Promoting active site renewal in heterogeneous olefin metathesis catalysts.

As an atom-efficient strategy for the large-scale interconversion of olefins, heterogeneously catalysed olefin metathesis sees commercial applications in the petrochemical, polymer and speciality chemical industries1. Notably, the thermoneutral and highly selective cross-metathesis of ethylene and 2-butenes1 offers an appealing route for the on-purpose production of propylene to address the C3 shortfall caused by using shale gas as a feedstock in steam crackers2,3. However, key mechanistic details have remained ambiguous for decades, hindering process development and adversely affecting economic viability4 relative to other propylene production technologies2,5. Here, from rigorous kinetic measurements and spectroscopic studies of propylene metathesis over model and industrial WOx/SiO2 catalysts, we identify a hitherto unknown dynamic site renewal and decay cycle, mediated by proton transfers involving proximal Brønsted acidic OH groups, which operates concurrently with the classical Chauvin cycle. We show how this cycle can be manipulated using small quantities of promoter olefins to drastically increase steady-state propylene metathesis rates by up to 30-fold at 250 °C with negligible promoter consumption. The increase in activity and considerable reduction of operating temperature requirements were also observed on MoOx/SiO2 catalysts, showing that this strategy is possibly applicable to other reactions and can address major roadblocks associated with industrial metathesis processes.

molSimplify: A toolkit for automating discovery in inorganic chemistry.

We present an automated, open source toolkit for the first-principles screening and discovery of new inorganic molecules and intermolecular complexes. Challenges remain in the automatic generation of candidate inorganic molecule structures due to the high variability in coordination and bonding, which we overcome through a divide-and-conquer tactic that flexibly combines force-field preoptimization of organic fragments with alignment to first-principles-trained metal-ligand distances. Exploration of chemical space is enabled through random generation of ligands and intermolecular complexes from large chemical databases. We validate the generated structures with the root mean squared (RMS) gradients evaluated from density functional theory (DFT), which are around 0.02 Ha/au across a large 150 molecule test set. Comparison of molSimplify results to full optimization with the universal force field reveals that RMS DFT gradients are improved by 40%. Seamless generation of input files, preparation and execution of electronic structure calculations, and post-processing for each generated structure aids interpretation of underlying chemical and energetic trends. © 2016 Wiley Periodicals, Inc.

Ni nanoparticles supported on Al2O3 + La2O3 yolk-shell catalyst for photo-assisted thermal decomposition of methane.

Catalytic methane decomposition (CMD) has emerged as an appealing technology for large-scale production of H2 and carbon nanostructures from natural gas. As the CMD process is mildly endothermic, the application of concentrated renewable energy sources such as solar energy under a low-temperature regime could potentially represent a promising approach towards CMD process operation. Herein, Ni/Al2O3-La2O3 yolk-shell catalysts are fabricated using a straightforward single-step hydrothermal approach and tested for their performance in photothermal CMD. We show that the morphology of the resulting materials, dispersion and reducibility of Ni nanoparticles, and nature of metal-support interactions can be tuned by addition of varying amounts of La. Notably, the addition of an optimal amount of La (Ni/Al-20La) improved the H2 yield and catalyst stability relative to the base Ni/Al2O3 material, while also favoring base growth of carbon nanofibers. Additionally, we show for the first time a photothermal effect in CMD, whereby the introduction of 3 suns light irradiation at a constant bulk temperature of 500 °C reversibly increased the H2 yield of catalyst by about 1.2 times relative to the rate in the dark, accompanied by a decrease in apparent activation energy from 41.6 kJ mol-1 to 32.5 kJ mol-1. The light irradiation further suppressed undesirable CO co-production at low temperatures. Our work reveals photothermal catalysis as a promising route for CMD while providing an insightful understanding of the roles of modifier in enriching methane activation sites on Al2O3-based catalysts.

A Machine Learning Approach to Zeolite Synthesis Enabled by Automatic Literature Data Extraction.

