Search for: All records

In this paper, we discuss the concept and properties of variance-based global sensitivity analysis, as an expansion of local sensitivity metrics (such as the degree of rate control), for modeling and design of catalytic reaction systems. Using an illustrative example and supporting theory, we show that: (i) for small variations in the parameters, global sensitivities are similar to local derivatives; (ii) for larger variations in the parameters (i.e., a larger parameter space), the global sensitivities provide a ranking of importance of parameters and impose a rigorous bound on the errors that arise from fixing one or more parameters to nominal values; and (iii) in general, the global sensitivities can be related to the extrema of local derivatives. We argue that the square root of the total global sensitivity of a parameter, computed by summing the global sensitivity of that parameter acting independently and in combination with others, is a “global” degree of rate control for catalytic systems. 

Ab initio microkinetic modeling, parameterized using density functional theory (DFT) energies, is a common tool to quantify reaction rates and analyze reaction mechanisms a priori in heterogeneous catalysis. Such models, however, often have large prediction errors even if they include plausible reaction steps and correctly model the active sites; this is partially due to the intrinsic inaccuracies of the chosen DFT functional. Borrowing concepts from Bayesian calibration theory, we show that transferable data-driven corrections to DFT energies in the form of Gaussian process models trained on single-crystal adsorption calorimetry data can improve the accuracy of microkinetic models substantially. Specifically, we demonstrate that such corrections improve the predictive accuracy of the microkinetic model of the water-gas shift reaction on single-crystal Cu(111) surface by 3 orders of magnitude. We finally show that Gaussian process corrections serve as informed priors in a Bayesian experimental design framework to learn an accurate a posteriori microkinetic model from few kinetic experiments. We posit that these results suggest that even infusing small, related, high-fidelity thermochemistry data, when available, can systematically and substantially improve the predictive accuracy of microkinetic models. 

Ethylene oxidation by Ag catalysts has been extensively investigated over the past few decades, but many key fundamental issues about this important catalytic system are still unresolved. This overview of the selective oxidation of ethylene to ethylene oxide by Ag catalysts critically examines the experimental and theoretical literature of this complex catalytic system: (i) the surface chemistry of silver catalysts (single crystal, powder/foil, and supported Ag/α-Al2O3), (ii) the role of promoters, (iii) the reaction kinetics, (iv) the reaction mechanism, (v) density functional theory (DFT), and (vi) microkinetic modeling. Only in the past few years have the modern catalysis research tools of in situ/operando spectroscopy and DFT calculations been applied to begin establishing fundamental structure−activity/selectivity relationships. This overview of the ethylene oxidation reaction by Ag catalysts covers what is known and what issues still need to be determined to advance the rational design of this important catalytic system.