One promising application in the emission control is the Selective Catalytic Reduction (SCR) system for the reduction of nitric oxides from exhaust emissions. Previous works at the institute have highlighted the importance of accurate CFD turbulence modeling with respect to the turbulent mixing of ammonia vapor . With the help of Laser Doppler Anemometry (LDA) measurements it was confirmed that RANS approaches are capable of predicting the velocity field adequately. In contrast, the turbulence level was underestimated for all RANS approaches .Based on this work the paper at hand presents CFD results using Large Eddy Simulation (LES). The sensitivity of the solution with respect to spatial and temporal resolution as well as the boundary conditions is demonstrated. In accordance with the Kolmogorov theory grid sizes ranging from 3.2 to 20 million cells were investigated using LES methodology. The representation of the velocity field was found to be accurate independent of the mesh size. The simulation results are compared with the measured spatial velocity vectors and Reynolds stress tensors. In general, all LES calculations are in good agreement with the measured Turbulent Kinetic Energy (TKE) data. The prediction of the TKE was improved significantly compared to the RANS calculations. The applicability of the LES methodology is shown for cold and hot operating conditions.Therefore it can be concluded that it is a promising approach to improve the accuracy of the scalar mixing process. Further work will concentrate on the implementation and validation of the liquid phase dynamics.