Pyrolysis of waste polyvinyl chloride (PVC) is considered a promising and highly efficient treatment method. This work aims to investigate the kinetics, and thermodynamics of the process of PVC pyrolysis. Thermogravimetry of PVC pyrolysis at three heating rates (5, 10, and 20 K/min) showed two reaction stages covering the temperature ranges of 490-675 K, and 675-825 K, respectively. Three integral isoconversional models, namely Flynn-Wall-Qzawa (FWO), Kissinger-Akahira-Sunose (KAS), and Starink, were used to obtain the activation energy (Ea), and pre-exponential factor (A) of the PVC pyrolysis. On the other hand, the Coats-Redfern non-isoconversional model was used to determine the most appropriate solid-state reaction mechanism/s for both stages. Values of Ea, and A, obtained by the isoconversional models, were very close and the average values were, for stage I: Ea = 75 kJ/mol, A = 1.81 × 106 min-1; for stage II: Ea = 140 kJ/mol, A = 4.84 × 109 min-1. In addition, while the recommended mechanism of the first stage reaction was P2, F3 was the most suitable mechanism for the reaction of stage II. The appropriateness of the mechanisms was confirmed by the compensation effect. Thermodynamic study of the process of PVC pyrolysis confirmed that both reactions are endothermic and nonspontaneous with promising production of bioenergy. Furthermore, a highly efficient artificial neural network (ANN) model has been developed to predict the weight left % during the PVC pyrolysis as a function of the temperature and heating rate. The 2-10-10-1 topology with TANSIG-LOGSIG transfer function and feed-forward back-propagation characteristics was used.
Keywords: artificial neural networks (ANN); kinetics; polyvinyl chloride (PVC); pyrolysis; thermodynamics; thermogravimetric analysis (TGA).