Thermo-kinetics, thermodynamics, and ANN modeling of the pyrolytic behaviours of Corn Cob, Husk, Leaf, and Stalk using thermogravimetric analysis

Abstract

Abstract In this study, we investigate the thermal stabilities, thermo-kinetic, and thermodynamic behaviours of Corn Cob (CC), Husk (CH), Leaf (CL), and Stalk (CS) during pyrolysis using the Thermogravimetric Analysis (TGA) at a single heating rate of 10 °C/min. Thermo-kinetics and thermodynamic parameters were evaluated for two temperature regions, region I (100–350 °C) and region II (350–500 °C) by employing the Coats–Redfern (CR) integral method to fit the TGA data to sixteen kinetic models. Results showed that diffusion models (D1, D1, D3, and D1) best suited the decomposition of CC, CH, CL, and CS in region I with Ea values of 109.90, 186.01, 129.4, and 78.7 kJ/mol respectively. Similarly, D1, third order model (F3), D3, and nucleation model (P4) with Ea values of 68.50 (CC), 177.10 (CH), 62.10 (CL), and 127.70 (CS) kJ/mol respectively best described residues’ decomposition in region II. Furthermore, kinetic parameters were used to compute the thermodynamic parameters; change in enthalpy (∆H), Gibbs free energy (∆G), and change in entropy (∆S) values for both regions. To study the pyrolytic behaviours of the residues, Artificial Neural Network (ANN) was employed to develop models to predict weight losses in samples by determining the coefficient of determination (R 2) and minimum Mean Square Error (MSE). Results showed ANN as a very important tool for predicting the pyrolytic behaviours of corn residues and other biomass samples.