron shows a hump on the temperature-yield stress curve at the temperature range between 170 K and room temperature. The hump is suppressed by dissolved carbon, and also absent in impure iron. Hydrogen causes softening at the temperature range of the hump and hardening outside the temperature range. Hardening due to hydrogen often reported previously is an effect in impure iron. The softening and hardening due to carbon is additive to the softening due to hydrogen.

View full abstract

Investigations of channel-type reactors are of importance not only for the development of continuous steelmaking processes but also for the pretreatment of pig iron and posttreatment of crude steel. By means of residence time measurements and light section analysis, a channel-type reactor model specially designed for experiments with two (water-air, water-oil) and three phases (water-oil-air) has been developed. Special emphasis was thereby placed on the prevalence of piston flow during top-blowing with several lances in order to increase the area of reaction by emulsification at the interface between the two liquid phases. In the first experiments the flows of the liquid phases (water, oil) were optimized. Caprylic acid was used to simulate the mass transfer of tramp elements from the pig iron into the slag. The mass transfer was measured through conductivity, transferring the experience of discontinuous experiments to those which were conducted continuously. The different phase contacts: permanent, reverse transitoric (water is flowing continuously along permanent remaining oil layer) and countercurrent, were investigated in a channel-type reactor concerning the effect of mass transfer.

View full abstract

The rate of decarburization of carbon-saturated liquid iron with CO₂-H₂O gas mixtures has been studied to bettter understand the reaction mechanism between liquid iron containing carbon with post-combusted gas in an iron bath smelting reactor. The measurements were done under conditions in which the effect of mass transfer is negligible. The experimental temperature was 1773–K and the sulfur content in the metal was fixed at 0.2 mass%. The rate was measured for a partial pressure of CO₂ in the range of 0.5 to 1 atm for CO₂-Ar, 0.7 to 0.9 atm for CO₂-H₂O and 0 to 0.95 atm for CO₂-5%H₂O-Ar gas mixtures.
The measured rates agreed well with the predicted rates by assuming an ideal additivity of CO₂ and H₂O reaction rates individually measured in previous work. It was concluded that there were no special effects retarding or accelerating the reaction rate with liquid iron and the rates were simply additive. It was also found that iron oxide did not form on the surface as long as the surface concentration of carbon did not become very low. The critical carbon content at which FeO forms on the surface was expressed as a function of partial pressure of oxidant, sulfur content in the metal, mass transfer coefficient and temperature and was found to be in agreement with experimental observations.

View full abstract

One dimensional heat transfer analysis during centrifugal casting of aluminum-silicon base composite containing graphite particles has been studied. For this purpose, a finite difference technique has been adopted. The results indicate that the thicknesss of the region in which dispersed graphite particles are segregated due to centrifugal force, is strongly influenced by speed of rotation of the mold.
The thickness of graphite rich region near the inner periphery decreases with increase in speed thereby increasing the volume fraction of dispersion. The solidification time of the casting is also dependent upon the speed of rotation of the mold, and it decreases with increase in speed. This study also indicates that the presence of particles increase the solidification time of the casting.

View full abstract

The flow pattern of molten steel in a slab continuous caster mold is very important for removing the nonmetallic inclusions in the molten steel and preventing the entry of mold flux/slag into the molten steel. Different flow patterns are created by different operating conditions and submerged nozzle configurations. In this study, a three dimensional analysis system for turbulent fluid flow problems with free top surface has been developed to numerically simulate the molten steel flow in the mold. It can evaluate the effects of operating condition and nozzle design on the flow pattern and thus on the flotation of inclusions and flctuation of melt surface layer. The results of the simulations are very useful in determining the optimum operating condition and submerged nozzzle design to obtain cleaner continuous by cast slabs.

View full abstract

The first-order (e_O^Ca) and the second-order (r_O^Ca, r_O^{Ca, O}) interaction parameters between calcium and oxygen in liquid iron were determined at 1873 K by using the previous data obtained in slag-metal equilibrium experiments, on the basis of the reciprocal and conversion relationships between mass percent and mole fraction scales. The Values for e_O^Ca, r_O^Ca and r_O^{Ca, O} were obtained as -3600, 5.7×10⁵ and 2.9×10⁶ in the range of {[mass%Ca]+2.51[mass%O]}<0.005 and -990, 4.2×10⁴ and 2.1×10⁶ in the range of {[mass%Ca]+2.51[mass%O]}≥0.005, respectively, and the validity of these values was discussed.

