Huang Wenqiang, Jin Yucheng, Li Zhemin, Yao Lin, Chen Yun, Luo Zheng, Zhou Shen, Lin Jinguo, Liu Feng, Gao Zhifeng, Cheng Jun, Zhang Linfeng, Ouyang Fangping, Zhang Jin, Wang Shanshan
School of Advanced Materials, Peking University Shenzhen Graduate School, Shenzhen, China.
Department of Materials Science and Engineering, College of Aerospace Science and Engineering, National University of Defense Technology, Changsha, China.
Nat Commun. 2025 Mar 25;16(1):2927. doi: 10.1038/s41467-025-58160-3.
The high-resolution visualization of atomic structures is significant for understanding the relationship between the microscopic configurations and macroscopic properties of materials. However, a rapid, accurate, and robust approach to automatically resolve complex patterns in atomic-resolution microscopy remains difficult to implement. Here, we present a Trident strategy-enhanced disentangled representation learning method (a generative model), which utilizes a few unlabelled experimental images with abundant low-cost simulated images to generate a large corpus of annotated simulation data that closely resembles experimental results, producing a high-quality large-volume training dataset. A structural inference model is then trained via a residual neural network which can directly deduce the interlayer slip and rotation of diversified and complicated stacking patterns at van der Waals (vdW) interfaces with picometer-scale accuracy across various materials (e.g. MoS, WS, ReS, ReSe, and 1 T'-MoTe) with different layer numbers (bilayer and trilayers), demonstrating robustness to defects, imaging quality, and surface contaminations. The framework can also identify pattern transition interfaces, quantify subtle motif variations, and discriminate moiré patterns that are difficult to distinguish in frequency domains. Finally, the high-throughput processing ability of our method provides insights into a vdW epitaxy mode where various thermodynamically favorable slip stackings can coexist.
原子结构的高分辨率可视化对于理解材料微观结构与宏观性质之间的关系具有重要意义。然而，一种快速、准确且稳健的方法来自动解析原子分辨率显微镜中的复杂图案仍然难以实现。在此，我们提出一种三叉戟策略增强的解缠表征学习方法（一种生成模型），它利用少量未标记的实验图像与大量低成本的模拟图像来生成大量与实验结果极为相似的带注释模拟数据，从而产生高质量的大容量训练数据集。然后通过残差神经网络训练一个结构推理模型，该模型能够以皮米级精度直接推断范德华（vdW）界面处各种材料（如MoS、WS、ReS、ReSe和1 T'-MoTe）不同层数（双层和三层）的多样且复杂堆叠图案的层间滑移和旋转，展现出对缺陷、成像质量和表面污染的鲁棒性。该框架还能够识别图案转变界面、量化细微的 motif 变化，并区分在频域中难以区分的莫尔条纹图案。最后，我们方法的高通量处理能力为一种vdW外延模式提供了见解，在这种模式下各种热力学上有利的滑移堆叠可以共存。
