Vanadium removal from mining ditch water using commercial iron products and ferric groundwater treatment residual-based materials

• DOI: 10.1016/j.chemosphere.2021.131817
• Publication Date: 2021-08-06T00:25:26Z
• Authors: Ruichi Zhang; Jinmei Lu; Mark Dopson; Tiina Leiviskä
• Keywords: Waste iron product; 550; Iron; 0211 other engineering and technologies; Vanadium uptake; Water; Vanadium; 02 engineering and technology; Hydrogen-Ion Concentration; Miljövetenskap; 6. Clean water; Water Purification; Hazard quotient; Iron sorbent; Adsorption; Mining effluent; Groundwater; Environmental Sciences; Water Pollutants, Chemical

Removal of vanadium from liquid waste streams protects the environment from toxic vanadium species and promotes the recovery of the valuable metal. In this study, real mining ditch water was sampled from a closed vanadium mine (V-Fe-Ti oxide deposit, Finland) and used in sorption experiments at prevailing vanadium concentration (4.66-6.85 mg/L) and pH conditions (7.02-7.83). The high concentration of vanadium in the water represents a potential health concern according to the initial risk assessment carried out in this study. Vanadium was efficiently removed using four different iron sorbents: ferric oxyhydroxide with some goethite (CFH-12), poorly crystallized akaganéite (GEH 101), ferric groundwater treatment residual (GWTR), and GWTR-modified peat (GWTR-Peat). Higher dosage (6 g/L with 24 h contact time) and longer contact time (72 h using 1 g/L dosage) resulted in removal efficiencies of higher than 85%. Kinetic data were well represented by the Elovich model while intra-particle diffusion and Boyd models suggested that the sorption process in a real water matrix was significantly controlled by both film diffusion and intra-particle diffusion. Column studies with CFH-12, GEH 101, and GWTR-Peat showed that the breakthrough started earlier with the mining ditch water compared to a synthetic vanadium solution (investigated only with CFH-12), whereas GEH 101 proved to have the best performance in column mode. The Thomas and Yoon-Nelson column models were found to agree with the experimental data fairly well with the 50% breakthrough time being close to the experimental value for all the studied sorbents.