Loading...

Investigation of 2-nitrophenol solar degradation in the simultaneous presence of K2S2O8 and H2O2: Using experimental design and artificial neural network

Bararpour, S. T ; Sharif University of Technology | 2018

767 Viewed
  1. Type of Document: Article
  2. DOI: 10.1016/j.jclepro.2017.11.191
  3. Publisher: Elsevier Ltd , 2018
  4. Abstract:
  5. In this study, the interaction effect of K2S2O8 (PDS) and H2O2 (as two powerful oxidants) was investigated on the solar degradation of 2-nitrophenol (2-NF) in the two systems, i.e. absence and presence of Ag/S/TiO2 photocatalyst. Experiments were designed based on the central composite design, and two methods of response surface methodology (RSM) and artificial neural network (ANN) were developed for modeling of the systems. Concentrations of PDS and H2O2 were considered as independent variables and 2-NF degradation efficiency was selected as the response. It was revealed that the predictive capacity of ANN model is more than RSM model according to their corresponding R2, R2 adj RMS, MAE, and AAD values. Therefore, ANN model was utilized to analyze the effects of the independent variables. Moreover, it was found that, by increasing the amount of either PDS or H2O2, less amount of the other oxidant was required to reach the highest degradation efficiency which shows the contributive effect of the oxidants on each other. By adding Ag/S/TiO2 visible-light-sensitive photocatalyst to the solution containing PDS and H2O2, a significant enhancement in the 2-NF degradation efficiency was observed. Using an ANN-genetic algorithm (ANN-GA) approach, the values of 172.1 mg/l and 80.9 mg/l were respectively obtained as optimum concentrations for PDS and H2O2 (in the presence of the photocatalyst). Under this condition, 2-NF degradation efficiency was predicted to be 96% after only 45 min of solar light irradiation that is in a good agreement with the actual degradation efficiency of 97.1% obtained at the optimum concentrations. In addition, the relative importance of the independent variables was studied using Garson method and it was found that PDS had more impact on the response in comparison with H2O2. © 2017 Elsevier Ltd
  6. Keywords:
  7. Artificial neural network ; Oxidants ; Photocatalytic degradation ; Degradation ; Efficiency ; Genetic algorithms ; Light ; Neural networks ; Optimization ; Silver ; Central composite designs ; Degradation efficiency ; Independent variables ; Optimum concentration ; Photo catalytic degradation ; Predictive capacity ; Response surface methodology ; Solar light irradiation ; Design of experiments
  8. Source: Journal of Cleaner Production ; Volume 176 , 2018 , Pages 1154-1162 ; 09596526 (ISSN)
  9. URL: https://www.sciencedirect.com/science/article/pii/S0959652617328640