Centrifugally spun magnetic nanofibers for anionic dye removal: Optimization and predictive modeling with RSM and ANN

Abstract

Response surface methodology (RSM) and artificial neural network (ANN) approach were used for the modeling-optimization of dye removal by adsorption using poly(ethylene terephthalate)/Fe₃O₄ nanocomposites nanofibers. The nanoparticles of Fe₃O₄ were synthesized by the chemical co-precipitation method in the presence of linoleic acid as a stabilizer. The polymer matrix was obtained by recycling PET from plastic water bottles. We used a simple homemade centrifugal spinning device to prepare PET/Fe₃O₄ nanofibers and the resulting material was characterized by XRD, FTIR and SEM. The performance of the nanofiber material as an adsorbent was examined in water treatment to remove Naphthol Green B anionic dye in batch mode. Box–Behnken-RSM and ANN were employed to optimize the effect of pH, contact time, initial dye concentration, and adsorbent mass. The optimized conditions leading to the maximum predicted dye removal were a pH of 2, a contact time of 60 min, an adsorbent mass of 0.125 g/L, and a dye concentration of 20 mg/L, and. The results demonstrated that both methods showed good predictions of dye removal. The pseudo-first-order kinetic and Langmuir isotherm models were found to be best fitted with the experimental data and rePET/Fe3O4 exhibited a maximum NGB dye adsorption capacity up to 155 mg/g. This study also indicates that magnetite nanoparticles combined with recycled PET in the form of nanocomposite nanofibers could be used for the treatment of dye wastewater.

Graphical abstract

Keywords:

• PET/Fe₃O₄ composite nanofibers
• anionic dye
• adsorption
• optimization
• response surface methodology
• artificial neural networks

Acknowledgments

The authors would like to thank Professor Tahar Abass Miloud for his help in the XRD analysis of the samples, and the engineers Mr. Zerrokate Khaled, and Mr. Kermal Ibrahim for their assistance during fieldwork.

Disclosure Statement

No potential conflict of interest was reported by the author(s).