Phenol adsorption onto olive pomace activated carbon: modelling and optimization

Abstract

Purpose. To depollute water contaminated by phenol using the adsorption process in a batch reactor with valorization of olive pomace for the preparation of activated carbon to be used as an adsorbent. Methodology. In this present work, study on the influence of four principal parameters on the adsorption yield during the treatment of polluted water by the process of adsorption were studied, namely: the activated carbon mass, the contact time, the phenol initial concentration and the stirring speed. In addition, the process was optimized with full factorial designs using the response surface methodology by the MINITAB software. Findings. The phenol removal by adsorption on the activated carbon made from olive pomace makes it possible to achieve an adsorption efficiency of 91 % with the use of following optimal conditions: adsorbent mass of 0.48 g, a contact time of 110.80 min, a phenol concentration of 100.98 mg/L, and a stirring speed of 462.89 rpm. Contact time and adsorbent mass showed a positive effect on phenol removal efficiency. The principal effect results illustrate that all four examined factors significantly affected the phenol removal by olive pomace activated carbon with a confidence level of 95 %. Originality. The experimental data of the phenol adsorption on the olive pomace activated carbon were examined by adjustment of a second-degree polynomial model. This model is validated by a statistical method using an analysis of variance (­ANOVA). Numerical optimization was performed by the desirability function to identify the optimal parameters for maximum phenol recovery. Practical value. In order to highlight a process for treating waters contaminated with phenol, we have chosen the processes that are considered to be best suited, which is adsorption with the recovery of a waste such as an adsorbent, which is prepared by physical and chemical activation of olive pomace. The full factorial design applied makes it possible to show the influence of each parameter independently and their dependencies, as well as to find the optimal experimental conditions quickly which lead to the realization of this process.