6 Search Results

The complex reaction network of catalytic biomass conversions often involves hundreds of surface intermediates and thousands of reaction steps, greatly hindering the rational design of metal catalysts for these conversions. Here, we present a framework of machine learning (ML)-accelerated first-principles studies for the hydrodeoxygenation (HDO) of propanoic acid over transition metal surfaces. The microkinetic model (MKM) is initially parametrized by ML-predicted energies and iteratively improved by identifying the rate-determining species and steps (RDS), computing their energies by density functional theory (DFT), and reparameterizing the MKM until all the RDS are computed by DFT. The Gaussian process (GP) model performs significantlymore » better than the linear ridge regression model for predicting both the adsorption free energies and transition state free energies. Parameterized with energies from the GP model, only 5–20% of the full reaction network has to be computed by DFT for the MKM to possess DFT-level accuracy for the TOF and dominant reaction pathway. While the linear ridge regression model performs worse than the GP model, its performance is greatly improved when only transition states are predicted by the regression model and adsorption energies are computed by DFT. Overall, we find that a high accuracy in adsorption free energies is more important for a reliable MKM than a high accuracy in TS free energies. Lastly, based on the GP model with GOH and GCHCHCO as catalyst descriptors, we build two-dimensional volcano plots in activity and selectivity that can help design promising alloy catalysts for HDO reactions of organic acids.« less

Silica-supported tungsten oxides are widely used industrial catalysts for olefin metathesis due to their low cost and robustness, yet the mechanisms for heterogeneously catalyzed metathesis reactions remain comparatively less understood. In this work, density-functional theory (DFT) calculations were used to study the model reactions of propene metathesis. Our calculations confirm that the metathesis reactions catalyzed by WO_x/SiO₂ largely follow the Chauvin cycle, with an overall energetic barrier of 142 kJ/mol. To understand how the initial alkylidene active sites are generated, three mechanisms were examined: The pseudo-Wittig mechanism was found to be most favorable and proceeds with a metallacycle intermediate, whilemore » the allylic and vinylic C-H activations are much more difficult and require the reduction of surface sites to W(+4). Relative to the adsorbed reactant state, the overall intrinsic barriers for three mechanisms were computed to be 193, 261, and 355 kJ/mol, respectively. In conclusion, the higher barriers for active-site formation than for the metathesis cycle are consistent with the difficult, high-temperature pretreatment required in experiments to activate WO_x/SiO₂ catalysts.« less

Cu-based alloy catalysts have recently been investigated experimentally for the hydrodeoxygenation (HDO) of biomass-derived organic acids. Here, the HDO of propanoic acid (PAc) has been studied over Cu(111) by mean-field microkinetic modeling based on parameters obtained from first-principles calculations. Models were developed for the gas- and liquid-phase HDO in condensed water and 1,4-dioxane. In agreement with experimental observations, the gas-phase PAc conversion rate is low at 573 K and increases in liquid water by 1 order of magnitude. In all reaction environments, the decarboxylation mechanism is dominant at low hydrogen partial pressures less than 0.1 bar, and the C–COO bondmore » dissociation is the rate-controlling elementary step. This observation contrasts with the rate-controlling step identified over most group VIII metal surfaces, which is the C–OH bond dissociation in the decarbonylation mechanism. At high hydrogen (H₂) partial pressures greater than 10 bar, the HDO of PAc produces propionaldehyde that can readsorb and further react through decarbonylation to produce C₂ alkane products, which is conceptually different from the low H₂ partial pressure scenario. At high H₂ partial pressures, the initial hydrogenation at the carbonyl carbon of PAc becomes the rate-controlling elementary step.« less

Metal-free boron nitride has recently been reported to exhibit high catalytic activity and selectivity towards oxidative dehydrogenation of propane (ODHP). Despite several experimental studies, the exact nature and function of the active sites in this emerging catalyst are unknown. In the present work, density functional theory calculations are combined with microkinetic modeling to systematically explore the ODHP on oxygen passivated boron nitride nanoribbons (BNNRs). Here, the relative stabilities of different edge structures of BNNR were examined using ab initio thermodynamic analysis, and the most stable oxygen passivated edge structure was used for a mechanistic study. Microkinetic analysis revealed that amore » N₂O or NO_x-type active site is active and selective for the conversion of propane to propene. In addition to a heterogeneous catalytic cycle, the proposed N₂O/NO_x-type active sites are expected to be able to generate gas phase C₃H₇˙, which in turn can trigger gas phase reactions of ODHP as experimentally speculated.« less

Computational catalyst discovery involves identification of a meaningful model and suitable descriptors that determine the catalyst properties. We study the impact of combining various descriptors (e.g., reaction energies, metal descriptors, and bond counts) for modeling transition-state energies (TS) based on a database of adsorption and TS energies across transition-metal surfaces for the decarboxylation and decarbonylation of propionic acid, a chemistry characteristic for biomass conversion. Results of different machine learning models for more than 1572 descriptor combinations suggest that there is no statistically significant difference between linear and nonlinear models when using the right combination of reactant energies, metal descriptors, andmore » bond counts. However, linear models are inferior when not including bond count and metal descriptors. In conclusion, when there are missing data for reaction steps on all metals, conventional linear scaling is inferior to linear and nonlinear models with proper choice of descriptors that are surprisingly robust.« less