Experimental and Modelling Research of Slag Properties in Nickel Laterite Smelting

"Knowledge of slag properties is required for the optimisation of smelting operations. The phase equilibria and viscosities of the Al2O3-CaO-FeO-Fe2O3-MgO-MnO-NiO-Cr2O3-SiO2 slags at conditions of interest to Koniambo Ni-laterite smelting operations have been investigated. High-temperature re-equilibration phase equilibrium experiments of industrial slags followed by rapid quenching and direct measurement of phase compositions using electron probe X-ray microanalysis (EPMA) have been undertaken in the olivine and pyroxene primary phase fields at temperatures between 1475 and 1600°C and SiO2/MgO weight ratios between 1.57 and 2.09. Phase equilibria predictions were performed using FactSage thermodynamic software. A slag viscosity model available in the FactSage thermodynamic package has been used to calculate the viscosities of the liquid slags. The Einstein-Roscoe equation combined with the predicted phase equilibria data have been used to estimate the viscosities of the slag slurries (slag + solids). The information provided can be used to assist in the evaluation of the furnace performance, to assess further optimisation of slag composition and to explore possibilities of targeted feed changes.INTRODUCTION One of the key steps in the production of ferronickel from nickel laterite ore at Koniambo Nickel SAS involves reduction and smelting of pre-dried nickel-bearing saprolitic ore (high Mg and Si) in an electric furnace at temperatures up to 1650°C. The characterisation of the slag properties as functions of key operational parameters, such as, temperature and composition, including liquidus, the proportions of solids and viscosity, is essential for improving control and optimisation of the operations. The slag produced in the smelting process is a complex oxide system with the major chemical components including Al2O3, CaO, Cr2O3, “FeO”, MgO, “MnO”, “NiO” and SiO2. Characterisation of these multi-component systems using a solely experimental approach becomes increasingly complex as the number of components is increased."