Modeling heat transfer in packed bed processes such as iron ore pelletization is essential for optimizing process operation and furnace design. In these systems, materials like magnetite iron ore undergo thermal and chemical transformations, where heat and mass transfer are often coupled with heat effects from multiple processes — including both exothermic chemical reactions (e.g., oxidation) and endothermic physical changes (e.g., drying and sintering). As the industry moves towards fossil-free ironmaking, it becomes increasingly important to isolate pure heat transfer behavior, independent of chemical reactions, to support the development of sustainable process schemes.

This study investigates convective heat transfer in a packed bed by evaluating several established correlations against the conventionally used modified Ranz–Marshall correlation. Pilot pot-scale experiments were performed by isothermally heating 120 kg of already indurated iron ore pellets at 300 °C to avoid chemical reactions, with additional experiments at 500 °C and 700 °C to assess performance at elevated temperatures typical of industrial pelletization.

Results show that the modified Ranz–Marshall correlation underpredicts heat transfer rates under conditions isolating convective heat transfer, reinforcing the need for this investigation. The Wakao–Funazkri and Rowe–Claxton correlations provided the best agreement with experimental data, particularly at 300 °C. Minor deviations at higher temperatures suggest the influence of unaccounted variables, warranting further study. Sensitivity analysis identified gas velocity as the most significant parameter affecting heat transfer. The results suggest that adopting the Wakao–Funazkri and Rowe–Claxton correlations can provide a stronger basis for predictive pellet heat transfer modeling in future process design and simulation work.