Optimization of Stirring Speed and Irganox 1076 Concentration for Enhanced Physical Properties of ABS Using Artificial Neural Networks

This research was undertaken to investigate the efficacy of Irganox 1076 antioxidant in enhancing the impact strength of acrylonitrile butadiene styrene (ABS) polymer. Specifically, the study focused on ABS grade SD-0150, a product of Tabriz Petrochemical Company, which served as the base polymer. Irganox 1076 was strategically incorporated into the polymer matrix, functioning as both an impact modifier and a heat stabilizer, with the aim of improving the material's overall performance. To comprehensively evaluate the antioxidant's influence on the polymer's properties, a series of ABS/Irganox 1076 blends were meticulously prepared. These blends were then subjected to a battery of characterization tests, including the determination of melt flow index (MFI), Vicat softening temperature, and Izod impact strength. All experimental results were rigorously compared against the properties of the pristine ABS-SD0150, establishing a baseline for analysis. Furthermore, a multilayer perceptron (MLP) neural network, a sophisticated machine learning tool, was employed to model the collected experimental data. This computational approach facilitated the prediction of blend properties, offering insights beyond direct experimental measurements. The results of this study conclusively demonstrated that the incorporation of Irganox 1076 led to a significant improvement in the physical properties of ABS. Notably, the blend containing 2% by weight of Irganox 1076 exhibited markedly enhanced thermal stability, a critical factor for many applications. Additionally, the polymer's inherent resistance to environmental stress, was also substantially improved. These findings underscore the potential of Irganox 1076 as an effective additive for enhancing the performance characteristics of ABS polymers.