Morphology significantly affects material's electronic, catalytic, and magnetic properties, especially for 2D crystals. Abundant achievements have been made in the morphology engineering of high-symmetry 2D materials, but for the emerging low-symmetry ones, such as ReS 2 , both the morphology control technique and comprehension are lacking. Here, the lateral shape and vertical thickness engineering of 2D ReS 2 by tailoring the growth temperature and the substrate symmetry using chemical vapor deposition, is reported. The temperature increase induces an isotropic-to-anisotropic transition of domain shapes, as well as a monotonic decrease of the domain thickness, which promotes the electrocatalytic performance. The substrate rotational symmetry determines the shape anisotropy of polycrystalline ReS 2 monolayers via a diffusion-limited mechanism, leading to highly oriented square, triangular, and strip-like domains synthesized on the fourfold symmetry SrTiO 3 (001), threefold symmetry c-sapphire, and twofold symmetry a-sapphire substrates, respectively. Various stacking configurations in bilayers are unclosed at the atomic scale. Some are predicted to adopt a type-II band alignment with great potential in photovoltaics. The results give insights into the morphological engineering of a unique class of 2D material with low in-plane lattice symmetry and weak interlayer coupling, which are crucial for their high-quality synthesis and industrial applications.

Alloying is widely applied to tailor properties of 2D materials. Herein, a space‐confined chemical vapor deposition (CVD) strategy to homogeneously grow 100 μm‐sized monolayer 1T’‐MoTe2 in batches is developed. Aberration‐corrected annular dark‐field scanning transmission electron microscopy combined with density functional theory calculations is applied to investigate the atomic structural alteration of 1T’‐MoTe2 alloyed with sulfur. 1T’‐to‐2H phase transition is observed, triggered by both thermodynamic (stability improvement) and kinetic (phase transition barrier reduction) reasons. The alloying degrees of MoS2xTe2(1−x) grown at different temperatures display homophilic and random configurations, respectively, providing a feasible approach to tailor the atomic mixture of the alloying elements. The intralayer reconstruction and the interlayer displacement take place as defect concentration increases at elevated alloying temperatures. In contrast, 1T’‐MoTe2 displays high susceptibility in the presence of oxygen. An encapsulation method is developed using CVD‐grown monolayer MoS2, extending the lifetime of monolayer tellurides from several minutes to at least 24 h. These results gain fundamental insights into the structural change in both the beneficial (sulfurization) and detrimental (oxidation) alloying processes of atomically thin 1T’‐MoTe2 and provide a new degree of freedom for the controllable growth of 2D alloys (i.e., tailoring the alloying degree).

Developing promoters that can boost the growth quality, efficiency, and robustness of two-dimensional (2D) transition metal dichalcogenides is significant for their industrial applications. Herein, a new group (group IIA) of...

While a growing body of literature has been studying new Graph Neural Networks (GNNs) that work on both homophilic and heterophilic graphs, little has been done on adapting classical GNNs to less-homophilic graphs. Although the ability to handle less-homophilic graphs is restricted, classical GNNs still stand out in several nice properties such as efficiency, simplicity, and explainability. In this work, we propose a novel graph restructuring method that can be integrated into any type of GNNs, including classical GNNs, to leverage the benefits of existing GNNs while alleviating their limitations. Our contribution is threefold: a) learning the weight of pseudo-eigenvectors for an adaptive spectral clustering that aligns well with known node labels, b) proposing a new density-aware homophilic metric that is robust to label imbalance, and c) reconstructing the adjacency matrix based on the result of adaptive spectral clustering to maximize the homophilic scores. The experimental results show that our graph restructuring method can significantly boost the performance of six classical GNNs by an average of 25% on less-homophilic graphs. The boosted performance is comparable to state-of-the-art methods.

No abstract

Developing a convenient method to determine the complete structure of single‐walled carbon nanotubes (SWNTs) is important to achieve the fully controlled growth of this nanomaterial. However, approaches that can identify handedness at the atomic level with simple equipment, operation, and data analysis are still lacking. Here, the SWNTs/graphene (Gr) vertical heterostructures are artificially constructed with aligned interfaces to realize the lattice interpretation of SWNT upper and lower walls separately by only one transmission electron microscopy image, thus transforming the 3D handedness information to projected 2D space. Gr displays prominent out‐of‐plane deformation at the interface, promoting the energetic advantage for the aligned interface construction. The interfacial alignment between the SWNT and Gr shows no obvious dependence on either the helical angle or diameter of SWNTs. The half‐wrapping of SWNTs by deformed Gr also triggers diversified alterations in electronic structures based on theoretical calculations. 27 specimens with SWNTs prepared by two disparate methods are examined, implying equal handedness distribution in the randomly aligned SWNTs grown on quartz and potential handedness enrichment in horizontal SWNT arrays grown on a‐sapphire. This work provides a simple strategy for chiral discrimination and lays a characterization foundation for handedness‐selective growth of nanomaterials.

While a growing body of literature has been studying new Graph Neural Networks (GNNs) that work on both homophilic and heterophilic graphs, little work has been done on adapting classical GNNs to less-homophilic graphs. Although lacking the ability to work with less-homophilic graphs, classical GNNs still stand out in some properties such as efficiency, simplicity and explainability. We propose a novel graph restructuring method to maximize the benefit of prevalent GNNs with the homophilic assumption. Our contribution is threefold: a) learning the weight of pseudo-eigenvectors for an adaptive spectral clustering that aligns well with known node labels, b) proposing a new homophilic metric that measures how two nodes with the same label are likely to be connected, and c) reconstructing the adjacency matrix based on the result of adaptive spectral clustering to maximize the homophilic scores. The experimental results show that our graph restructuring method can significantly boost the performance of six classical GNNs by an average of 25% on less-homophilic graphs. The boosted performance is comparable to state-of-the-art methods.