NOx reduction in a single representative channel of an SCR converter is modeled accounting for convective mass transfer along the channel, diffusive mass transfer and chemical reactions within the catalytic layer, coupled via mass transfer boundary condition at the wall. The proposed 1D+1D code for the channel has different parameters involving rate expression constants for the surface chemistry and properties of the catalytic and washcoated layer. A comprehensive set of experimental data on the Fe-Zeolite washcoated monolith was available for different feed concentrations, temperatures and flow rates. A genetic algorithm was used as an optimization technique to calibrate the model parameters by minimizing the error between simulation results and transient responses of the real system. Dynamic tests showing the response of the system during fast transients were subsequently used to assess the predictive quality of the model with the calibrated parameters. In particular, different rate expressions for the surface reactions were compared to find the best fit for the tested catalyst, providing insight with respect to the importance of the different reaction pathways and the chemico-physical properties of the system. Based on the findings, an optimization strategy could be proposed both in terms of sequence as well as minimum required experimental tests for each sub-system according to the parameters which can be determined separately.

This content is only available via PDF.
You do not currently have access to this content.