Estimation of rate parameters for reduction of iron ore–graphite composite pellets in packed bed reactor using genetic algorithm

Abstract

A semiempirical kinetic model has been developed to determine the course of reduction of iron ore–graphite composite pellets over time in a laboratory scale side heated packed bed reactor attached with a tailor made bottom hanging thermogravimetric set-up. The rate parameters in the model, especially the three sets of apparent activation energy values and frequency factors associated with the reduction of iron oxides in three elementary steps, namely hematite to magnetite, magnetite to wustite and wustite to iron, have been estimated based on experimental data by employing an optimisation tool, the genetic algorithm (GA). The difference between the predicted and experimental degree of reduction is minimised to obtain the rate parameters. The experimental degree of reduction is calculated based on mass loss data during reduction and the exit gas analysis. Estimated values of apparent rate parameters were found to be of the same order of magnitude to their intrinsic counterparts reported in literature. Finally, by using the predicted rate parameters the temporal evolution of various oxide phases as well as pure iron has been evaluated.