metal/crucible interface using FE-SEM (Field Emission Scanning Electron Microscope).9) The composition was also analyzed quantitatively with EDS by measuring the solid solubility of Fe, Cr and/or Ni oxides in Ti 3 O 5 at metal/crucible interface.Numeral analysis by Redlich-Kister type polynomial on titanium deoxidation of molten Fe-Ni alloy was carried out in order to extend applicable composition range of quadratic formalism in the previous report.8) The experimental result of titanium deoxidation in molten Fe-Cr alloy has been numerically analyzed by applying Redlich-Kister type polynomial to the excess Gibbs free energy change of mixing for Fe-Cr-Ti-O system in the present work. In the sequence of numerical analysis, the interaction parameter between Cr and Ti (W Cr-Ti ) of Redlich-Kister type polynomial was assessed using the experimental result in Fe-Cr-O system.10) Also, binary interaction parameter of Redlich-Kister type polynomial between Cr and Ti (W Cr-Ti ) was deduced. Finally, validity of those evaluated parameters was confirmed by comparing the present experimental results with the evaluated equilibrium relation between titanium and oxygen concentration in molten Fe-Cr-Ni alloy.
ExperimentalDeoxidation of Fe-Cr and Fe-Cr-Ni alloys with titanium were carried out in a Ti 3 O 5 crucible to fix the activity of titanium oxide and to avoid contamination of those alloy from crucible by induction furnace. Details of Ti 3 O 5 crucible production and an experimental method of titanium deoxidation are given previous report. 8) Temperature of the melt was monitored by an infrared radiation pyrometer calibrated by the melting temperatures of Fe and Cu. The samples of Fe-Cr and Fe-Cr-Ni alloys were quenched by turning off the power of the furnace and impinging helium gas on the surface of the melt after the predetermined equilibrium time of 60 min and 90 min, respectively.The quenched sample was cut, polished by abrasive paper of SiC and then electrolytically-polished for 3 min under the condition of 5 V-0.5 A. Mixed acid of 80 mass% acetic acid and 20 mass% perchloric acid was used for the electrolyte. The sample was ultrasonically cleaned in acetone and immediately supplied to O analysis by inert gas fusion infrared absorptionmetry to avoid effect of oxygen adsorption on the sample surface. Analysis of Cr, Ni, Fe and Ti contents was conducted by induction coupled plasma emission spectrometry.All the samples were analyzed by FE-SEM EBSD in order to determine the oxide coexisting with the alloy. The solid solubility of Fe, Cr and Ni oxide in titanium oxide phase was also analyzed by EDS, quantitatively.
Experimental ResultsTitanium deoxidation experimental results in molten Fe-Cr and Fe-Cr-Ni alloys are listed in Tables 1 and 2 and the results are reproduced in Figs. 2 to 5 with the results in molten pure Fe.6,7) The curves in Figs. 2 to 5 show the calculated relation between Ti and O based on analysis by Redlich-Kister type polynomial and they will be discussed in Chap. 4. It is shown in Fig. 2 that the oxygen c...