No abstract

Many chemical reactions take place simultaneously during the induration of iron ore pellets produced from magnetite concentrates. Two of the most important are magnetite oxidation and calcination of carbonate fluxes. The first reaction consumes oxygen diffusing into the pellet, while the second reaction produces carbon dioxide that must diffuse out of the pellet. A mathematical model combining the two reactions and gaseous diffusion within the pellet has been developed to quantify the interaction between the two reactions. This combined mathematical model showed that current induration plant mathematical models for the mass and energy balance around a pellet furnace are inaccurate in treating magnetite oxidation and flux calcination as separate reactions. Assuming separate reactions can lead to an error of up to 20 pct conversion of magnetite at the end of the preheat stage. This combined mathematical model, confirmed by experiments with single pellets, also demonstrated that calcination of fluxes also tends to follow a ''shrinking core'' model rather than reacting simultaneously across the pellet, as existing whole plant models assume. Modifying induration plant mathematical models in accordance with the findings of this article could lead to further savings in energy costs for pellet plants.

Although extensive development has occurred in recent years for models of induration of iron ore pellets, none of these models have taken into account the partial melting of some of the raw materials. To determine the importance of partial melting and melt phase formation to the energy balance for induration, estimates have been made of the thermodynamic properties of silicoferrite of calcium, SFC, using published techniques. SFC was used in this paper as an example of the initial melt forming minerals in the pellets. Owing to the complexity of the structure of SFC, there was some doubt as to the accuracy of these estimates as they suggested that SFC was not a thermodynamically stable phase, though it exists between 1 050 and 1 250°C. As no experimental data was available, however, these 'best available' estimates have been incorporated into a mathematical model to determine the effect of melt phase formation on the induration process. It was found that while there was little effect on the pellet temperature profiles, the overall amount of energy required to indurate the pellets increased by about 1-1.5 % when melt phase formation was included. This suggests that experimental determination of the thermodynamic properties of SFC and other phases produced from the melt would be of benefit in modelling the energy requirements for induration of pellets more accurately.

A solidified layer on the inside of a cooled flow channel can be used to control the flow rate of molten material through that channel. This concept can be used for flow rate control of molten furnace products in the metallurgical industry. In this study, internal solidification of molten metal flows has been modeled mathematically for both steady-state and dynamic cases. The model predicted solidified layer thickness and metal flow rate. Experimental verification of the mathematical model was obtained using molten tin. Novel design features of the experimental apparatus included the use of boiling heat transfer and the vertical mounting of the cooling section. Engineering knowledge regarding the design, operation, and control of a pilot scale (24 kg/s) molten metal circuit was obtained during the construction, commissioning, and operation of the experimental apparatus. Experimental results for tin flow rate from the experimental apparatus were within experimental error of the predictions of the mathematical model.

During induration of iron ore pellets produced from magnetite concentrates, there are many reactions occurring simultaneously. While the magnetite is being oxidised to haematite, carbonate fluxes are being calcined, leading to the formation of complex oxides. Calcination of fluxes and the formation of calcium and magnesium ferrites and silicates during the preheat stage can influence the magnetite oxidation process. As the calcination reaction proceeds quicker than oxidation at temperatures experienced during the preheat stage (700-1200uC), the bulk flow of CO 2 emanating from within the pellet should disrupt the diffusion of O 2 to the magnetite reaction as the gas fluxes are similar in magnitude. This increases the time required for the oxidation of the pellet. Using an infrared rapid heating furnace, experiments were performed on heating individual magnetite green balls in an air atmosphere. These experiments showed that time played an important role in the extent of oxidation and calcination and temperature significantly influenced the initial evolution of high temperature phases such as magnesioferrite, calcium ferrites and calcium silicates. Understanding the kinetics of all these reactions for a particular pellet feed and their interaction is important for optimising the design of the induration furnace.