Effect of Surfactant Concentration on Zeolite Modification for Nitrate Removal and Its Simulation with Kinetics Models and ANFIS in Fixed-Bed Column

Abstract

Nitrate removal is important for reuse of agricultural drainage, controlling nutrition and reducing its environmental impacts. For this purpose, the effect of zeolite modification with concentrations of 0, 10, 15, 20 and 25 mg/L of HDTMABr surfactant, in columns with height of 30 cm and 32 mm diameter were investigated. In order to modelling the breakthrough curve of adsorption column, three models of Bohart-Adams, Thomas and modified dose-response plus Artificial intelligence model of ANFIS were evaluated. The results showed that when the equilibrium adsorption capacity is increased from 0.07 to 0.74 mg/g, the breakthrough and exhausting times are also increased. Modification of natural zeolite with a concentration of 25 mg/L of surfactant, was selected as the best surfactant concentration. Among the adsorption models, Thomas model and the modified dose-response model were able to predict the dynamic absorption coefficients, but the modified dose-response model, was more efficient and more accurate. The results of the study showed that Increasing the concentration of surfactant leads to an increase in the equilibrium adsorption capacity. ANFIS with average relative error of 23.3%, root mean square and chi-square of 3.9 and 7.2 mg/L, had a higher efficiency than the modified response dose model, and in total has been shown higher performance to mathematical adsorption models.

Keywords

• Adsorption capacity
• Artificial intelligence
• Adsorption Model
• Cationic Modifier
• modified zeolite