Artificial neural network modeling of the influence of sol-gel synthesis variables on the photocatalytic activity of TiO2 nanoparticles in the removal of Acid Red 27

作者：Behnajady Mohammad A^*; Eskandarloo Hamed; Eskandarloo Farzaneh

来源：Research on Chemical Intermediates, 2015, 41(9): 6463-6476.

DOI：10.1007/s11164-014-1753-z

• TiO2 nanoparticles
• Photocatalytic activity
• Sol-gel method
• Synthesis variables
• Artificial neural network

In this study, titanium dioxide (TiO2) nanoparticles were prepared by the sol-gel method in different synthesis conditions. The effect of synthesis variables (including water: titanium alkoxide molar ratio, reflux temperature, reflux time, gelation pH, and stirring speed) were studied in the removal of Acid Red 27 as a model contaminant from textile industry under UV light irradiation. For the first time, we report modeling of the effects of synthesis variables on the photocatalytic activity of TiO2 nanoparticles by an artificial neural network (ANN). Five effective synthesis variables were inserted as the input of the network and reaction rate constants (kap) were introduced as the output of the network. The results showed that the predicted data from the designed ANN model were in good agreement with the experimental data, with a correlation coefficient (R-2) of 0.9655 and mean square error of 0.00148. The designed artificial neural network provided a reliable method for modeling the photocatalytic activity of TiO2 nanoparticles prepared under different synthesis conditions. Furthermore, the relative importance of each synthesis variable was calculated based on the connection weights of ANN model. The reflux time and reflux temperature were the most significant variables in the photocatalytic activity of TiO2 nanoparticles, followed by the water: titanium alkoxide molar ratio and gelation pH.

• 出版日期2015-9