Rubens Teles Monteiro
◽
Rayany Magali da Rocha Santana
◽
Ana Maria Ribeiro Bastos da Silva
◽
Alex Leandro Andrade de Lucena
◽
Léa Elias Mendes Carneiro Zaidan
◽
...
The growth of pollution in aquatic environments increases every day, causing compounds like pharmaceuticas to be detected in surface waters. Thus, tecniques such as advanced oxidation processes (AOP) have been used to degrade this compounds. In this work, the efficiency of AOP in the degradation of nimesulide and ibuprofen pharmaceuticals was evaluated through chromatographic analysis as well as organic matter through the levels of chemical oxygen demand (COD) and total organic carbon (TOC). It was verified that the photo-Fenton process presented the bests results, degrading 89.70% of nimesulide and 93.35% of ibuprofen. This same process managed to reduce COD by 91.60% and mineralize 90.04% of the TOC. The kinetic study showed a good linear fit (R2=0.993) for the clustered kinetic model, as well as a good fit to the mathematical model of artificial neural networks (ANNs), with a value of R2=1.000 for the MLP4-4-1 BFGS 4567 model. Finally, the toxicity of the solution after treatment was verified against the seeds of Lactuta sativa, Cichorium endívia, Ocimum basilicum and American Hard grain. It was found that the seeds that received the solution before treatment had a lower germination amount than the ones where the post AOP treatment solution was added. Then, the root growth was evaluated, in which a relative toxic effect was observed.
Prediction of Chemical Oxygen Demand from The Chemical Composition of Wastewater by Artificial Neural Networks