Data-Centric Heterogeneous Catalysis: Identifying Rules and Materials Genes of Alkane Selective Oxidation.

Artificial intelligence (AI) can accelerate catalyst design by identifying key physicochemical descriptive parameters correlated with the underlying processes triggering, favoring, or hindering the performance. In analogy to genes in biology, these parameters might be called "materials genes" of heterogeneous catalysis. However, widely used AI methods require big data, and only the smallest part of the available data meets the quality requirement for data-efficient AI. Here, we use rigorous experimental procedures, designed to consistently take into account the kinetics of the catalyst active states formation, to measure 55 physicochemical parameters as well as the reactivity of 12 catalysts toward ethane, propane, and n-butane oxidation reactions. These materials are based on vanadium or manganese redox-active elements and present diverse phase compositions, crystallinities, and catalytic behaviors. By applying the sure-independence-screening-and-sparsifying-operator symbolic-regression approach to the consistent data set, we identify nonlinear property-function relationships depending on several key parameters and reflecting the intricate interplay of processes that govern the formation of olefins and oxygenates: local transport, site isolation, surface redox activity, adsorption, and the material dynamical restructuring under reaction conditions. These processes are captured by parameters derived from N2 adsorption, X-ray photoelectron spectroscopy (XPS), and near-ambient-pressure in situ XPS. The data-centric approach indicates the most relevant characterization techniques to be used for catalyst design and provides "rules" on how the catalyst properties may be tuned in order to achieve the desired performance.

Green synthesis of propylene oxide directly from propane.

The chemical industry faces the challenge of bringing emissions of climate-damaging CO2 to zero. However, the synthesis of important intermediates, such as olefins or epoxides, is still associated with the release of large amounts of greenhouse gases. This is due to both a high energy input for many process steps and insufficient selectivity of the underlying catalyzed reactions. Surprisingly, we find that in the oxidation of propane at elevated temperature over apparently inert materials such as boron nitride and silicon dioxide not only propylene but also significant amounts of propylene oxide are formed, with unexpectedly small amounts of CO2. Process simulations reveal that the combined synthesis of these two important chemical building blocks is technologically feasible. Our discovery leads the ways towards an environmentally friendly production of propylene oxide and propylene in one step. We demonstrate that complex catalyst development is not necessary for this reaction.

The Influence of the Chemical Potential on Defects and Function of Perovskites in Catalysis.

A Sm-deficient Sm0.96MnO3 perovskite was prepared on a gram scale to investigate the influence of the chemical potential of the gas phase on the defect concentration, the oxidation states of the metals and the nature of the oxygen species at the surface. The oxide was treated at 450°C in nitrogen, synthetic air, oxygen, water vapor or CO and investigated for its properties as a catalyst in the oxidative dehydrogenation of propane both before and after treatment. After treatment in water vapor, but especially after treatment with CO, increased selectivity to propene was observed, but only when water vapor was added to the reaction gas. As shown by XRD, SEM, EDX and XRF, the bulk structure of the oxide remained stable under all conditions. In contrast, the surface underwent strong changes. This was shown by AP-XPS and AP-NEXAFS measurements in the presence of the different gas atmospheres at elevated temperatures. The treatment with CO caused a partial reduction of the metals at the surface, leading to changes in the charge of the cations, which was compensated by an increased concentration of oxygen defects. Based on the present experiments, the influence of defects and concentration of electrophilic oxygen species at the catalyst surface on the selectivity in propane oxidation is discussed.

Materials genes of heterogeneous catalysis from clean experiments and artificial intelligence.

ABSTRACT: The performance in heterogeneous catalysis is an example of a complex materials function, governed by an intricate interplay of several processes (e.g., the different surface chemical reactions, and the dynamic restructuring of the catalyst material at reaction conditions). Modeling the full catalytic progression via first-principles statistical mechanics is impractical, if not impossible. Instead, we show here how a tailored artificial-intelligence approach can be applied, even to a small number of materials, to model catalysis and determine the key descriptive parameters ("materials genes") reflecting the processes that trigger, facilitate, or hinder catalyst performance. We start from a consistent experimental set of "clean data," containing nine vanadium-based oxidation catalysts. These materials were synthesized, fully characterized, and tested according to standardized protocols. By applying the symbolic-regression SISSO approach, we identify correlations between the few most relevant materials properties and their reactivity. This approach highlights the underlying physicochemical processes, and accelerates catalyst design. IMPACT STATEMENT: Artificial intelligence (AI) accepts that there are relationships or correlations that cannot be expressed in terms of a closed mathematical form or an easy-to-do numerical simulation. For the function of materials, for example, catalysis, AI may well capture the behavior better than the theory of the past. However, currently the flexibility of AI comes together with a lack of interpretability, and AI can only predict aspects that were included in the training. The approach proposed and demonstrated in this IMPACT article is interpretable. It combines detailed experimental data (called "clean data") and symbolic regression for the identification of the key descriptive parameters (called "materials genes") that are correlated with the materials function. The approach demonstrated here for the catalytic oxidation of propane will accelerate the discovery of improved or novel materials while also enhancing physical understanding. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1557/s43577-021-00165-6.

