Solidification of melts in the CaO-“Fe2O3”-SiO2 system

Abstract

In order to provide further insight into the chemical reactions and phase transformations taking place during iron ore sintering fundamental studies have been undertaken on the solidification of liquids in the “Fe2O3”-CaO-SiO2 system in air. A new experimental technique and analysis methodology has been developed that enables the phases and microstructures formed at selected temperatures to be identified. The technique involves the melting and solidification of a small mass of an oxide powder mixture at a controlled cooling rate followed by rapid quenching. The resulting microstructures and phases present in the sample are characterised through the use of electron probe micro X-ray analysis (EPMA) measurements. The new approach makes it possible to systematically investigate the influence of key process variables and to accurately measure the compositions of the phases and their sequence of formation. In addition, an improved experimental technique has been used to measure the temperatures and oxygen partial pressures experienced by the materials in the sinter bed during pilot scale sintering. The phases formed during the cooling of high iron melts have been shown to critically depend on the thermodynamic properties of the system, the crystal structures of the phases and the kinetics of the solidification processes. Of particular interest to industrial iron ore sintering operations is the formation of the silico-ferrite of calcium (SFC) phase, which is isomorphic with silico-ferrite of calcium and aluminium (SFCA). The focus of the experimental study has been on bulk liquid compositions for which it is anticipated that the silico-ferrite of calcium (SFC) phase would form on equilibrium solidification. The effects of melt composition, the initial primary phase field, Fe2O3 concentration, CaO/SiO2 ratio, and cooling rate were investigated. The phases identified in the solidified products include, hematite, dicalcium silicate, calcium diferrite, calcium monoferrite, SFC-I and Ca7.2Fe2+ 0.8Fe3+ 30O57. The results of the experimental studies show that the solidification paths do not follow those anticipated from equilibrium or Sheil-Gulliver cooling. Significantly, the SFC phase does not form from the solidification of the melts at any of the conditions investigated in the “Fe2O3”-CaO-SiO2 system in air. The complex morphologies and phase assemblages in the final products are shown to be the result of solidification of the melt and include; equiaxed facetted hematite and magnetite; dendritic di-calcium silicate; divorced and coupled structures of di-calcium silicate and hematite; individual calcium diferrite crystals; coupled structures of calcium diferrite and di-calcium silicate; individual calcium monoferrite crystals; coupled structures of calcium monoferrite and di-calcium silicate; coupled structures of calcium diferrite, di-calcium silicate and calcium monoferrite. The sequences of phases formed on solidification of iron rich compositions are more closely described by Sheil-Gulliver cooling of a metastable liquidus of the “Fe2O3”-CaO-SiO2 system that does not include the SFC primary phase field. The heterogeneous nucleation of SFC on hematite appears not to be favoured. The SFCA-I phase, having a crystal structure related to SFCA, is observed to nucleate homogeneously from the melt and preferentially the surfaces of magnetite under conditions where this latter phase is present. The formation of secondary hematite in industrial sinter has been shown to take place through the oxidation of magnetite at sub-liquidus and sub-solidus temperatures. The study lays the foundation for further systematic studies in the system “Fe2O3”-CaO-SiO2 system in the iron rich region including minor concentrations of Al2O3 and MgO, for compositions that more closely resemble those produced in industrial processing operations.