Análise da co-pirólise do pseudocaule de bananeira e resíduos plásticos por termogravimetria: caracterização cinética e modelagem por redes neurais artificiais
Nenhuma Miniatura disponível
Data
2023-09-27
Autores
Nepomucena, Thâmara Vieira
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Federal do Espírito Santo
Resumo
The high oxygen content in bio-oil obtained from the pyrolysis of lignocellulosic biomass may make its use as a fuel unviable. One strategy to mitigate this issue is deoxygenation through co-pyrolysis of biomass with plastic waste. Thus, the general objective of this work was to analyze the co-pyrolysis of banana pseudostem and plastic waste by thermogravimetry. Initially, the results of the chemical characterization indicated polypropylene as the most suitable plastic exhaust for bio-oil improvement, compared to polyethylene, as it contains greater amounts of hydrogen and the absence of oxygen. Subsequently, all parameters used to characterize co-pyrolysis, through thermogravimetric analysis, increased as the proportion of plastics in the mixture increased. This suggests that the initial temperature, maximum mass loss rate, peak temperature, final residual mass, and pyrolysis performance are specific primarily to plastics. Even so, artificial neural networks showed excellent prediction capacity for thermogravimetry data, with brightness coefficients greater than 0.999. The activation energies estimated with the data predicted by the neural networks were very close to those discovered experimentally. Finally, as the main contribution made to the topic of this research and future application of this process, the best conditions presented for the reports: a mixture containing 25% polyethylene mass and 75% banana pseudostem mass, using a heating rate of 30 K/min, and another mixture with 50% polypropylene mass and 50% banana pseudostem mass, using a heating rate of 10 K/min.
Descrição
Palavras-chave
Energia renovável , Gestão de resíduos sólidos , Polietileno tereftalato , Polipropileno , Modelo cinético , Perceptron multicamadas