Values for interaction parameters in liquid iron alloys collected by the 19th Committee (Steelmaking) of The Japan Society for the Promotion of Science are shown for some elements to describe the present state of available data. It is shown that though the interaction parameters have been repeatedly measured, the data are limited to some particular elements. Even for these elements, the data are not satisfactory. The periodicity of interaction parameter with atomic number is examined. The distinct periodicity reported by several investigators is not apparently revealed because of the scatter of data, although some tendencies are observed.
The literature collected by the committee are reviewed for the experimental and the data processing and analyzing methods. Numerous interesting methods were devised in the effort to determine the interaction parameters.

View full abstract

The effect of Na₂O additions to the CaO-CaF₂-SiO₂ system on the partition of vanadium, niobium, manganese and titanium between CaO- CaF₂-SiO₂ melts doubly saturated with CaO and 3CaO•SiO₂ and molten carbon saturated iron was investigated at 1300°C. The partition of vanadium, niobium and titanium increased with increasing content of Na₂O. On the other hand the manganese partition decreased, indicating that manganese oxide behaves as a basic oxide in the CaO-Na₂O-CaF₂-SiO₂ system.
Since the vanadium, niobium and titanium have various valencies in ionic melts of slag, the ionic species of these three elements were determined. Each element was shown to exist in slag mainly as Nb⁵⁺, V⁴⁺ and Ti⁴⁺, respectively.
Addition of a small amount of Na₂O to the CaO-CaF₂-SiO₂ system enhances the capacity of removing niobium, vanadium and titanium from hot metal to slag. It was proved that the yield of manganese increases and the recovery of niobium, vanadium and titanium to slag increases by addition of Na₂O to CaO-CaF₂-SiO₂ flux for hot metal treatment.
Vanadium, niobium, manganese and titanium capacities were calculated and the relationships between these capacities and carbonate and phosphate capacities are discussed.

View full abstract

The dephosphorization reaction of low carbon Fe-P alloy with Na₂CO₃ has been studied at 1300°C under an argon atmosphere.
The results obtained are summarized as follows:
(1) As carbon monoxide is evolved during the reaction and the carbon content in the alloy increases after the reaction, it is concluded that the dephosphorization reaction can be expressed as follows: Na₂CO₃+2/3[P]=2/3Na₃PO₄+1/3CO+2/3[C]
(2) The range of the average oxygen partial pressure, P_O2 (atm), during the reaction estimated on the basis of the Na-C-O₂ phase diagram and of the phosphate capacity as follows: log P_O2 (atm)=-17--19

View full abstract

Phosphorus distribution between MgO_sat.-BaO-Cr₂O₃-Fe_tO(-BaF₂) slags and liquid iron containing Cr ([%Cr]=0-16) has been measured in the temperature range from 1550 to 1650°C. The refining limit of dephosphorization in steel containing Cr and hot metal was discussed by using the values for phosphate capacity C_P(=(%P)/[a_P•a^5/2_O]). Thermodynamic analysis indicates that the phosphorus partition ratio increases with C up to a critical value and decreases beyond that.

View full abstract

Experiments were done on the distribution of each element between liquid Fe-Cr alloys and CaO-MgO-Al₂O₃-SiO₂ slags having the compositions near the CaO-Al₂O₃ binary system in a magnesia crucible at temperatures of 1575, 1600 and 1650°C. The distribution ratios of chromium and sulphur, the oxygen content of liquid alloy and the activity coefficient of CrO in slag were expressed as functions of temperature and compositions of liquid alloy and slag. The major equations derived are as follows. log (wt%O)=0.413 log N_CrO+0.0065 [wt%Cr] -0.091 ∑K_i•N_i-13044/T+5.574 log (wt%Cr)/[wt%Cr]=2.04 log [wt%O] -0.048 [wt%Cr]+0.292 ∑Ki•Ni+26 805/T-10.487 log γ_CrO=-0.561 log N_CrO-0.096∑K_i•N_i -3044/T+1.528 log (wt%S)/[wt%S]=-0.47 log N_CrO+0.0087 [wt%Cr] +2.835 ∑K_i•N_i+6902/T-4.28 where, ∑K_i•N_i=N_CaO+0.1N_MgO-0.8N_Al2O3-N_SiO2

