Towards modeling the generation of the transformation products of carbamazepine in polishing and oxidation water treatment processes

Abstract

Carbamazepine (CBZ) is an anticonvulsant, one of the most commonly detected in surface water, even in areas with low anthropogenic pressure. The CBZ contained in wastewater is poorly removed by conventional secondary processes and requires a polishing stage; oxidation processes in this case are a tried-and-tested options. However, a multitude of reaction pathways occurs and may lead to the generation of many transformation products (TPs) with different physicochemical properties. This work presents an ordinary differential equation (ODE)-based computer model predicting the generation of TPs generated from CBZ oxidation. The aim is to improve our knowledge of the nature, fate and dynamics of such TPs. After calibration using individual datasets taken from the literature, we proposed a global model able to predict the fate of 11 TPs. A sensitivity analysis revealed that 4 primary TPs are instrumental in the model quality: EP-CBZ, 9CHO-CBZ, 2OH-CBZ and 10OH-CBZ. TPs far from CBZ in the reaction pathway are less accurately predicted. Furthermore, the work showed that determining stoichiometric parameters (k_stoi) is essential to predict TPs as they exhibit greater sensitivity to them than kinetic constants (k).