Evidence for the Active Phase of Heterogeneous Catalysts through In Situ Reaction Product Imaging and Multiscale Modeling

Kinetic Monte Carlo simulations for heterogeneous catalysis: Fundamentals, current status, and challenges

Kinetic Monte Carlo (KMC) simulations in combination with first-principles (1p)-based calculations are rapidly becoming the gold-standard computational framework for bridging the gap between the wide range of length scales and time scales over which heterogeneous catalysis unfolds. 1p-KMC simulations provide accurate insights into reactions over surfaces, a vital step toward the rational design of novel catalysts. In this Perspective, we briefly outline basic principles, computational challenges, successful applications, as well as future directions and opportunities of this promising and ever more popular kinetic modeling approach.

Active learning-based framework for optimal reaction mechanism selection from microkinetic modeling: a case study of electrocatalytic oxygen reduction reaction on carbon nanotubes

The elucidation of complex electrochemical reaction mechanisms requires advanced models with many intermediate reaction steps, which are governed by a large number of parameters like reaction rate constants and charge transfer coefficients. Overcomplicated models introduce high uncertainty in the choice of the parameters and cannot be used to obtain meaningful insights on the reaction pathway. We describe a new framework of optimal reaction mechanism selection based on the mean-field microkinetic modeling approach (MF-MKM) and adaptive sampling of model parameters. The optimal model is selected to provide both the accurate fitting of experimental data within the experimental error and low uncertainty of model parameters choice. Generally, this approach can be applied for any complex heterogeneous electrochemical reaction. We use the "2e^-" electrocatalytic oxygen reduction reaction (ORR) on carbon nanotubes (CNTs) as a representative example of a sufficiently complex reaction. Rotating disk electrode (RDE) experimental data for both ORR in O₂-saturated 0.1 M KOH solution and hydrogen peroxide oxidation/reduction reaction (HPRR/HPOR) in Ar-purged 0.1 M KOH solution with different HO₂^- concentrations were used to show the dependence of the model parameters uniqueness on the completeness of the experimental dataset. It is demonstrated that the optimal reaction mechanism for ORR on CNT and available experimental data consists of O₂ adsorption step on the electrode surface and effective step of two-electron reduction to HO₂^- combined with its desorption from the electrode. The low uncertainty of estimated model parameters is provided only within the 2-step model being applied to the full available experimental dataset. The assessment of elementary step mechanisms on electro-catalytic materials including carbon-based electrodes requires more diverse experimental data and/or higher precision of experimental measurements to facilitate more precise microkinetic modeling of more complex reaction mechanisms.

Perspective on theoretical methods and modeling relating to electro-catalysis processes

Theoretical methods and models for the description of thermodynamics and kinetics in electro-catalysis, including solvent effects, externally applied potentials, and many-body interactions, are discussed.

Combining Planar Laser-Induced Fluorescence with Stagnation Point Flows for Small Single-Crystal Model Catalysts: CO Oxidation on a Pd(100)

A stagnation flow reactor has been designed and characterized for both experimental and modeling studies of single-crystal model catalysts in heterogeneous catalysis. Using CO oxidation over a Pd(100) single crystal as a showcase, we have employed planar laser-induced fluorescence (PLIF) to visualize the CO2 distribution over the catalyst under reaction conditions and subsequently used the 2D spatially resolved gas phase data to characterize the stagnation flow reactor. From a comparison of the experimental data and the stagnation flow model, it was found that characteristic stagnation flow can be achieved with the reactor. Furthermore, the combined stagnation flow/PLIF/modeling approach makes it possible to estimate the turnover frequency (TOF) of the catalytic surface from the measured CO2 concentration profiles above the surface and to predict the CO2, CO and O2 concentrations at the surface under reaction conditions.

Addressing global uncertainty and sensitivity in first-principles based microkinetic models by an adaptive sparse grid approach

In the last decade, first-principles-based microkinetic modeling has been developed into an important tool for a mechanistic understanding of heterogeneous catalysis. A commonly known, but hitherto barely analyzed issue in this kind of modeling is the presence of sizable errors from the use of approximate Density Functional Theory (DFT). We here address the propagation of these errors to the catalytic turnover frequency (TOF) by global sensitivity and uncertainty analysis. Both analyses require the numerical quadrature of high-dimensional integrals. To achieve this efficiently, we utilize and extend an adaptive sparse grid approach and exploit the confinement of the strongly non-linear behavior of the TOF to local regions of the parameter space. We demonstrate the methodology on a model of the oxygen evolution reaction at the Co₃O₄ (110)-A surface, using a maximum entropy error model that imposes nothing but reasonable bounds on the errors. For this setting, the DFT errors lead to an absolute uncertainty of several orders of magnitude in the TOF. We nevertheless find that it is still possible to draw conclusions from such uncertain models about the atomistic aspects controlling the reactivity. A comparison with derivative-based local sensitivity analysis instead reveals that this more established approach provides incomplete information. Since the adaptive sparse grids allow for the evaluation of the integrals with only a modest number of function evaluations, this approach opens the way for a global sensitivity analysis of more complex models, for instance, models based on kinetic Monte Carlo simulations.

