Characterization of polyethylene terephthalate wastes/Acrylonitril-Butadiene styrene (PETW/ABS) composites with applications of artificial neural networks

Abstract

During the processes done in petrochemical plants, high amount of polyethylene terephthalate (PET) as PET Waste remains. In this research in order to reuse PET and reduce the environmental problems, PET Waste in different weight percentages (0, 5, 10, 15) has been combined with ABS in different conditions of temperature, time and speed of extruder. Different tests have been performed on polymer mixture like: Impact Test, Elongation at Break, Modular Melt Flow, vicat softening temperature and Yellow Index. The results declared that in condition with constant temperature and time and speed of mixing with increasing PET Waste in ABS the number of Impact Test and Elongation at Break decreases and the number of Modular Melt Flow, vicat softening temperature and Yellow Index increases.When the percentage of PET Waste in ABS is constant, according to the change in each of the conditions of the mixing, it is possible that the amount of the change for different tests to be investigated and the laboratory data to be simulated by the MLP artificial neural networks. The results of the simulation covered the laboratory data perfectly and the information of this article can be used as a reference in order to improve the specifications of the mixed polymer according to the need of increasing polymer properties.