Highly efficient simultaneous ultrasonic assisted adsorption of brilliant green and eosin B onto ZnS nanoparticles loaded activated carbon: Artificial neural network modeling and central composite design optimization

Abstract

In this work, central composite design (CCD) combined with response surface methodology (RSM) and desirability function approach (DFA) gives useful information about operational condition and also to obtain useful information about interaction and main effect of variables concerned to simultaneous ultrasound-assisted removal of brilliant green (BG) and eosin B (EB) by zinc sulfide nanoparticles loaded on activated carbon (ZnS-NPs-AC). Spectra overlap between BG and EB dyes was extensively reduced and/or omitted by derivative spectrophotometric method, while multi-layer artificial neural network (ML-ANN) model learned with Levenberg–Marquardt (LM) algorithm was used for building up a predictive model and prediction of the BG and EB removal. The ANN efficiently was able to forecast the simultaneous BG and EB removal that was confirmed by reasonable numerical value i.e. MSE of 0.0021 and R2 of 0.9589 and MSE of 0.0022 and R2 of 0.9455 for testing data set, respectively. The results reveal acceptable agreement among experimental data and ANN predicted results. Langmuir as the best model for fitting experimental data relevant to BG and EB removal indicates high, economic and profitable adsorption capacity (258.7 and 222.2mgg−1) that supports and confirms its applicability for wastewater treatment.