Coal fly ash (CFA) obtained from pulverized coal furnaces is a highly refractory waste that can be used for alumina and rare-earth elements (REEs) extraction. The REEs in this type of CFA are associated with a mullite and amorphous glassy mass that forms a core-shell structure. In this research, it was shown that complete dissolution of amorphous aluminosilicates from the mullite surface with the formation of the low-alkali mullite concentrate prior to sulfuric acid leaching with the addition of (NH₄)₂SO₄ helps to accelerate the extraction of REEs. The extraction degree of Sc and other REEs reaches 70–80% after 5 h of leaching at 110 °C and acid concentration of 5 M versus less than 20% for the raw CFA at the same conditions. To study the leaching kinetics of the process, the effects of temperature (90–110 °C), liquid-to-solid ratio (5–10), and leaching time (15–120 min) on the degrees of Al and rare-earth elements (REEs) extraction were evaluated. After 120 min of leaching at 110 °C and L/S ratio = 10, the extraction of Al was found to be lower than 30%. At the same time, total REEs (TREE) and Fe extraction were greater than 60%, which indicates that a part of the TREE was transferred into the acid soluble phase. After leaching, the residues were studied by laser diffraction (LD), X-ray diffraction (XRD), X-ray fluorescence (XRF), and scanning electron microscopy (SEM-EDS) to evaluate the leaching mechanism and the solubility of Al- and Fe-containing minerals, such as mullite, hematite, and amorphous aluminosilicate.

Rare earth elements (REEs) and Sc are concentrated in aluminum production byproducts. The novel REEs recovery approach, which involves leaching with acid at a pH > 3 in the presence of MgSO₄, results in the formation of a pregnant leach solution (PLS) with a low concentration of iron (Fe) and titanium (Ti) and a large number of valuable elements. This work studies the application of chelating resin Puromet MTS9580 in the sorption recovery of Sc from sulfate solutions. To analyze the static Sc sorption data, Langmuir, Freundlich, and Temkin isotherm models were used. The Langmuir isotherm model was the best fitted to the experimental data, with a coefficient of determination (R²) of 0.983. The dynamic adsorption experiment was conducted using a PLS and a simulated solution without contaminants. Adsorption of Sc from the simulated solution was better fitted to the Thomas model with a Sc capacity greater than 6.4 mg mL⁻¹. Because Ti had a gradual decrease in C/C₀, which the Thomas model was unable to simulate, the modified dose-response (MDR) model fitted better with PLS with a Sc capacity greater than 3.8 mg mL⁻¹. The NaHCO₃ solution (200 g L⁻¹) effectively desorbed Sc (>98%) from simulated and PLS solutions after 1.5 h of stirring in a batch mode. After 1.5 h of desorption, the concentration of Sc in the desorption solution was 461.5 mg L⁻¹, while the concentration of Mg and Ti was lower than 200 mg L⁻¹ and 50 mg L⁻¹, respectively.

Rare earth elements (REEs) and Sc are concentrated in aluminum production byproducts. The novel REEs recovery approach, which involves leaching by acid at a pH > 3 in the presence of MgSO4, results in the formation of a pregnant leach solution (PLS) with a low concentration of iron (Fe) and titanium (Ti) and a large quantity of valuable elements. The ion exchange method was acknowledged as the most favorable method used to separate the REEs from PLS. This work studies the application of chelating resin Puromet МТS9580 in the sorption recovery of Sc from sulfate solutions. To analyze the static Sc sorption data, Langmuir, Freundlich, and Temkin isotherm models were used. The Langmuir isotherm model was the best fitted to the experimental data, with a coefficient of determination (R2) of 0.983. The dynamic adsorption experiment was conducted using a PLS and a simulate solution without contaminants. Adsorption of Sc from the simulate solution was better fitted to the Thomas model with a Sc capacity greater than 7 mg mL–1. Because Ti had a gradual decrease in C/C0, which the Thomas model was unable to simulate, the modified dose-response (MDR) model fitted better with PLS. The NaHCO3 solution (200 g L–1) effectively desorbed Sc (>98 %) from simulated and PLS solutions after 1.5 h of stirring in a batch mode.