Analysis and modeling of multicomponent sorption of heavy metals on chicken feathers using Taguchi's experimental designs and artificial neural networks

In this study, we have used an integrated approach based on Taguchi's experimental designs and artificial neural networks (ANNs) for the analysis and modeling of the simultaneous removal of cadmium (Cd2+), nickel (Ni2+), and lead (Pb2+) ions from ternary aqueous solutions using chicken feathers. Our results indicated that the multicomponent sorption of these heavy metals on chicken feathers is a complex antagonistic process. Specifically, chicken feathers showed a strong preference for the removal of Pb2+ ions in multicomponent solutions and the presence of these ions affected significantly the multicomponent removal of Cd2+ and Ni2+. This antagonistic sorption effect is more significant at pH 5 and the sorption preference of chicken feathers during heavy metal removal depends on the solution pH. Results of X-ray absorption near edge structure suggested that sulfide and carboxylic groups of chicken feathers appear to play a relevant role for the removal of heavy metal ions using this biomass. On the other hand, the desorption process using diluted acidic solutions is effective for the recovery of both Pb2+ and Cd2+ from metal-loaded chicken feathers indicating the feasibility of sorbent regeneration. Finally, ANNs model offers a better performance and more advantages for modeling the sorption of heavy metals in multicomponent solutions than those obtained using Langmuir- and Sips-type multicomponent isotherm equations. This ANNs model is capable of modeling and predicting the sorbent performance at different conditions of pH. In summary, the application of Taguchi's experimental designs and ANNs models is promising for data analysis and modeling of multicomponent pollutant removal for wastewater and water treatment.