Continuous Removal of Fluoride by Fe/Al-Anchored Bamboo Charcoal in Fixed Bed Column and Prediction of Breakthrough Curve using Artificial Neural Network Model

The fluoride removal from drinking water gets world attention. To effectively manage the fluoride content in potable water, a low-cost, renewable iron/aluminum-anchored bamboo charcoal (Fe/Al-BC) was synthesized using waste bamboo at first time. The continuous fluoride removal efficiency was assessed by a fixed-bed column, and the entire adsorption process was predicted employing an artificial neural network (ANN) model. The breakthrough time (t_b), exhaustion time (t_e), and adsorption capacity (q_eq) of the column bed all increased as the height of the column bed rises, while they decreased as the influent flow rate increases. However, q_eq differed from t_b and t_e in that q_eq improves with increasing the concentration of influent and it achieved 4.93 mg/g when the fluoride concentration was 20.45 mg/L. The ANN model accurately predicted the entire breakthrough behavior of fluoride removal by Fe/Al-BC in column experiments, achieving a high correlation coefficient (R² = 0.9990) and demonstrating potential to guide large-scale field applications. The parameters fitted by the ANN model indicated that the total effluent time has the highest relative importance for q_eq, at 36.70%, followed by flow rate > initial concentration > column height. Although the q_eq decreased with the increase of regeneration times, the regeneration efficiency still reached 25.5% after four cycles. The lifetime of Fe/Al-BC avoiding the t = 0 to reach t_b was 5.7 cycles, whereas, the q_eq dropped to zero after 4.7 cycles. The primary adsorption mechanisms of Fe/Al-BC for fluoride involve the formation of inner-sphere surface complexes of M-F (M: Al, Fe) through ligand exchange and ion exchange reactions, along with contributions of electrostatic attraction and hydrogen bonding.

Graphical Abstract