Kinetic modeling and thermogravimetric investigation of Phoenix dactylifera and Phyllanthus emblica non-edible oil seeds: artificial neural network (ANN) prediction modeling

In the present study, Phoenix dactylifera (PD) or Indian Amla and Phyllanthus emblica (PE) or Dry dates were thermally studied to determine their potential as pyrolysis feedstocks using theoretical kinetic modeling, then the artificial neural network (ANN) was used on TGA data to predict the best fitting. TG experiments were conducted at three different heating rates of 10, 20, and 30 degrees C/min for both the seeds. Kinetic modeling of the PD and PE seeds was also performed using FWO, KAS, FRM, KN model-free isoconversional methods and C-R model fitting method. Further, the thermodynamic parameters were also determined for the aforementioned heating rates. The average values of activation energy were found to be in incremental pattern of KN<FRM<C-R<KAS<FWO methods and KN<KAS<FWO<C-R<FRM methods, respectively for PD and PE seeds. The values of G, S, and H were observed in the similar range as other oilseeds utilized in pyrolysis process. The TGA data were accurately simulated by ANN modeling at three different heating rates as the value of R-2 were found in the range of 0.99-1 for both the seeds. The kinetic modeling, thermodynamic parameters, and ANN modeling also proved that the PD and PE waste biomass seeds could definitely be used to obtain liquid fuels or other valuable chemicals besides reducing the environmental degradation and value-addition to waste seeds towards a circular economy.