Local-metrics error-based Shepard interpolation as surrogate for highly non-linear material models in high dimensions

Many problems in computational materials science and chemistry require the evaluation of expensive functions with locally rapid changes, such as the turn-over frequency of first principles kinetic Monte Carlo models for heterogeneous catalysis. Because of the high computational cost, it is often desirable to replace the original with a surrogate model, e.g., for use in coupled multiscale simulations. The construction of surrogates becomes particularly challenging in high-dimensions. Here, we present a novel version of the modified Shepard interpolation method which can overcome the curse of dimensionality for such functions to give faithful reconstructions even from very modest numbers of function evaluations. The introduction of local metrics allows us to take advantage of the fact that, on a local scale, rapid variation often occurs only across a small number of directions. Furthermore, we use local error estimates to weigh different local approximations, which helps avoid artificial oscillations. Finally, we test our approach on a number of challenging analytic functions as well as a realistic kinetic Monte Carlo model. Our method not only outperforms existing isotropic metric Shepard methods but also state-of-the-art Gaussian process regression.

Novel in Situ Techniques for Studies of Model Catalysts

Motivated mainly by catalysis, gas-surface interaction between single crystal surfaces and molecules has been studied for decades. Most of these studies have been performed in well-controlled environments and have been instrumental for the present day understanding of catalysis, providing information on surface structures, adsorption sites, and adsorption and desorption energies relevant for catalysis. However, the approach has been criticized for being too far from a catalyst operating under industrial conditions at high temperatures and pressures. To this end, a significant amount of effort over the years has been used to develop methods to investigate catalysts at more realistic conditions under operating conditions. One result from this effort is a vivid and sometimes heated discussion concerning the active phase for the seemingly simple CO oxidation reaction over the Pt-group metals in the literature. In recent years, we have explored the possibilities to perform experiments at conditions closer to those of a technical catalyst, in particular at increased pressures and temperatures. In this contribution, results from catalytic CO oxidation over a Pd(100) single crystal surface using Near Ambient Pressure X-ray Photo emission Spectroscopy (NAPXPS), Planar Laser-Induced Fluorescence (PLIF), and High Energy Surface X-ray Diffraction (HESXRD) are presented, and the strengths and weaknesses of the experimental techniques are discussed. Armed with structural knowledge from ultrahigh vacuum experiments, the presence of adsorbed molecules and gas-phase induced surface structures can be identified and related to changes in the reactivity or to reaction induced gas-flow limitations. In particular, the application of PLIF to catalysis allows one to visualize how the catalyst itself changes the gas composition close to the model catalyst surface upon ignition, and relate this to the observed surface structures. The effect obscures a straightforward relation between the active phase and the activity, since in the case of CO oxidation, the gas-phase close to the model catalyst surface is shown to be significantly more oxidizing than far away from the catalyst. We show that surface structural knowledge from UHV experiments and the composition of the gas phase close to the catalyst surface are crucial to understand structure-function relationships at semirealistic conditions. In the particular case of Pd, we argue that the surface structure of the PdO(101) has a significant influence on the activity, due to the presence of Coordinatively Unsaturated Sites (CUS) Pd atoms, similar to undercoordinated Ru and Ir atoms found for RuO₂(110) and IrO₂(110), respectively.

Surface science under reaction conditions: CO oxidation on Pt and Pd model catalysts

Application of surface-science techniques, such as XPS, SXRD, STM, and IR spectroscopy under catalytic reactions conditions yield new structural and chemical information. Recent experiments focusing on CO oxidation over Pt and Pd model catalysts were reviewed.

2D and 3D imaging of the gas phase close to an operating model catalyst by planar laser induced fluorescence

In recent years, efforts have been made in catalysis related surface science studies to explore the possibilities to perform experiments at conditions closer to those of a technical catalyst, in particular at increased pressures. Techniques such as high pressure scanning tunneling/atomic force microscopy (HPSTM/AFM), near ambient pressure x-ray photoemission spectroscopy (NAPXPS), surface x-ray diffraction (SXRD) and polarization-modulation infrared reflection absorption spectroscopy (PM-IRAS) at semi-realistic conditions have been used to study the surface structure of model catalysts under reaction conditions, combined with simultaneous mass spectrometry (MS). These studies have provided an increased understanding of the surface dynamics and the structure of the active phase of surfaces and nano particles as a reaction occurs, providing novel information on the structure/activity relationship. However, the surface structure detected during the reaction is sensitive to the composition of the gas phase close to the catalyst surface. Therefore, the catalytic activity of the sample itself will act as a gas-source or gas-sink, and will affect the surface structure, which in turn may complicate the assignment of the active phase. For this reason, we have applied planar laser induced fluorescence (PLIF) to the gas phase in the vicinity of an active model catalysts. Our measurements demonstrate that the gas composition differs significantly close to the catalyst and at the position of the MS, which indeed should have a profound effect on the surface structure. However, PLIF applied to catalytic reactions presents several beneficial properties in addition to investigate the effect of the catalyst on the effective gas composition close to the model catalyst. The high spatial and temporal resolution of PLIF provides a unique tool to visualize the on-set of catalytic reactions and to compare different model catalysts in the same reactive environment. The technique can be applied to a large number of molecules thanks to the technical development of lasers and detectors over the last decades, and is a complementary and visual alternative to traditional MS to be used in environments difficult to asses with MS. In this article we will review general considerations when performing PLIF experiments, our experimental set-up for PLIF and discuss relevant examples of PLIF applied to catalysis.