Appraisal of wastewater-grown biomass of Leptochloa fusca to produce bioenergy and biochemicals through pyrolysis, TG-FTIR-GCMS, and machine learning tools

Muhammad Ishfaq,Azeem Asghar,Imtiaz Ali,Aqib Zafar Khan,Ayesha Shahid,Ning Wang,Hui Zhu,Hesham R. El-Seedi,Muhammad Aamer Mehmood,Chen-Guang Liu

DOI: https://doi.org/10.1016/j.enconman.2024.118229

IF: 10.4

2024-03-01

Energy Conversion and Management

Abstract:Mass scale production of lignocellulosic biomass to produce clean energy needs huge amounts of freshwater and land which conflicts with their need for food and agricultural crops. Here, a land-free cultivation method was employed to produce Leptochloa fusca biomass using industrial and urban wastewater, which was then converted into bioenergy through pyrolysis for the first time. Both biomasses were thermally degraded in a thermogravimetric analyzer (TGA) at 10 − 50℃/min of heating rates under nitrogen inert atmosphere. Major thermal degradation of both biomasses occurred at 200 °C − 450 °C in the active pyrolysis zone at conversion (α) 0.275 to 0.700. Urban wastewater (UWW) biomass produced 19.27 ± 0.09 MJ/kg while industrial wastewater (IWW) biomass produced 18.92 ± 0.17 MJ/kg of heating energy. Similarly, activation energies calculated through FWO, Friedman, and KAS methods were 169.22 kJ/mol, 165.14 kJ/mol, and 167.82 kJ/mol for UWW-grown biomass and 181.15 kJ/mol, 182.94 kJ/mol, and 180.40 kJ/mol for IWW-grown biomass respectively, demonstrating the appropriateness of both biomasses for pyrolysis. Smaller difference (<4 kJ/mol) between the activation energy (Ea) and enthalpy change (ΔH) indicated favorable reaction thermodynamics. TG-FTIR-GCMS analysis indicated the presence of hydrocarbons, alcohols, organic acids, ethers, esters, ketones, aldehydes, amines, and amides in produced pyrolytic gases. UWW cultivation had a more pronounced effect on improving energy efficiency and value of the biomasses during pyrolysis. Later, the ANN model data were shown to be aligned with TG-based pyrolysis interpretation, analyzed by MLP-based ANN regression model accurately determined activation energies. R2 values (0.99) indicated the reliability of training, testing, and validation data. M−DAEM model also had good fit (R2  > 0.99) with experimental data. Conclusively, adoption of land-free cultivation system emerges as a promising strategy to generate renewable biomass with optimized pyrolysis potential, where pollution gets cleaned up and resources get recycled offering a win–win scenario for energy-water-environment nexus sustainability.

energy & fuels,thermodynamics,mechanics