Autothermal reforming of methane gas—Modelling and experimental validation

Abstract

Autothermal reforming of methane (CH 4) is a complex thermo-chemical process. It consists of steam reforming (STR) and partial oxidation (POX), while operating under thermal-neutral condition. This paper presents an enhancement to the two-dimensional (2-D) kinetic model, developed in the previous study, by including (1) non-plug flow, radial varying gas velocity in the axial direction; (2) radial varying porosity of the reactor; and (3) effectiveness of gas diffusing into the porous catalyst pallets. In order to validate the predictive capability of the model, extensive experiments were conducted using a reforming system that was developed in-house. The results were used to verify the mathematical model running under identical operating parameters and identical reactor's geometric parameters. These experimental results are more aligned to that expected of the mathematical model than previous works. Maximum error of 9% for temperature profile, 7% for hydrogen and 8% for carbon monoxide was shown between the experimental and current predicted results. Further insights into the autothermal reforming process were obtained by studying the effect of varying inlet gas temperatures and reformer radius. Varying the inlet gas temperatures did not produce significant difference in hydrogen (H 2) yield and CH 4 conversion, when the inlet gas temperature was increased to 623 K. It was also noted that POX was dominant in the front portion, thermal-neutral in the mid portion, while slightly stronger presence of STR at the rear portion of the reactor. In addition, this study showed that the reactor reaches a steady state at a faster rate when the inlet gas temperature was lowered. Besides this, the study revealed that changes in reactor radius neither produced significant effect on the H 2 yield nor CH 4 conversion. It is suspected that within the range of radii used in this study, the resident time of the gases remains sufficient to negate the difference in radius change. Therefore, the difference in H 2 yield was insignificant at the exit. It was also noted that the time required for the reactor to reach steady state is shorter when the radius is smaller.