ACS Catalysis 最新文献

In computational electrochemistry, the adsorption strength of key intermediates is widely considered to be the primary factor determining product selectivity. However, in CO₂ electroreduction, the *OCHO* intermediate leading to HCOOH adsorbs more strongly than *COOH leading to CO on most d-block metals, including CO-selective ones. This discrepancy has been attributed to the instability of the *OCO* precursor kinetically hindering *OCHO* formation. Interestingly, Cu exhibits comparable selectivity for both HCOOH and CO, prompting investigation of the underlying mechanisms. Herein, using ab initio molecular dynamics simulations combined with explicit solvation and kinetic simulations, we reveal that the modulated hydrogen bond network induced by surface hydrogen (*H) plays a critical role in controlling the reaction pathways on Cu. This reorganized network induces a correlated motion between CO₂ and *H, which facilitates the formation of *OCHO*, resulting in similar activation barriers and partially overlapping pathways for CO and HCOOH. These findings offer a unified mechanism for understanding the divergence and connection of the two-electron pathways on Cu, and highlight the importance of interfacial structure in electrocatalysis and future catalyst design.

Alkali metals are recognized as effective additives to enhance catalytic activity in heterogeneous catalysis. Nevertheless, the traditional alkali metal promotion effect as electron donors cannot circumvent the linear scaling relationship between intricate reaction intermediates, and achieving highly efficient catalytic processes is still challenging. Herein, we propose an alkali metal template effect for the Co–Mo ensemble, which induces phase transition and intermetallic nanoparticle exsolution to construct a metal-nitride heterostructure. Detailed studies reveal that alkali metal atoms are embedded into the lattice of the oxide precursor rather than being electrostatically deposited over its surface, thus weakening the metal–oxygen interaction and facilitating nanoparticle exsolution from host materials. The prepared Co₃Mo/Co₂Mo₃N catalyst delivers a superior activity of 12.3 mmol·g_cat^–1·h^–1 for ammonia synthesis at 400 °C and 0.9 MPa, outperforming the well-recognized Co₃Mo₃N monophase nitride. We demonstrate that the enhanced activity is attributed to a dual-site mechanism for the independent activation of reactants on nitride and intermetallic surfaces rather than an electronic effect. These findings open perspectives for understanding the potential promotion effect of alkali metals and surpassing the Sabatier optimality.

Cu^I-catalyzed hydroboration of alkynes is a cost-effective route to trans-alkenyl boronates (valuable intermediates for C–C cross-coupling) but controlling regioselectivity in these reactions is challenging. Here, we address this challenge by developing a machine learning model, using high-throughput computational-derived descriptors and a combined experimental/literature data set to predict regioselectivity (expressed as ΔΔG^‡). The resulting support vector regression (SVR) model achieved high accuracy (cross-validated R² > 0.8, RMSE ∼0.4–0.6 kcal/mol) and revealed mechanistically relevant trends through feature importance analysis. For example, the model rationalizes why certain N-donor ligands that fail to promote Cu-catalyzed hydroboration can effectively catalyze alkyne hydrosilylation, linking this divergence to differences in geometric ligand descriptors. Moreover, ML-guided screening identified promising ligands that improved hydroboration outcomes (increased yield and regioselectivity), as confirmed in experiments (e.g., enabling reduced catalyst loading without sacrificing selectivity). Overall, this integrated ML strategy offers a powerful tool for understanding and predicting regioselectivity in Cu^I-mediated reactions and, with appropriate calibration, could be extended to other organocopper systems.

The persistent challenge of catalyst deactivation in methanol-to-olefins (MTO) conversion, primarily arising from restricted molecular transport and subsequent coke accumulation in conventional microporous SAPO-34 zeolites, necessitates innovative structural solutions. Herein, we demonstrate a synthesis strategy that employs sucrose-derived carbon as a hard template combined with vapor-phase transport to fabricate hierarchical SAPO-34 single crystals, thereby overcoming the low crystallinity and poor pore connectivity that have previously plagued hierarchical SAPO-34 zeolites. This intracrystalline hierarchical architecture with highly interconnected pores demonstrates molecular highway functionality and exhibits a 77.5% increase in propylene diffusion coefficient compared to conventional samples. Such hierarchical molecular highway architecture effectively regulates coke distribution within SAPO-34 crystals during MTO reactions. The optimized hierarchical SAPO-34 exhibits a 505 min operational lifetime, a 3.3-fold enhancement in catalytic durability over traditional microporous systems. Our findings establish a materials design paradigm for overcoming diffusion-reaction trade-offs in zeolite catalysis, with implications extending beyond conventional MTO processes.

