Pyrolysis characteristics, artificial neural network modeling and environmental impact of coal gangue and biomass by TG-FTIR

The harm done to the environment by coal gangue was very serious, and it is urgent to adopt effective methods to dispose of coal gangue in order to prevent further environmental damage. Co-pyrolysis experiments of coal gangue (CG) and peanut shell (PS) were carried out using thermogravimetry-Fourier transform infrared spectroscopy (TG-FTIR) under nitrogen atmosphere. The heavy metal was detected using the inductively coupled plasma-optical emission spectroscopy (ICP-OES). CG and PS were mixed according to the mass ratio of 1:0, 3:1, 1:1, 1:3 and 0:1. The samples were heated to 1000 degrees C at the heating rate of 10 degrees C/min, 20 degrees C/min and 30 degrees C/min. The comprehensive pyrolysis index (CPI) of CG, C3P1, C1P1, C1P3 and PS is 0.17 x 10(-8), 9.75 x 10(-8), 35.47 x 10(-8), 100.94 x 10(-8) and 192.72 x 10(-8)%(2).min(-2).degrees C-3. The kinetic parameters were calculated by model-free methods (Flynn-Wall-Ozawa and Kissinger-Akahira-Sunose). The gas products generated at different temperatures during the pyrolysis experiment were detected by Fourier transform infrared spectrometer. The heating rate, temperature and mixing ratio are the input parameters of artificial neural network (ANN), and the remaining mass percentage of sample during the pyrolysis is the output parameter. The ANN model was established and used to predict thermogravimetric experimental data. The ANN18 model is the best model for predicting the co-pyrolysis of CG and PS. (C) 2020 Elsevier B.V. All rights reserved.

Automated summary

The co-pyrolysis experiments of coal gangue and peanut shell were conducted using TG-FTIR and ICP-OES, with kinetic parameters calculated by model-free methods and an ANN model established for predicting thermogravimetric experimental data.