Artificial Neural Network (ANN) Method for Modeling of Sunset Yellow Dye Adsorption Using Nickel Sulfide Nanoparticle Loaded on Activated Carbon: Kinetic and Isotherm Study

Abstract

In this paper, the NiS nanoparticles are prepared and characterized by x-ray powder diffraction and scanning electron microscopy. The NiS nanoparticles showed the excellent adsorption properties toward sunset yellow (UA) dye. The effect of solution pH, adsorbent dosage (0.005–0.020 g), contact time (0.5–30 minutes), and initial UA concentration (5–40 mg L−1) on the extent of adsorption was investigated and modeled by artificial neural network. The experimental equilibrium data was analyzed by Langmuir, Freundlich, Tempkin, and D–R isothermal models. It was seen that the data was well presented by Langmuir model with a maximum adsorption capacity of 333.3 mg g−1 at 26°C. Kinetic studies at various adsorbent dosages and initial UA concentrations show that high removal percentage (>90%) was achieved within 15 minutes. The adsorption of UA follows the pseudo-second-order rate model. The experimental data were applied to train the multilayer feed-forward neural network with three inputs and one output with Levenberg–Marquart algorithm and different numbers of neurons in the hidden layer. The minimum mean square error of 0.0003 and determination coefficient of (R2) 0.99 were found.