Microkinetic models for heterogeneously catalyzed processes are essential to unravel the complexity of the chemical reactions and identify measures for intensification. One such process is the reduction of pollutants from exhaust gas emissions via oxidation in catalytic converters. Next to CH₄, emissions are C₂H₄, C₂H₆, CO, H₂ and NO. Due to the high computational cost of DFT, it is not feasible to explore all the possible chemistry for this complex system. However, this can be easily done by automated mechanism generation software using the open-source Reaction Mechanism Generator (RMG). The study aims to provide new insights into the oxidation mechanism of real exhaust gas mixtures through automated mechanism generation and multiscale modeling. Mechanisms are generated for Pt(111) for exhaust gas mixtures from typical stoichiometric combustion conditions in gasoline engines. The generated microkinetic models are used in a multiscale modeling approach and compared with experiments.

RMG identifies the essential chemistry from more than 800 considered intermediates and 2000 reactions for this system. A hierarchical refinement scheme is employed by using the degree of rate control to identify the most important intermediates and elementary steps. In a combined bottom-up and top-down approach, the thermochemistry of adsorbates is determined from DFT calculations, while activation barriers are tuned to match the experiments. The figure shows the simulation of a light-off curve using the generated microkinetics after the iterative hierarchical refinement procedure. A good agreement between simulation and experiments is observed and new insights into the complex emission oxidation chemistry are obtained. The study highlights the capabilities of automated mechanism generation for heterogeneously catalyzed processes and provides a universal framework for other reactions.

Comparison of experimental and simulated light-off curves.

Olaf Deutschmann

Appearances