The use of medium and low-grade iron ore is gradually becoming more important due to the depletion of high-grade iron ore reserves and stringent environmental acts/rules. Slimes from iron ore washing were discarded in tailing dams; however, there is currently consideration for recovering iron values from ultra-fines as well. There are enormous fine dumps that are still unutilised. Hence, this study attempt to delve the optimization of iron ore slimes an indeed requirement for manufacturing and design in industries. Leveraging a flocculation process, coupled with the implementation of an Artificial Neural Network (ANN) predictive model, the Kiriburu processing plant serves as the primary source for iron ore slime samples. Chemical analyses of the collected iron samples reveal a composition featuring 58.24 % iron content, 3.47 % Al2O3, 4.72 % SiO2, and 5.18 % LOI (Loss on Ignition). The investigation explores the performance of the flocculation technique under varying pH levels, different pulp densities, and diverse flocculant dosages. Furthermore, the varying parameters selected are pH from 6 to 11, pulp density from 1 % to 15 %, and flocculant dose from 0.03 to 0.27 mg/g. The study's findings showcase a substantial improvement in the Fe grade of iron ore, escalating from 58.24 % to 66.12 %, with an impressive recovery rate of 82.54 % achieved using a flocculant dosage of 0.09 mg/g at pH 10. Additionally, a performance assessment of the selective flocculation method for iron ore slimes is conducted using an ANN predictive model, with recovery as the pivotal parameter. The input parameters for this model encompass pH, pulp density, and flocculant dosages. Employing a three-layer ANN model with a 3–3–1 architecture and utilizing feed-forward back propagation, the study demonstrates a close alignment between predicted values and experimental data, confirming the model's effectiveness for practical manufacturing applications. Information regarding the potential applications of the model's iron ore slime beneficiation efficacy for the manufacturing sector should be considered. This could entail lower waste, more effectiveness, or cost savings. Emphasise any possible ramifications for sustainability or the environment that would make the study pertinent in a larger perspective.
Read full abstract