Zeolites are porous, aluminosilicate materials with many industrial and "green" applications. Despite their industrial relevance, many aspects of zeolite synthesis remain poorly understood requiring costly trial and error synthesis. In this paper, we create natural language processing techniques and text markup parsing tools to automatically extract synthesis information and trends from zeolite journal articles. We further engineer a data set of germanium-containing zeolites to test the accuracy of the extracted data and to discover potential opportunities for zeolites containing germanium. We also create a regression model for a zeolite's framework density from the synthesis conditions. This model has a cross-validated root mean squared error of 0.98 T/1000 Å3, and many of the model decision boundaries correspond to known synthesis heuristics in germanium-containing zeolites. We propose that this automatic data extraction can be applied to many different problems in zeolite synthesis and enable novel zeolite morphologies.

Unifying Exchange Sensitivity in Transition-Metal Spin-State Ordering and Catalysis through Bond Valence Metrics.

Accurate predictions of spin-state ordering, reaction energetics, and barrier heights are critical for the computational discovery of open-shell transition-metal (TM) catalysts. Semilocal approximations in density functional theory, such as the generalized gradient approximation (GGA), suffer from delocalization error that causes them to overstabilize strongly bonded states. Descriptions of energetics and bonding are often improved by introducing a fraction of exact exchange (e.g., erroneous low-spin GGA ground states are instead correctly predicted as high-spin with a hybrid functional). The degree of spin-splitting sensitivity to exchange can be understood based on the chemical composition of the complex, but the effect of exchange on reaction energetics within a single spin state is less well-established. Across a number of model iron complexes, we observe strong exchange sensitivities of reaction barriers and energies that are of the same magnitude as those for spin splitting energies. We rationalize trends in both reaction and spin energetics by introducing a measure of delocalization, the bond valence of the metal-ligand bonds in each complex. The bond valence thus represents a simple-to-compute property that unifies understanding of exchange sensitivity for catalytic properties and spin-state ordering in TM complexes. Close agreement of the resulting per-metal-organic-bond sensitivity estimates, together with failure of alternative descriptors demonstrates the utility of the bond valence as a robust descriptor of how differences in metal-ligand delocalization produce differing relative energetics with exchange tuning. Our unified description explains the overall effect of exact exchange tuning on the paradigmatic two-state FeO+/CH4 reaction that combines challenges of spin-state and reactivity predictions. This new descriptor-sensitivity relationship provides a path to quantifying how predictions in transition-metal complex screening are sensitive to the method used.

Where Does the Density Localize? Convergent Behavior for Global Hybrids, Range Separation, and DFT+U.

Approximate density functional theory (DFT) suffers from many-electron self-interaction error, otherwise known as delocalization error, that may be diagnosed and then corrected through elimination of the deviation from exact piecewise linear behavior between integer electron numbers. Although paths to correction of energetic delocalization error are well-established, the impact of these corrections on the electron density is less well-studied. Here, we compare the effect on density delocalization of DFT+U (i.e., semilocal DFT augmented with a Hubbard U correction), global hybrid tuning, and range-separated hybrid tuning on a diverse test set of 32 transition metal complexes and observe the three methods to have qualitatively equivalent effects on the ground state density. Regardless of valence orbital diffuseness (i.e., from 2p to 5p), ligand electronegativity (i.e., from Al to O), basis set (i.e., plane wave versus localized basis set), metal (i.e., Ti, Fe, Ni), and spin state, or tuning method, we consistently observe substantial charge loss at the metal and gain at ligand atoms (∼0.3-0.5 e or more). This charge loss at the metal is preferentially from the minority spin, leading to increasing magnetic moment as well. Using accurate wave function theory references, we observe that a minimum error in partial charges and magnetic moments occurs at higher tuning parameters than typically employed to eliminate energetic delocalization error. These observations motivate the need to develop multifaceted approximate-DFT error correction approaches that separately treat density delocalization and energetic errors to recover both correct density and orbital energy-derived properties.