Escaping the trap of complication and complexity in multiscale microkinetic modelling of heterogeneous catalytic processes

Abstract

The hierarchical multiscale approach allows for escaping the trap of complication and complexity in multiscale microkinetic modelling of heterogeneous catalytic processes.

Constrained Chemical Dynamics of CO Dissociation/Hydrogenation on Rh Surfaces

“…Because of the underlying inaccuracies in the density functionala pproximationsa nd the models ystemsu sed in the calculations,hierarchicalr efiningofmechanismsusing semiempirical methods and/or experimental data is anotherc ommon approacht om echanism development. Thisi ncludes the works of Karakayaa nd Deutschmann supported by experimental data, [10,[18][19][20] the workso fV lachos andM aestri [7,[21][22][23] based on a hierarchical approach, the model of Mei et al [6] developed using both DFT and experiments, and our previouss emiempirical work on Rh. [9] Despite these efforts, agreement between the barriers obtained for the studiedp athways can vary considerably (see Ta ble 1), even thought he underlying ab initio and semiempirical methods used are similar.…”

“…Because of the underlying inaccuracies in the density functional approximations and the model systems used in the calculations, hierarchical refining of mechanisms using semiempirical methods and/or experimental data is another common approach to mechanism development. This includes the works of Karakaya and Deutschmann supported by experimental data, the works of Vlachos and Maestri based on a hierarchical approach, the model of Mei et al . developed using both DFT and experiments, and our previous semiempirical work on Rh .…”

Multiscale modelling of CO2 reduction to methanol over industrial Cu/ZnO/Al2O3 heterogeneous catalyst: Linking ab initio surface reaction kinetics with reactor fluid dynamics

“…Nevertheless, DFT is very useful in establishing trends, reaction mechanisms, reaction equilibria and gaining valuable insight into the microscopic behaviour during the reaction. In this work, we resorted to the following constrains for modifying first-principle microkinetic model: (i) finding a minimum set of elementary reactions to describe the observed conversion and selectivity; (ii) adjusting the activation barriers and reaction energy within the accuracy of the DFT calculations (error between 0.2 -0.7 eV for PBE functionals (Maestri, 2017)); (iii) all changes resulting in the same equilibrium state as experimentally measured. With an approach akin to (Grabow and Mavrikakis, 2011), a list of correction factors, as presented in Table 1, were obtained for reactions 4, 7 and 10 (fitting procedure on a single-pore approximation for the PBR showed that altering only kinetic parameters for the three reactions is sufficient to satisfy all the constrainssee section 4.2.3).…”

“…Despite extensive DFT calculations, their mean-field microkinetic model required some corrections of DFTcalculated values for the results to match the experimental data. In a step forward towards industrial application, Maestri presented the feasibility of coupling mean-field microkinetic model with CFD (Maestri, 2017). In their most recent work Maestri et al presented CO oxidation on ruthenium oxide, showcasing the capability of the approach in making the multiscale simulation of complex chemical reactors with tabulated KMC model possible (Bracconi and Maestri, 2020).…”

Prediction of morphological changes of catalyst materials under reaction conditions by combined ab initio thermodynamics and microkinetic modelling

“…The modelling of the crystal habit of the catalyst material under reaction conditions and its effect on the activity easily result in a high level of complexity 13. This is related to the necessity of both (i) building microkinetic models not only for one facet, but for several facets (potentially including corners, edges, and defects), and (ii) coupling the prediction of the structure and shape of the nanoparticle to the local variation of the chemical potential in the reactor.…”