Oxide Solubility Minimum in Liquid Fe-M-O Alloy

Abstract

The origin of the solubility minimum of oxide (M{sub x}O{sub y}) in liquid Fe-M-O alloy was investigated, and the minimum was predicted based on thermodynamic calculations. Due to the characteristic property of activities of M and O in the liquid, a maximum exists in the product between the two activities if the affinity of M to O is significantly high, as most deoxidizing elements are. A critical activity product is defined, which is an indicator of the solubility minimum of the M{sub x}O{sub y} in the liquid Fe-M-O alloy according to the following relationship: max(a{sub M}{sup x}×a{sub O{sub -}{sup y}})=K{sub M{sub xO{sub y}}}×a{sub M{sub xO{sub y}}}, where the a{sub M{sub xO{sub y}}} is unity if the alloy is in equilibrium with the pure M{sub x}O{sub y}. The origin of the solubility minimum was explained using the change of the activity product by composition. Available CALPHAD assessments for several binary Fe-M liquid alloys and Wagner’s solvation shell model were combined to calculate the activity product in the Fe-M-O alloy, which can be used to predict the solubility minimum of M{sub x}O{sub y}. A favorable agreement was obtained when M=Al, B, Cr, Mn, Nb, Si, Ta, Ti, V, and Zr. The Gibbs energy of dissolution of O in pure liquid M (Δg{sub O{sub -(M)}{sup ∘}}) and the Gibbs energy of the formation of M{sub x}O{sub y} per mole of atoms (Δg{sub M{sub xO{sub y}{sup ∘}}}/(x+y)) play important roles in determining the solubility minimum, as long as an interaction between Fe and M is less significant than the interaction between metal (Fe and M) and O. Predictions of the solubility minima of CaO and MgO were not satisfactory, requiring further improvement of the present analysis.