Feasibility of biogas and oxy-fuel combustion in steam cracking furnaces: Experimental and computational study

• Published in: Fuel (Guildford) Vol. 304; p. 121393
• Main Authors: Zhang, Yu, Vangaever, Stijn, Theis, Gilles, Henneke, Mike, Heynderickx, Geraldine J., Van Geem, Kevin M.
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ltd 15.11.2021; Elsevier BV
• Subjects: Biogas; Carbon dioxide; Combustion; Computational fluid dynamics; Computer applications; Dilution; Experimental data; Experiments; Feasibility studies; Fuel combustion; Furnaces; Heat flux; Heat transfer; Industrial scale; Mathematical models; Natural gas; Natural gas industry; Numerical models; Oxy-fuel; Oxy-fuel combustion; Radiative heat transfer; Retrofitting; Simulation; Steady state models; Steam; Three dimensional models; Biogas; Industrial scale; Computational fluid dynamics; Oxy-fuel combustion; Radiative heat transfer
• ISSN: 0016-2361; 1873-7153
• DOI: 10.1016/j.fuel.2021.121393

•Biogas and oxy-fuel combustion in steam cracking furnaces has been demonstrated.•Experimental studies are conducted in the industrial-scale JZHC test furnace.•CFD models for the JZHC test furnace simulation are developed and refined.•A modified EDC model is used for the oxy-fuel combustion scenario.•Biogas and oxy-fuel combustion yield similar heat flux as in traditional furnaces. This work evaluates the feasibility of biogas air-fuel combustion and natural gas oxy-fuel in steam cracking furnaces. Four cases, namely air-fuel combustion of pure natural gas, 20% CO2, 40% CO2 diluted natural gas, and oxy-fuel combustion of natural gas are investigated both experimentally and numerically. The John Zink Hamworthy Combustion test furnace, representing a section of a steam cracking furnace, is used for experimental studies. A three-dimensional steady-state CFD model is also developed to simulate the test furnace. The simulation results of the air-fuel combustion scenarios are in good agreement with the experimental data, with the maximum and average relative errors of furnace temperature of 3.86% and 1.78%, respectively. The reduction of flame length with increasing CO2 mole fraction in the fuel is observed in both experiments and simulations. It is shown that CO2 dilution has minor effect on the overall heat flux profile, which is beneficial for retrofitting existing furnaces. On the other hand, the oxy-fuel combustion simulation using default EDC model predicts a significant flame lift-off and incident radiative heat flux shift towards the higher elevations which was not observed in the experiments. This can be mainly attributed to the reduced reaction rate in a CO2 and H2O enriched combustion environment. Adjusting the EDC model parameters helps to achieve better agreement between simulation results and experimental data, while additional lab-scale experiments are essential for further validation of the numerical model. Moreover, it is of particular interest to study the optimal mole fraction of O2 in oxy-fuel combustion scenario.