Catalytic pyrolysis of sewage sludge with HZSM5 and sludge-derived activated char: A comparative study using TGA-MS and artificial neural networks

The catalytic effects of HZSM5 and sludge-derived activated char (AC) on thermal decomposition of sewage sludge (SS) were invested in the present work. The outcomes including thermal behaviour, kinetic study, and evolution of pyrolysis gases during non-catalytic and catalytic pyrolysis were comparatively examined. The results revealed that the catalytic and non-catalytic pyrolysis involves four stages, and the primary decomposition reaction occurred in the range of 200-600 degrees C. Coats-Redfern method has been adapted in the second and third devolatilization stages in order to estimate the reaction order and activation energy from the mass loss data. The average activation energies for the reaction orders n = 1, 2, and 3 are in the range of 6.67-20.37 kJ.mol(-1) for SS; 1.51-6.87 kJ.mol(-1) for HZSM5; and 2.29-9.17 kJ.mol(-1) for AC, respectively. Moreover, we found that both HZSM5 and sludge-derived AC promote SS pyrolysis's reaction rate by abridging the average activation energy. Interestingly, both catalysts effectively reduce the evolved gaseous generated during the SS pyrolysis (eg., CH4, CO2, HCN, NO2, SO2 and CH3SH). Finally, an artificial neural network (ANN) model was developed to simulate the kinetics of both catalytic and non-catalytic pyrolysis based on the thermal decomposition data. The findings showed an excellent agreement between the experimental and predicted values (R-2 > 0.99). Thus, sludge-derived AC is an efficient catalyst for SS pyrolysis and can be considered cost-effective compared to HZSM5.

Automated summary

The study showed that both HZSM5 and sludge-derived activated char are efficient catalysts for promoting sewage sludge pyrolysis reactions, reducing average activation energy, and effectively decreasing the generated gases. An artificial neural network model successfully simulated the kinetics of pyrolysis with high agreement between experimental and predicted values.