Optimization of the pelletizing process of nickel-iron ores in a rotary disc using a novel binder agent

Abstract

This study evaluates Poços de Caldas clay (ARG Caldas) as a potential bentonite substitute for pelletizing pyrometallurgical nickel‑iron residues (i.e., rotary kiln dust). The investigation employed a two-stage experimental approach: an initial 2k Factorial Design followed by a Central Composite Design (CCD). The 2k design assessed the effects of disc filling percentage (%p), binder mass percentage (%b), and rotational speed (V) on mean particle diameter (d50) and compressive strength, achieving R2 values of 0.94 and 0.95, respectively, confirming the influence of these variables. Subsequently, the CCD incorporated the ARG Caldas mass percentage (%ARG) as a fourth variable, confirming the significant influence of all parameters, where R2 for d50 was 0.88, and R2 for compressive strength was 0.81 at a 10% significance level. Optimization aimed to maximize compressive strength while maintaining d50 between 9.0 and 16.0 mm. The optimal conditions identified were a d50 of 16.0 mm and a compressive strength of 68.45 N. Validation experiments showed high predictive accuracy, with measured values of 15.6 ± 0.2 mm and 65.0 ± 3.2 N (relative errors of 2.50% and 5.04%). These results validate ARG Caldas as a technically viable complementary binder, provided that pellet diameter and strength are optimized according to established operational parameters.