Modeling of Methylene Blue Adsorption on the Surfactant-Modified Tea Waste by the Artificial Neural Network

The adsorption process modeling of Methylene Blue dye (MB+) on the modified tea waste (TW) with anionic surfactant (SDS) was performed by the artificial neural network. The FTIR and EDS analyses were used to investigate the presence of SDS molecules on the adsorbent surface and it was determined that the SO3- anions help to MB+ adsorption by the ion exchange mechanism. The adsorption isotherm data were fitted to Langmuir and Freundlich isotherm equations and the Langmuir adsorption capacity (Qmax) was calculated respectively, equal to 124.3 and 156.2 mg/g for TW and STW. It was observed that the adsorption kinetics of MB follows the pseudo-second-order kinetic model. The findings of this investigation suggest that physical adsorption also plays a role in controlling the sorption rate of MB. The feedforward ANN model with 5 input parameters and one hidden layer having 15 neurons was designed to predict the adsorptive behavior of sorbents, and it was observed that the model is able to predict the MB removal percentage for the test data series with great R2>0.97.