Abstract
A four-stage industrial ammonia production reactor comprising four fixed beds and one internal heat exchanger was simulated in this work. The flows are quenched in between the stages in order to keep their temperatures at desired values. The beds are assumed heterogeneous when modeling the reactor. Two different techniques to compute effectiveness factor are reported in the literature. One is writing heat and mass transfer equations for catalyst particle and the other is employing an empirical correlation of temperature, pressure and conversion ratio. The results of this research proved, due to the random geometry of the catalyst, the empirical correlation is superior. A two-dimensional model was developed for each bed to predict temperature and concentrations in any radial or axial position. Model equations were solved applying a forth order Rounge Kutta method in MATLAB environment. Partial differential equations of the two-dimensional model were discretized applying finite difference method. The resulting algebraic equations were then solved using Gauss technique. The catalytic beds were also simulated employing FEMLAB CFD package. The results indicated there is a considerable temperature gradient near the beds wall. The model predictions were compared with industrial data. The error in prediction of temperatures varies from 2°C to 6°C in different beds while the difference in concentration and conversion ratio is negligible.
