Thermodynamic calculations of Ellingham diagrams and the forming oxides have been performed relevant to the Pt-based alloys Pt₈₂Al₁₂M₆ (M = Cr, Hf, Pt, and Ta). The predicted Ellingham diagrams indicate that the elements Hf and Al are easy to oxidize, followed by Ta and Cr, while Pt is extremely difficult to oxidize. Oxidation experiments characterized by X-ray diffraction (XRD) and electron probe micro-analyzers verify the present thermodynamic predictions, showing that the best alloy with superior oxidation resistance is Pt₈₂Al₁₂Cr₆, followed by Pt₈₈Al₁₂ due to the formation of the dense and continuous α-Al₂O₃ scale on the surface of alloys; while the worse alloy is Pt₈₂Al₁₂Hf₆ followed by Pt₈₂Al₁₂Ta₆ due to drastic internal oxidation and the formation of deleterious HfO₂, AlTaO₄, and Ta₂O₅. The present work, combining computations with experimental verifications, provides a fundamental understanding and knowledgebase to develop Pt-based superalloys with superior oxidation resistance that can be used in ultra-high temperatures.

The atomic structures of solid-solid interfaces in materials are of fundamental importance for understanding the physical properties of interfacial materials, which is, however, difficult to determine both in experimental and theoretical approaches. New theoretical methodologies utilizing various global optimization algorithms and machine learning (ML) potentials have emerged in recent years, offering a promising approach to unraveling interfacial structures. In this review, we give a concise overview of state-of-the-art techniques employed in the studies of interfacial structures, e.g., ML-assisted phenomenological theory for the global search of interface structure (ML-interface). We also present a few applications of these methodologies.

Large-scale atomistic simulations of two-dimensional (2D) materials rely on highly accurate and efficient force fields. Here, we present a physics-infused machine learning framework that enables the efficient development and interpretability of interatomic interaction models for 2D materials. By considering the characteristics of chemical bonds and structural topology, we have devised a set of efficient descriptors. This enables accurate force field training using a small dataset. The machine learning force fields show great success in describing the phase transformation and domain switching behaviors of monolayer Group IV monochalcogenides, e.g., GeSe and PbTe. Notably, this type of force field can be readily extended to other non-transition 2D systems, such as hexagonal boron nitride (hBN), leveraging their structural similarity. Our work provides a straightforward but accurate extension of simulation time and length scales for 2D materials.

The Haber-Bosch (H-B) process, which is widely used in industry to synthesize ammonia, leads to serious energy and environment-related issues. The electrochemical nitrogen reduction reaction (eNRR) is the most promising candidate to replace H-B processes because it is more energy-efficient and environmentally friendly. Atomic-level catalysts, such as single-atom and double-atom catalysts (SACs and DACs), are of great interest due to their high atomic utilization and activity. The synergy between the metal atoms and two-dimensional (2D) support not only modulates the local electronic structure of the catalyst but also controls the catalytic performance. In this article, we explored the eNRR performance of 2D Fe₃@N_xC_20-x (x = 0~4), whose structure was based on the experimentally synthesized Ag₃@C₂₀ sheet, by means of density functional theory calculations. Through calculations, we found that the 2D Fe₃@N₄C₁₆ with Fe₂ site coordinated with four N is a promising eNRR catalyst: the limiting potential is as low as -0.45 V, and the competing hydrogen evolution reaction can be effectively suppressed. Our work not only confirms that the coordination environment of the metal site is crucial for the electrocatalytic activity but also provides a new guideline for designing low-cost eNRR catalysts with high efficiency.

The exploration of efficient electrocatalysts for carbon dioxide reduction reaction (CO₂RR) with viable activity and superior selectivity remains a great challenge. The efficiency of CO₂RR over traditional transition metal-based catalysts is restricted by their inherent scaling relationships, so breaking this scaling relationship is the key to improving the catalytic performance. In this work, inspired by the recent experimental progress in the synthesis of dual atom catalysts (DACs), we reported a rational design of novel DACs with two transition metal atoms embedded in defective MoS₂ with S vacancies for CO₂ reduction; 21 metal dimer systems were selected, including six homonuclear catalysts (MoS₂-M₂, M = Cu, Fe, Ni, Mn, Cr, Co) and 15 heteronuclear catalysts (MoS₂-M₁M₂). First-principles calculations showed that the MoS₂-NiCr system not only breaks the linear relationship of key intermediates but also possesses a low overpotential of 0.58 V and superior selectivity in the process of methane generation, which can be used as a promising catalyst for methane formation from CO₂ electroreduction. Notably, by combining random forest regression machine learning study, we found that the CO₂RR activity of DACs is essentially controlled by some fundamental factors, such as the distance between metal centers and the number of outer electrons in the metal atoms. Our findings provide profound insights for the design of efficient DACs for CO₂RR.

The electrocatalytic process of nitrogen reduction reactions (NRR) offers a promising approach towards achieving sustainable ammonia production, acting as an environmentally friendly replacement for the conventional Haber-Bosch method. Density functional theory calculations have been utilized to design and investigate a set of catalysts known as triple-atom catalysts (TACs) for electrochemical NRR, which are supported on graphite-C₃N₃ nanosheets. Herein, we have systematically evaluated these TACs using stringent screening to assess their catalytic performance. Among the candidates, supported Pt₃, Re₃, and Ru₃ trimers emerged as highly active with decent selectivity, involving a limiting potential range of -0.35~-0.11 V. According to analysis of electronic properties, we determined that high NRR activity stems from the d-π* electron-accepting and -donating mechanism. Significantly, the correlation between chemical activity of TACs and electronic structure was established as a pivotal physical parameter, which has led to the conclusion that we can precisely control the catalytic behavior of transition metal trimer clusters by selecting appropriate metal elements and designing moderate cluster-substrates interactions. In summary, these theoretical studies not only enhance our understanding of how catalytic properties are governed by metal-support interactions, regulating stability, activity, and selectivity, but also offer a useful method for screening and designing novel TACs for NRR.