Few-layered potassium nickel and cobalt oxides show drastic differences in catalytic activity based on metal ion preorganization. Uniform compositions [(CoO₂/K)₆ or (NiO₂/K)₆] show limited activity, while homogeneously mixed-metal cobalt/nickel oxides [(Co_nNi_(1–n)O₂/K)₆] display moderate improvement. However, a layer-by-layer arrangement of alternating cobalt and nickel oxide sheets [e.g., (CoO₂/K/NiO₂/K)] provides superior catalytic performance, reducing the oxygen evolution overpotential by ∼200–400 mV. Density functional theory simulations provide an illustration of the electronic properties (density of states and localization of orbitals) that promote catalysis in the layer-segregated materials over those of homogeneous composition. This study reveals that atomic preorganization of metal ions within layered catalysts plays a more crucial role than the overall metal composition in enhancing catalytic efficiency for oxygen evolution.

Developing machine learning (ML) models for yield prediction of chemical reactions has emerged as an important use case scenario in very recent years. In this space, reaction datasets present a range of challenges mostly stemming from imbalance and sparsity. Herein, we consider chemical language representations for reactions to tap into the potential of natural language processing models such as the ULMFiT (Universal Language Model Fine-Tuning) for yield prediction, which is customized to work across such distribution settings. We contribute a reaction dataset with more than 860 manually curated reactions collected from literature spanning over a decade, belonging to a family of catalytic meta-C(sp²)–H bond activation reactions of high contemporary importance. Taking cognizance of the dataset size, skewness toward the higher yields, and the sparse distribution characteristics, we developed (i) a time- and resource-efficient pretraining strategy for downstream transfer learning, and (ii) the CFR (classification followed by regression) model, which provided more accurate yield predictions than the traditional direct regression (DR) methods. Instead of the prevailing pretraining practice of using a large number of unlabeled molecules (1.4 million) from the ChEMBL dataset, we first created a pretraining dataset SSP1 (0.11 million) by using substructure-based mining from the PubChem database, which is found to be equally effective and more time-efficient in offering enhanced performance. The CFR model with the ULMFiT-SSP1 regressor achieved a notable RMSE of 8.40 ± 0.12 for the CFR-major and 6.48 ± 0.29 for the CFR-minor class in yield prediction on the title reaction, with a class boundary of yield at 53%. Furthermore, the CFR model is highly generalizable as evidenced by the significant improvement over the previous benchmarks on Buchwald–Hartwig coupling, Suzuki coupling, nickel-catalyzed C–O coupling, and USPTO reaction datasets. Our model provides a scalable and efficient framework that can assist in reaction discovery.

Transition metal alkyl species are widespread intermediates and prefer to undergo the β-H elimination process to release HX, which results in β-H loss. Alkyl electrophiles are usually used as radical donors in alkyl Heck reaction, and the undesired β-H elimination needs to be suppressed. Herein, we report a β-hydrogen elimination-enabled PdH-catalyzed hydrofunctionalization of alkyl halides with gem-fluoroalkenes or N-tosylhydrazones under visible-light irradiation. β-H elimination of the alkyl-palladium intermediates (generated from alkyl halide partners) access catalytic Pd–H species and alkenes. This is followed by regioselective migratory insertion of gem-fluoroalkene or carbene into the Pd–H bond and subsequent alkyl Heck reaction with the in situ generated alkene (from alkyl halide) to afford the final hydroalkenylation or hydroalkylation product. These products can undergo a variety of valuable transformations. Notably, two drug compounds, flobufen and imazodan, can be synthesized to demonstrate the utility of this protocol. Experimental and computational studies provide insight into the reaction mechanism. Compared to systems using alkyl halides as radical precursors, this system switches alkyl halides from radical donors to acceptors (alkenes) and provides a strategy for the intermolecular hydrogen-transfer alkyl Heck reaction of alkyl halides.