View full abstract

The design of thermomechanical processing schedules to control microstructures requires the knowledge of the austenite-to-ferrite transformation start temperature (Ar₃). In this industrial process, during deformation, the temperature usually decreases continuously. Thus, two new methods to determine the Ar₃, based on continuous cooling compression (CCC) and continuous cooling torsion (CCT), have been developed. While the former is applicable for low strains only, the latter can be used for low and high strain processes.
The aim of this investigation was to determine the effect of deformation in the single phase austenite and two phase austenite plus ferrite region on the transformation and dynamic transformation behaviour of austenite-to-ferrite. CCC tests were carried out on a low carbon steel and the influence of strain was examined.
As expected, deformation in the single phase austenite region increased the kinetics of the austenite-to-ferrite transformation, raising the Ar₃. The faster kinetics leads to a finer polygonal ferrite grain size after transformation. Straining in the two phase region causes strain concentration on the softer ferrite and, consequently, recrystallization of this phase. Deforming close to the Ar₃ maximizes the strain effect on dynamically transformed ferrite.

View full abstract

All-weld blocks of C-Mn steel were prepared by multipass submerged arc weld deposit, using different welding currents (500-750 A) and speeds (40-60 cm/min), filler wire of 4 mm diameter and a basic bonded type flux. The dendrite content of the weld blocks was estimated by quantitative metallography technique. The fatigue crack growth rate of the weld at LT direction of the block was evaluated by using C-T specimens and correlated with the area fraction of dendrite (D_f) of the matrix. The increase in dendrite content of the matrix was found to reduce the crack growth rate by influencing significantly the material constant C and exponenet m of Paris law, da/dN=c(ΔK)^m, where, 'a' is the crack length, 'N' is the number of loading cycle and ΔK is the stress intensity factor range. The modification of C and m with D_f, was found to provide a satisfactory basis for the evaluation of fatigue crack growth rate of multipass C-Mn steel weld deposit having a given amount of dendrite.

View full abstract

Plane bending fatigue tests were conducted on smooth and notched specimens made with combinations of tensile strength and thickness of 440 N/mm² and 2.6 mm, 590 N/mm² and 2.3 mm, and 780 N/mm² and 2.0 mm, respectively. A change in the fatigue limit load (moment) was investigated. In smooth specimens, the fatigue limit moment did not change with the variation of steel. In the notched specimen having an 8 mm^φ pierced hole, however, the fatigue limit moment decreased drastically with the increase in tensile strength in combination with the reduction of thickness. An improvement in the fatigue limit stress of the notched specimen was then studied from the point of view of the stamping process. It was found that coining the pierced hole is one effective method.

View full abstract

The detrimental effect of Ni addition on sulfide stress cracking (SSC) resistance has been well known since the beginning of the study on SSC. However, effects of Ni itself and untempered martensite which can be typically formed for Ni bearing steel have not been clearly established. In this paper, the effects of Ni addition on SSC resistance are investigated using constant load tests and DCB tests for low alloy tempered martensite steels. SSC threshold stress (σ_th) decreases with an increase in Ni content at a constant yield strength. This is probably caused by the fact that fissures work as initiation sites of SSC. On the other hand, an addition of Ni raises slightly the value of K_ISSC. Each K_ISSC value is supposed to be determined by the hydrogen content. Hydrogen content when crack propagation stops decreases with an increase in Ni content, while hydrogen content increases in Ni content then the steel starts being exposed to a sour environment. Untempered martensite is cormed during tempering at a temperature just below A_C1. Steels with untempered martensite show lower values of σ_th and K_ISSC than those without untempered martensite at a constant yield strength. So an addition of Ni promotes the formation of fissures and accelerates SSC initiation even without untempered martensite. Furthermore, untempered martensite decreases even more SSC resistance.

View full abstract