Lu Wang

Carbon coating rigidifies the electric double layer on copper, promoting CO diffusion and enhancing its electroreduction to C ₂₊ products at industrial current densities, demonstrated in a kilowatt-level electrolyzer for green CO/CO ₂ valorization.

We introduce a graph neural network framework that integrates density functional theory (DFT) calculations and scanning tunneling microscopy experimental data to predict molecular self-assembly behavior on metal surfaces. We constructed a data set comprising 20 diverse aromatic precursor molecules and their corresponding assembled structures on Au, Ag, and Cu substrates. Leveraging DFT-derived descriptors, feature importance analysis identified the molecule–substrate interactions and interfacial charge transfer as the dominant factors governing assembly behavior. A modified graph attention network model, trained on this multisource data set, achieved exceptional predictive accuracy, which exceeded 95% in classifying molecular arrangements and attained a coefficient of determination (R²) of 0.985 for adsorption energy regression. The model’s generalizability was further validated by accurately classifying the self-assembled layers of three previously unseen molecules. This study establishes a machine learning framework that bridges computational and experimental insights, paving the way for the rational design of surface-supported functional nanostructures.

Multiscale modelling links theory and experiment, showing that mesoscale mass transport controls oxygen reduction on catalyst surfaces and that transport-governed local pH and active site density determine electrocatalytic activity.

We develop a multiscale theoretical model integrating atomic-scale density functional theory (DFT), microkinetic simulations, and continuum transport models to elucidate the mass transport effects on the kinetics of the OER on IrO₂ and RuO₂ catalysts, which bridges atomic-level insights with macroscopic electrochemical behavior. By coupling constant potential DFT simulations of reaction energetics with microkinetic equations and Nernst-Planck-Poisson transport physics, we resolve atomic-scale reaction pathways under experimentally relevant concentration gradients. During the reaction process, significant variations in the concentrations of H⁺, OH^-, and adsorbates are identified at the electrode surface under different potentials, leading to the local pH shifts and ion depletion at electrode interfaces, which fundamentally alter the rate-determining steps. The framework reproduces experimental Tafel slopes (39/146 mV/dec for RuO₂; 59/118 mV/dec for IrO₂) without empirical fitting, highlighting the critical role of ion concentration gradients in governing the reaction kinetics. Our unified framework emphasizes the mass transport imperative on revealing the reaction mechanisms in realistic electrochemical environments, offering a generalizable strategy for designing and evaluating the novel OER catalysts.

Two-dimensional (2D) indium selenide, with its low effective mass, high thermal velocity, and exceptional electronic mobility, is a promising semiconductor for surpassing silicon electronics, but grown films have not achieved performance comparable with that of exfoliated micrometer-scale flakes. We report a solid‒liquid‒solid strategy that converts amorphous indium selenide films into pure-phase, high-crystallinity indium selenide wafers by creating an indium-rich liquid interface and maintaining a strict 1:1 stoichiometric ratio of indium to selenium. The as-obtained indium selenide films exhibit exceptional uniformity, a pure phase, and a high crystallinity across an entire ~5-centimeter wafer. Transistor arrays based on the produced indium selenide wafers demonstrate outstanding electronic performance surpassing that of all 2D film-based devices, including an extremely high mobility (averaging as high as 287 square centimeters per volt-second) and a near-Boltzmann-limit subthreshold swing (averaging as low as 67 millivolts per decade) at room temperature.

Photocatalytic upcycling of waste polyolefins into value-added chemicals provides promise in plastic waste management and resource utilization. Previous works demonstrate that polyolefins can be converted into carboxylic acids, with CO₂ as the final oxidation product. It is still challenging to explore more transformation products, particularly mild-oxidation products such as alcohols, because of their instability compared with polymer substrates, which are prone to oxidation during catalytic reactions. In this work, we propose an efficient strategy to regulate the product type through precise control of radicals, intermediates, and reaction paths. Taking the commonly used photocatalyst C₃N₄ as an example, its major products are carboxylic acids and CO₂. When MoS₂ is introduced to construct a Z-scheme heterostructure, gas products are significantly reduced and alcohols appear with a high yield of 1358.8 μmol g_cat^-1 and a high selectivity up to 80.3%. This is primarily attributed to the presence of •OH radicals from oxygen reduction, acting a key role in alcohol formation while simultaneously suppressing the competing pathways oxygen to •O₂^- and ¹O₂, thus reducing the overoxidation products. The β-scission of the C-C bonds in the polymer chains generates intermediate alkyl species, followed by the combination with •OH to produce methanol, which is more energetically favorable for MoS₂/C₃N₄. In contrast, alkyl species couple with oxygen species to form formic acid, which is favorable for C₃N₄. This work provides new approaches for controlling the product types and offers new insights into the reaction pathways involved in polyolefin photorefinery.

A machine learning force field workflow efficiently accelerates ab initio molecular simulations and reveals the significance of solvent configurations in electrocatalytic simulations.