Abstract
This study examines the modeling of hydrocarbon pyrolysis in a Ni0.27Bi0.73 molten metal alloy reactor. The model is executed in two stages. The first stage investigates the effect of the physical properties of the gas and molten liquid on the bubble-size distribution, and determines the Sauter mean bubble diameter in the Ni0.27Bi0.73 column. In this stage, a population-balance-based model using the Euler-Euler approach is coupled with nonreactive computational fluid dynamics in the ANSYS Fluent V17.2 software package. After estimating the Sauter mean diameter, the next stage computes the overall decomposition kinetics of hydrocarbons (gas phase + melt interface) and couples them with an existing hydrodynamic model to determine the final H-2 output and selectivity. The Sauter mean diameter was found to increase with increasing superficial gas velocity (or flow rate), liquid density, surface tension, column diameter, and decrease with increasing liquid viscosity. The hydrogen selectivity improved when the model included the surface kinetics, and the hydrogen selectivity of higher hydrocarbons was comparable to (or even higher than) that of pure methane at 1000 ?.