ANN Optimization of Adsorption of Naphthalene on Composite Nanoparticles of Chitosan-CTAB-Sodium Bentonite Clay

Abstract

In the present study, nanoparticles of chitosan-cetyltrimethylammonium bromide (CTAB)-sodium bentonite clay were synthesized and characterized using EDX, SEM, FTIR, XRF and XRD techniques. The composite material was utilized as adsorbent for the treatment of contaminated aqueous solution containing naphthalene. The adsorption process was modeled and optimized using artificial neural network (ANN) and ANN–genetic algorithm respectively. The process variables considered were surfactant concentration, X1 , activation time, X 2 , activation temperature, X3 , and chitosan dosage, X4 . The predicted ANN models for % removal of naphthalene and adsorption capacity of the composite adsorbent fitted excellently the experimental adsorption data of naphthalene judging from high value of coefficient of determination, 2 R , amongst others and very low values of error functions. The optimum conditions obtained with ANN–GA were X1 = 70.7580 mg/L, X 2 = 2.9940 h, X3 = 99.9880o C, and X 4 = 2.0340 g. The predicted response variables of 99.1461% removal of naphthalene and 249.67 mg/g adsorption capacity of the composite adsorbent using the ANN-GA models were in excellent agreement with their corresponding experimental values of 99.35% and 250.16 mg/g with % errors of 0.2056 and 0.1960 respectively. Consequently, the ANN models and the ANN–GA optimized conditions can be reliably applied to the experimental adsorption data of naphthalene on the chitosan–CTAB–sodium bentonite clay composite nanoparticles as adsorbent. Moreover, the prepared adsorbent in this study is a viable alternative adsorbent for the treatment of industrial wastewater containing polycyclic aromatic compounds, especially naphthalene.