Study of Nitrous Oxide Production in the Nitric Acid Process

Abstract

This work contains investigations relevant for the study of catalytic chemical reactors using the detailed microkinetics approach, with the intent of improving the prediction of the product distribution. The study is comprised of two main topics. The first one consists of an investigation of computational simplification for the chemistry. Two main approaches are studied. Firstly is the Lookup Table approach, where different techniques are explored with their limitations. The second approach is Neural Networks, which was found to provide lower memory requirements for the test case, and it can be computed by at least two orders of magnitude faster than the lookup table and than the direct computation of the chemistry. The second main area of investigation of this thesis consists of the study of the ammonia oxidation reactor, considering a chemical mechanism that shows reliable results based on the available literature. Different domains are studied, and the Sh and Nu numbers are investigated for these cases. It was found that correlations can predict accurately the wire temperature and the mass transfer coefficient. The production of the greenhouse gas nitrous oxide is investigated as a function of geometry and process conditions. The effect of radiation was studied, which showed a decrease of the wire temperatures which in turn led to a higher nitrous oxide selectivity, however it did not imply a shift on the trend. Not only the temperature was found to drive the selectivity. Also, the surface composition showed an effect, and a sensitivity study was performed. The results showed that the selectivity is most sensitive to the surface composition of ammonia, which was found from the CFD calculations. Lastly, the effect of turbulence was studied in a single wire case. It was found that the results are not different from the laminar case results in terms of wire temperature and selectivity.