Regression model, artificial intelligence, and cost estimation for phosphate adsorption using encapsulated nanoscale zero-valent iron

Abstract

This study investigated the adsorption of PO43− onto encapsulated nanoscale zero-valent iron (nZVI). At initial PO43–: 10 mg · L−1, the optimum condition was initial pH: 6.5, nZVI dosage: 20 g · L−1, stirring-rate: 100 rpm, and time: 30 min, achieving PO43− removal of 42%. The effect of pH and time on the PO43− removal efficiency was quadratic-linear concave up, whereas the curve of nZVI dosage was quadratic-convex. Artificial neural network with a structure of 5−7−1 adequately predicted PO43− removal (R2: 97.6%), and the sensitivity analysis demonstrated that pH was the most influential input. The cost of the adsorption unit was 3.15 $USD · m−3.