Thermochemical Advanced Oxidation Process by DiCTT for the Degradation/Mineralization of Effluents Phenolics with Optimization using Response Surface Methodology and Artificial Neural Networks Modelling

Abstract

The actual work evaluated the effect of initial phenol concentration (CPh0) of 500, 1000 and 1500 mg.L-1, the molar stoichiometric ratio of Phenol/Hydrogen peroxide (RP/H) of 25, 50 and 75 % and time (t) of 30, 90 and 150 min on the oxidation of phenolic effluents by called Direct Contact Thermal Treatment (DiCTT). This process provides a novel means to induce degradation and mineralization of organic pollutants in water. The experimental studies were carried out at semi-industrial plant. The organic pollutant was degraded with a conversion higher than 99% and a Total Organic Carbon (TOC) mineralization exceeding 40%, to a (RP/H) of 75%, independent of the CPh0, that was identified as the optimal condition by thermochemical process. The initial phenol concentration was quantified and identified by the High Performance Liquid Chromatography (HPLC) technique followed by statistical design tools to optimization using Response Surface Methodology (RSM) and an analytical mathematical modelling via Artificial Neural Networks (ANNs). The results also showed the dynamic concentration evolution of the intermediates formed (catechol, hydroquinone and para-benzoquinone). Artificial Neural Networks were applied to model the step experimental of Phenol Degradation (PD) and Total Organic Carbon (TOC) conversion by DiCTT thermochemical process. For the ANN modelling, “statistic 8.0” software was used with a Multi-Layer Perceptron (MLP) feed-forward networks by input-output data using a back-propagation algorithm. The correlation coefficients R2 between the network predictions and the experimental results were in the range of 0.95–0.99.