Experimental Investigation and Proposal of Artificial Neural Network Models of Lead and Cadmium Heavy Metal Ion Removal from Water Using Porous Nanomaterials

Abstract

This study used porous nanomaterials MCM-41 and SBA-15, as well as their modified species, to remove lead and cadmium ions from water. We used X-ray diffraction (XRD), a scanning electron microscope (SEM), the Brunauer–Emmett–Teller (BET), and the Fourier transform infrared (FT-IR) method to investigate the characteristics of porous nanomaterials. Additionally, atomic absorption spectroscopy (AAS) measured the concentration of lead and cadmium ions. The stratigraphic analysis showed the samples’ isothermal shape to be type IV. This study investigated the amount, absorbent, pH changes, and adsorption time parameters. We observed that the adsorption efficiency of lead by the synthesized samples was higher than that of the adsorption of cadmium. Mesoporous structures also displayed increased adsorption efficiency due to the amino group. Four testing stages were conducted to determine the reproducibility of the adsorption by the synthesized samples, with the results showing no significant changes. As a result of the adsorption process, the structure of the recycled sample NH2-MCM-41 was preserved. We also used artificial neural networks (ANN) to propose predictive models based on the experimental results. The ANN models were very accurate, such that the mean absolute error (MAE) was less than 2% and the R2 was higher than 0.98.