Exploring sustainable fuel production through thermal behavior analysis using TGA and artificial neural network in the co-pyrolysis of polystyrene and coconut sawmill residue

Abstract

This study employs Thermogravimetric Analysis (TGA) to explore co-pyrolysis potential using polystyrene (PS) and coconut sawmill residue (CSR) for liquid fuel production. Two distinct degradation stages are observed in CSR-PS blends, mirroring pure CSR samples: the initial phase (200-400°C) decomposes biomass components, while the second stage (400-550°C) targets the synthetic polymer PS within CSR-PS blends. Analyzing thermal degradation parameters reveals insights. 100% PS exhibits the highest weight loss and activation energy, highlighting PS's formidable decomposition. Conversely, 100% CSR shows the lowest weight loss and activation energy due to its organic composition. Artificial Neural Network (ANN) modeling indicates varying correlation accuracies for different blend compositions. Surprisingly, 100% PS exhibits lower correlation accuracy in predicting weight loss compared to the 80% PS blend, which achieves a perfect correlation. Conversely, 100% CSR, with simpler decomposition, has the lowest correlation accuracy. These findings illuminate the complex thermal behavior of CSR-PS blends, emphasizing the distinct degradation characteristics of PS and CSR. Implications extend to material applications and disposal strategies, emphasizing tailored approaches based on blend compositions and thermal profiles. This research advances co-pyrolysis as a sustainable avenue for liquid fuel production, providing insights for future research and practical applications.