Enhancement of biofuel quality via conventional and catalytic co-pyrolysis of bamboo with polystyrene in a bubbling fluidized bed reactor: Coupled experiments and artificial neural network modeling

Abstract

Experiments and artificial neural network modeling were performed to investigate the effect of operating parameters (temperature, fluidization velocity, and particle size) and catalysts (HZSM-5, red mud, Fe2O3, and dolomite) on co-pyrolysis of bamboo with polystyrene (PS) in a fluidized bed reactor for upgrading bio-oil. The synergistic effect was revealed by analyzing products via various analytical techniques and differences between theoretical and actual co-pyrolysis results. Under the H2-donor source from PS, co-pyrolysis reduced the O content while enhancing the content of aromatic hydrocarbons and higher heating value (HHV) of oil. Depending on the type of catalyst, characteristics and yield of the co-pyrolysis oil were affected along with the proposed reaction pathways. Dolomite was assessed as the most effective catalyst for improving oil quality, with the highest HHV (34.1 MJ/kg) and highest pH value (5.0). An artificial neural network using a back propagation algorithm in Matlab software was applied to predict the liquid yield and HHV of oil. The ANN15 model (15 neurons in the hidden layer) was found to be the best model in validating experimental data with mean deviations of 1.75 % for liquid yield and 0.87 % for HHV of oil.