Multi-scale CFD+kMC simulation of methanol partial oxidation over ceria in a fixed bed
Qingang Xiong, Sreekanth Pannala, Stuart C. Daw, Aditya Savara, Steven H. Overbury
Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, USA
Abstract A research project aiming to achieve multi-scale computational models of catalytic methanol partial oxidation over ceria (CeO2) was initiated for better utilization of clean energy. Relevant reactor design and optimization of operating conditions require fundamental understanding on gas flow and heterogeneous reactions. Computational fluid dynamics (CFD) can complement experiment to provide spatiotemporal details for reactor design and optimization with a reduced developing cycle. However, the underlying reaction mechanisms of methanol partial oxidation on the surface of ceria remain unclear, which hinders an accurate prediction of reactor performance by CFD. In this study, a global reaction pathway which was proposed based on the information from both experiments and literature was employed with CFD to simulate a laboratory-scale fixed bed for isothermal ceria catalyzed methanol partial oxidation. In the corresponding experiment, CeO2 was diluted with silica sand to form the fixed bed, and 5 % methanol/ Ar was flowed from the reactor bottom. The reactor temperature was varied from 573 K to 673 K to realize six isothermal conditions. The pure macro-scale CFD results were corrected by the detailed information obtained from the micro-scale kinetic Monte Carlo simulation on the ceria surface. A compound wavelet matrix approach was employed to realize the multi-scale coupling. The corrected global reaction kinetics was then employed to simulate product composition and yields under different operating conditions and compared with experiment for the validation of the multi-scale coupling strategy.
Research sponsored by the Laboratory Directed Research and Development Program of Oak Ridge National Laboratory, managed by UT-Battelle, LLC, for the U. S. Department of Energy