Preparation of a new adsorbent for the removal of arsenic and its simulation with artificial neural network-based adsorption models. Issue 4 (August 2020)

• Record Type: Journal Article
• Authors: Rodríguez-Romero, J.A.; Mendoza-Castillo, D.I.; Reynel-Ávila, H.E.; de Haro-Del Rio, D.A.; González-Rodríguez, L.M.; Bonilla-Petriciolet, A.; Duran-Valle, C.J.; Camacho-Aguilar, K.I.
• Is Part Of: Journal of environmental chemical engineering. Volume 8:Issue 4(2020)
• Publication Date: 2020-08
• Subjects: Arsenic -- Artificial neural network -- Adsorption modeling -- Water treatment -- Opuntia ficus indica; Chemical engineering -- Environmental aspects -- Periodicals; Environmental engineering -- Periodicals; Chemical engineering -- Environmental aspects; Environmental engineering; Periodicals; 660.0286
• DOI: 10.1016/j.jece.2020.103928
• Languages: English
• ISSNs: 2213-2929

Abstract

Graphical abstract: Highlights: A mexican feedstock was used to prepare an adsorbent for arsenic removal. Hybrid models to simulate the arsenic adsorption were developed. Modeling results of adsorption kinetics and isotherms improved by using artificial neural networks. Arsenic adsorption can be performed with an adsorbent obtained from nopal biomass. Abstract: The preparation of an alternative material for the adsorption of arsenic from aqueous solution was studied. This adsorbent was obtained from the pyrolysis and ZnCl2 activation of Opuntia ficus indica biomass (widely known as nopal), which is a typical plant of the Mexican landscape. Preparation conditions of this adsorbent were improved to increase its arsenic adsorption properties. Experimental kinetic and isotherm data for the arsenic removal with the best adsorbent were quantified to analyze its performance. A detailed physicochemical characterization of this adsorbent was carried out to obtain insights about the arsenic adsorption mechanism. A set of new isotherm and kinetic equations were also developed for modeling the arsenic adsorption. These novel models were obtained from the hybridization of the traditional adsorption equations with an artificial neural network. The artificial neural network was used to improve the performance of the conventional kinetic and isotherm equations for the simulation of arsenic removal at different conditions of pH and temperature. Performance of these models was assessed using the arsenic adsorption experimental data obtained with tested adsorbent. Results showed that hybrid models outperformed the well-known kinetic and isotherm adsorption equations commonly used in water treatment allowing better calculations for process design. These models can be extended for the study and analysis of the adsorption of a variety of water pollutants.