No abstract

No abstract

Van der Waals epitaxy on the surface of two-dimensional (2D) layered crystals has gained significant research interest for the assembly of well-ordered nanostructures and fabrication of vertical heterostructures based on 2D crystals. Although van der Waals epitaxial assembly on the hexagonal phase of transition metal dichalcogenides (TMDCs) has been relatively well characterized, a comparable study on the distorted octahedral phase (1T′ or Td) of TMDCs is largely lacking. Here, we investigate the assembly behavior of one-dimensional (1D) AgCN microwires on various distorted TMDC crystals, namely 1T′-MoTe2, Td-WTe2, and 1T′-ReS2. The unidirectional alignment of AgCN chains is observed on these crystals, reflecting the symmetry of underlying distorted TMDCs. Polarized Raman spectroscopy and transmission electron microscopy directly confirm that AgCN chains display the remarkable alignment behavior along the distorted chain directions of underlying TMDCs. The observed unidirectional assembly behavior can be attributed to the favorable adsorption configurations of 1D chains along the substrate distortion, which is supported by our theoretical calculations and observation of similar assembly behavior from different cyanide chains. The aligned AgCN microwires can be harnessed as facile markers to identify polymorphs and crystal orientations of TMDCs.

Scanning transmission electron microscopy (STEM) is an indispensable tool for atomicresolution structural analysis for a wide range of materials. The conventional analysis of STEM images is an extensive hands-on process, which limits efficient handling of highthroughput data. Here we apply a fully convolutional network (FCN) for identification of important structural features of two-dimensional crystals. ResUNet, a type of FCN, is utilized in identifying sulfur vacancies and polymorph types of MoS2 from atomic resolution STEM images. Efficient models are achieved based on training with simulated images in the presence of different levels of noise, aberrations, and carbon contamination. The accuracy of the FCN models toward extensive experimental STEM images is comparable to that of careful hands-on analysis. Our work provides a guideline on best practices to train a deep learning model for STEM image analysis and demonstrates FCN's application for efficient processing of a large volume of STEM data.