The mechanism of the texture formation in low carbon steels is not fully understood as yed. To clarify this, a series of systematic investigations has been made. Summarizing these results and comparing these with those of other investigations, possible mechanisms of texture formation are discussed. In polycrystalline iron, deformation and recrystallization in grain boundary regions play the most important role in the texture formation. In this respect, results obtained on single and bi-crystals do not provide much informations. In polycrystals, crystal rotation occurs during rolling along different paths from that observed in single crystals, affected by grain boundary constraints significantly. Approaches to the stable end orientation seems to be closely related with the development of the stable dislocation substructures. The development of the rolling texture is therefore strongly affected by the metallurgical factors that influence the evolution of the dislocation substructures. {111} recrystallized grains seem to be formed from the grain boundary regions of the same orientation through the subgrain growth mechanism.

View full abstract

A mathematical model for estimating liquid and bubble rising velocities in metallurgical reactors subject to bottom gas injection was proposed. The mass conservation of gas and the momentum conservations of gas and liquid were employed as governing equations. Radial profiles of gas holdup and liquid velocity were assumed to be Gaussian distribution. Axial distributions of the center line value and the half-value radius of gas holdup were given by empirical correlations. The ratio of the half-value radius of gas holdup to that of liquid velocity was also given by empirical correlation. The slip flow model and the drift flux model were employed for estimating bubble rising velocity. Concerning the liquid velocity, the results of this study agreed well with experimental results published so far. For the bubble rising velocity, in the region near the nozzle where gas holdup on the center line, α_c/l, is higher than about 10%, it is hard to define which model is more appropriate, but in the region far from the nozzle where α_c/l is lower than about 10% as a result of very highly turbulent mixing, the slip flow model would be better.

View full abstract

A bubbling jet in a cylindrical bath with centric bottom gas injection rotated around the vessel axis under certain blowing conditions. There existed two kinds of swirl motions: one was closely related to so-called rotary sloshing and the other was induced by hydrodynamic instability of a large scale ring vortex enclosing the bubbling jet.
Conditions describing the occurrence and cessation of the swirl motions have been studied by many researchers. Meanwhile, the bubble characteristics, liquid flow characterstics, mass transfer from a solid body immersed in the bath, and mixing time of the bath have not been clarified in the presence of these swirl motions. This study was made to investigate the effects of the former swirl motion on the above-mentioned transport phenomena using a high-speed video camera, an electro-resistivity probe, a laser Doppler velocimeter, an electrochemical sensor, and an electric conductivity probe. In order to compare the results with those obtained in the absence of the swirl motion, a cylindrical pipe was used to stop the swirl motion. The swirl motion was found to enhance the mass transfer coefficient and reduce the mixing time significantly. This fact allows us to develop new metallurgical processes using swirl motions. It should be stressed, however, that the errosion of vessel walls also is enhanced by swirl motions.

View full abstract

In the transverse sections of low carbon continuously cast steel billets, samples for analyses of carbon and sulphur were collected by drilling at the billet centres as well as at the columnar-equiaxed transition (CET) boundaries, as revealed in etched macrostructures. Correlation of degrees of segregation for carbon (r_c) with those for sulphur (r_s) agreed closely with that predicted by Scheil's or modified Scheil's equation at the CET boundaries, but not at the billet centres. Variations of r_c and r_s with fractional solidification, although did not agree with predictions of the above equations, were in qualitative agreement with actual macrosegragation data reported in literature. Attempts were also made to correlated the experimental macrostructural and macrosegragation data with predictions based on conjugate fluid flow-heat transfer model developed earlier. Results of this exercise demonstrated close agreement of the predicted locations of CET boundaries with the measured values. This is being taken as additional confirmation of the reliability of the conjugate fluid flow-heat transfer model. Also, it seems that this model may be employed to estimate the size of equiaxed zone in continuous casting of steel.

View full abstract

Small-angle neutron scattering (SANS) measurements were carried out to investigate the structural change during the early stage of phase decomposition in Fe-Cu-Ni-Mn quaternary alloy. Data analysis of scattering intensities showed that the nuclear Guinier radius becomes larger than the magnetic one beyond about 2 nm. This is suggested to be attributable to the formation of segregated layer enriched by Ni and Mn around fcc copper-rich precipitate. This segregation was confirmed directly by atom-probe (AP) analysis and approved from a thermodynamic point of view. The calculated SANS intensities are well fitted to experimental data of both magnetic and nuclear components of neutron scatterings. Addition of Ni and Mn in Fe-Cu binary alloy was suggested to promote the precipitation reaction of ε-Cu phase.

View full abstract

A numerical model has been presented in this paper which simulates the austenite to pearlite phase transformation undergone by a 0.8% plain carbon steel. Both heat transfer and phase transformation were modelled and the model was verified by matching the complete cooling curve. The macro-microscopic model developted for the austenite-pearlite (eutectoid) reaction was in particular verified for its accurary by matching eutectoid hold time. However in this work the nucleation rate was not modelled and the growth rate was modelled with a fixed number of nucleants. The results including the undercooling obtained were satisfactory in spite of these drawbacks.

View full abstract

The behavior of precipitation hardening in two types of Fe-Cu alloys has been investigated by means of mechanical tests as well as small angle neutron scattering measurements. The integrated intensity increased first and reached a constant corresponding to the completion of precipitation reaction, while particle radius increased monotonically with aging time, where Vickers hardness and yield stress increased and reached maxima, then decreased. The interaction force with a dislocation due to each precipitate was very small compared with the force by the Orowan mechanism. After discussion based on three possible mechanisms in terms of coherency strain, elastic modulus change and interfacial energy, the strengthening was suggested to be caused from the coherency strain effect. The first increase of yield stress during aging is attributed to the growth in size of clusters and the decrease of yield stress after the maximum is mainly related to the decrease of number density of precipitates. It was found that the loss of coherency with the matrix greatly lowers the strengthening effect, whereby the structure of precipitates changes from bcc to fcc during aging.

View full abstract

In this paper, a study is presented about the prediction of the r-value of several low-carbon low-alloy steel sheets by means of Taylor-modelling. The purpose of this study was not to identify suitable models predicting the r-values for each type of steel separately, but to find one single model that would be able to predict r-values for any kind of low carbon steel sheet as well as possible.
It was shown that the instantaneous r-value diminishes during a tensile test. This evolution is coupled to a change in texture, which was measured and simulated by a Taylor-model using both ideal and real textures. The r-vaues were calculated in two different ways. The first method used was the series expansion method, the second method is one in which the texture evolution during tensile testing is simulated. A quantitative comparison between measured and calculated data was made, taking the errors introduced by tensile testing and texture measurements into account.
It will also be shown that for the purpose of quantitative data analysis, the q-value (contraction coefficient) deserves some preference over the commonly used r-value.

View full abstract