Key decisions in developing a model of a steam methane reformer were investigated in this paper. A comprehensive one-dimensional Python model of a reformer tube is proposed, including the effects of sintering and poisoning. Comparison between the Redlich−Kwong and ideal gas equations of state shows little difference upon reformed gas outlet properties. Different reaction kinetic models are tested over a variety of realistic conditions and compared against industrial data. Overall, these changes only impact conditions near the reactor inlet, as the approach to thermodynamic equilibrium near the reactor outlet ensures that all model-predicted process variables at the outlet are similar. Models from both Xu and Froment and Hou and Hughes are recommended for future reformer models where accurate outlet variables are required.

Results from a multiscale one-dimensional model of a steam methane reformer were applied to investigate carbon formation. Using a combined thermodynamic and kinetic approach, regions of likely carbon formation are predicted in line with industrial observations. This prediction is used to show that real gas considerations improve predictions of carbon formation regions. Kinetic reaction models were compared in their predicted carbon forming regions showing Xu and Froment's kinetics as the most applicable and realistic model with Hou and Hughes' kinetics showing carbon formation inconsistent with industrial observations. This deeper understanding of carbon formation mechanisms allows the development of a system to avoid carbon formation using CO 2 in place of/in conjunction with steam.

No abstract

No abstract