Site specific and localized structural displacements in open structured multimetallic oxides.

The structures of solids can locally differ from the macroscopic picture obtained by structural averaging techniques. This difference significantly influences the performance of any functional material. Measurements of these local structures are challenging. Thus, the description of defects is often disregarded. However, in order to understand the functionality, such irregularities have to be investigated. Here, we present a high resolution scanning transmission electron microscopic (STEM) study revealing local structural irregularities in open structured oxides using catalytically active orthorhombic (Mo,V,Te,Nb)Ox as a complex example. Detailed analysis of annular dark field- and annular bright field-STEM images reveal site specific local structural displacements of individual framework and channel sites in the picometer range. These experimental observables can be considered as an important structural addendum for theoretical modelling and should be implemented into the existing data in order to quantify site specific potential energies and stresses. This information can further be used to describe the impact of the structure on the catalytic performance in greater detail.

Catalytic activity, water formation, and sintering: Methane activation over Co- and Fe-doped MgO nanocrystals.

Microstructure, structure, and compositional homogeneity of metal oxide nanoparticles can change dramatically during catalysis. Considering the different stabilities of cobalt and iron ions in the MgO host lattice [M. Niedermaier et al., J. Phys. Chem. C 123, 25991 (2019)], we employed MgO nanocube powders with or without transition metal admixtures for the oxidative coupling of methane (OCM) reaction to analyze characteristic differences in catalytic activity and sintering behavior. Undoped MgO nanocrystals exhibit the highest C2 selectivity and retain the nanocrystallinity of the starting material after 24 h time on stream. For the Co-Mg-O nanoparticle powder, which exhibits the highest activity and COx selectivity and where OCM-induced coarsening is strongest, we found that the Co2+ ions remain homogeneously distributed over the MgO lattice. Trivalent Fe ions migrate to the surface of Fe-Mg-O nanoparticles where they form a magnesioferrite phase (MgFe2O4) with a characteristic impact on catalytic performance: Fe-Mg-O is initially less selective than MgO despite its lower activity. An increase in C2 selectivity and a decrease in the CO2/CO ratio with time on stream are attributed to the increasing fraction of coarsened particles that become depleted in redox active Fe. Surface water is a by-product of the OCM reaction, favors mass transport across the particle surfaces, and serves as a sintering aid during catalysis. The characteristic changes in size and morphology of MgO, Co-doped, and Fe-doped MgO particles can be consistently explained by activity and C2 selectivity trends. The original morphology of the nanocubes as a starting material for the OCM reaction does not impact the catalytic activity.

How strain affects the reactivity of surface metal oxide catalysts.

Highly dispersed molybdenum oxide supported on mesoporous silica SBA-15 has been prepared by anion exchange resulting in a series of catalysts with changing Mo densities (0.2-2.5 Mo atoms nm(-2) ). X-ray absorption, UV/Vis, Raman, and IR spectroscopy indicate that doubly anchored tetrahedral dioxo MoO4 units are the major surface species at all loadings. Higher reducibility at loadings close to the monolayer measured by temperature-programmed reduction and a steep increase in the catalytic activity observed in metathesis of propene and oxidative dehydrogenation of propane at 8 % of Mo loading are attributed to frustration of Mo oxide surface species and lateral interactions. Based on DFT calculations, NEXAFS spectra at the O-K-edge at high Mo loadings are explained by distorted MoO4 complexes. Limited availability of anchor silanol groups at high loadings forces the MoO4 groups to form more strained configurations. The occurrence of strain is linked to the increase in reactivity.

Aiding the self-assembly of supramolecular polyoxometalates under hydrothermal conditions to give precursors of complex functional oxides.

In situ Raman spectroscopy allows insight into molecular processes under hydrothermal conditions during synthesis of complex nanostructured MoVTeNb oxides (see picture: Nb yellow, Mo blue, V/Mo pale blue, Te red). Based on the knowledge acquired, the synthesis can be more efficiently directed towards the desired product with improved functionality.