View full abstract

Dephosphorization experiments under oxidizing condition were carried out to develop an inexpensive flux for the chromium-containing iron with a large dephosphorizing capacity. Among various halides tried in CaO-halide-Cr₂O₃ systems, NaF and BaF₂ are superior to others in their dephosphorization performance for chromium-containing iron though the fluxes are not inexpensive. From view points of both the dephosphorization ability and the flux cost, CaF₂ was selected among halides and BaO was added in order to contrive their compatibility. In the CaO-BaO-CaF₂-Cr₂O₃ system, various composition was tried and the composition was optimized. As the result, 90% of maximum dephosphorization degree was obtained with 100g of flux per kilogram of the metal at 2% C and 15% Cr.
Effect of carbon content and chromium content in the metal on dephosphorization was investigated with this CaO-BaO-CaF₂-Cr₂O₃ flux. As the result, it became possible to dephosphorize the iron with wider range of carbon (0.2 to 4% C) and chromium (15 to 22% Cr) contents. It was revealed that coexistence of calcium, barium, and sodium as cation was effective in dephosphorization. Fluorine as an anion was more effective in dephosphorization than chlorine.

View full abstract

In this work, the thermodynamic treatment of the equilibrium distribution coefficient k^X_o in iron base ternary alloys described in the previous work1, 2) was developed for iron base multi-component alloys and the effects of solute-interactions on the equilibrium distribution coefficient in multi-component iron alloys were discussed thermodynamically. A new paramete for the evaluation of such effects was introduced. Furthermore, the equilibrium distribution coefficients of Cr, Ni, Si, Mn and Mo at various compositions in Fe-Cr-Ni base alloys were measured. Small dependence of k^X_o of those elements on the concentrations of Cr and Ni in those alloys was found and those results were explained by the application of new parameter.

View full abstract

The dense wustite plates containing SiO₂ and/or Al₂O₃ have been reduced isothermally in a stream of hydrogen at temperatures between 670 and 930°C. The specimens used in this work consist of FeO phase with a dissolved part of added oxides, fayalite and hercynite.
Reduction of FeO is accelerated by the presence of SiO₂ in a small amount at temperatures above 800°C, but it is retarded at temperatures below 730°C. On the other hand, reduction of FeO is depressed by the presence of Al₂O₃ dissolved in the FeO phase, over the temperature range studied in this work. Unlike the single oxide dissolution, simultaneous dissolution of SiO₂ and Al₂O₃ in the FeO phase accelerates reduction over the entire range of temperature studied.
The acceleration effect of reduction due to the existence of oxides can be observed if the reduction rate is relatively slow.

View full abstract

A new method for analysis of solidification process of carbon steel is proposed. In the method, the equations of diffusion and heat conduction are solved simultaneously; δ/γ transformation, precipitation of MnS and change of dendrite arm spacing with cooling rate are included. Numerical examples are presented.

View full abstract

The lattice misfits between γ and γ' phase in single crystal Ni-base superalloys were accurately measured for filings of specimens by X-ray diffractometry combined with the deconvolution method. The (331) reflection peaks measured with a conventional X-ray diractometer were in a cluster of overlapping peaks because of the very small differences in the lattice parameters of both phases and the instrumental broadening due to X-ray optics including the spectral distribution of X-ray source such as CuKα doublet.
The deconvolution method was applied to remove the instrumental broadening from the (331) peaks measured with the conventional diffractometer. The instrumental window function, which is the most important factor for separation of the instrumental broadening and is necessary for the unfolding of clustered peaks, was calculated from (533) CuKα doublet reflection of Si standard by a non-linear least square method. The cluster of peaks was separated into individual peaks without previous knowledge of the number of peaks by the deconvolution with the instrumental window function and then the misfits between γ and γ' phase of specimens were accurately determined. Therefore, this method was a powerful tool for the measurement of the lattice misfit between γ and γ' phase in single crystal superalloys.

View full abstract