Adaptive neuro-fuzzy interference system modelling for chlorpyrifos removal with walnut shell biochar

Abstract

Accumulation of chlorpyrifos (CP), a pesticide, causes a significant environmental problem in food, surface/ground waters further to human health. The removal of the CP pollutant in surface/wastewater could be achieved by biochar due to the improved physical and chemical properties. In this work, the CP removal capacities of biochar samples derived from walnut shells at various temperatures from 450 to 900 °C were investigated. The experiments were performed as laboratory batch type study and the adsorption efficiency was determined at various conditions such as adsorbent dosage (10–500 mg/L), sorbate concentrations (100–1500 µg/L), contact time (0–300 min), initial pH (3–10), and the number of recycle. By subtracting the pyrolysis temperature from 450 °C to 900 °C, the surface areas were found to increase from 12.9 m 2 /g to 353.3 m 2 /g, respectively. The 143 experimental data were evaluated by a pair of kinetics and isotherm models and the Adaptive Neural Fuzzy Inference System (ANFIS). The developed ANFIS model was 98.56% successful in predicting the CP removal efficiency depending on the adsorption conditions. Walnut Shell Biochar (WSBC) can be applied for CP adsorption with 86.64% removal efficiency under optimum adsorption conditions (adsorbent = 250 µg/L, sorbate = 1000 µg/L, pH = 7.07 and contact time 15 min) thanks to its improved porosity. It was determined that the biochar samples could be reused 5 times. Equilibrium adsorption was observed to conform to the Langmuir isotherm, and the maximum adsorption capacity for WSBC@900 was 3.536 mg/g.