Modeling and simulation of a turbulentspray in urea-SNCR for biomas exhaust gas aftertreatment

Abstract

The combustion of biomass is considered renewable since it is carbon neutral, but the combustion still releases nitrogen oxides (NOx). Although not a greenhouse gas, the emission of NOx leads to acidification, eutrophication and smog. Therefore, it is interesting to evaluate techniques to remove NOx from biomass combustion. The technique looked into in this project is urea selective non-catalytic reduction(SNCR), where a spray consisting of urea and water is injected into the flue gas. The aim of this project is to create a model for of the urea-SNCR process that can be used for design and optimization of the technique. Since the process is dependent on many complex phenomena such as turbulence, multiphase flow, chemical kinetics and mixing, a variety of models are evaluated with varying complexity. Three models are evaluated, both reactor models and computational fluid dynamics (CFD) models. The evaluated models are: reactor models - continuous stirred-tank reactor and continuous plug flow reactor, and turbulence models - k- and detached eddy simulation (DES). In order to determine the accuracy of the models, a comparison is made with measurement data from a biomass boiler located in Rörvik, Sweden. From the measurement data two operating temperatures are chosen, one over the optimal NOx reduction temperature and one below. The results show that it is important to account for heat losses, and that there is a large difference between DES and k- for the higher temperature. This discrepancy is caused by the high sensitivity of the NOx reduction reaction to the reaction temperature. There are arguments for and against the use of DES instead of k-, but more measurement data are needed in order to arrive at a definite conclusion.