发表机构
University of Washington（华盛顿大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
通过微调通用原子模型并利用知识蒸馏，开发出MoiréMLIP及其精简版，实现百万原子莫尔晶格的高精度、可迁移模拟，并揭示交替扭转MoTe2中的层级极化网络。
AI 中文摘要
原子重构在多个尺度上重塑莫尔材料，从局部结构和极化纹理到全局电子拓扑，然而对于大超结构（如边缘扭转角莫尔和莫尔中的莫尔）的直接从头算建模变得不可行。我们通过在密度泛函理论标注的莫尔万花筒数据集上微调一个通用原子模型，开发了MoiréMLIP，该数据集涵盖过渡金属二硫化物组成、对称性、堆叠和扭转角。MoiréMLIP以6–8 meV/Å的力误差再现从头算重构，并可迁移到更小的扭转角和未见结构。知识蒸馏得到MoiréMLIP-mini，在保持精度的同时将单GPU推理扩展到一百万个原子。应用于交替扭转的MoTe2三层，它揭示了一个跨越数十纳米的层级极化网络，该网络源于沿莫尔中的莫尔畴壁尖锐局域化的大莫尔晶胞畸变。这些结果克服了现有原子模型在精度、可迁移性和扩展性方面的关键瓶颈，使得跨新兴莫尔长度尺度的预测模拟成为可能。
英文摘要
Atomic reconstruction reshapes moiré materials across multiple scales, from local structure and polarization textures to global electronic topology, yet direct \textit{ab initio} modeling becomes prohibitive for large superstructures such as marginal-twist-angle moirés and moiré-of-moirés. We develop MoiréMLIP by fine-tuning a universal atomistic model on the density functional theory labeled Moiré Kaleidoscope dataset, which spans transition metal dichalcogenide compositions, symmetries, stackings, and twist angles. MoiréMLIP reproduces \textit{ab initio} reconstruction with force errors of 6--8\,meV/Å and transfers to smaller twist angles and unseen structures. Knowledge distillation yields MoiréMLIP-mini, which retains this accuracy while extending single-GPU inference to one million atoms. Applied to an alternate-twist MoTe$_2$ trilayer, it reveals a hierarchical polarization network spanning tens of nanometers arising from large moiré-cell distortions sharply localized along the moiré-of-moiré domain walls. These results overcome key accuracy, transferability, and scaling bottlenecks of existing atomistic models and enable predictive simulations across emergent moiré length scales.
Comments14 pages, 6 figures, 5 tables
