化学性质不同的二维范德华双层中层层间结合与电荷再分布的扩散量子蒙特卡洛基准

中文摘要 AI 辅助

二维材料中的层间相互作用可以产生其组成单层中不存在的涌现现象，包括非常规磁序、多铁性和拓扑磁相。预测此类涌现行为需要同时解析长程色散、短程轨道杂化和电子关联相互作用，这些相互作用本质上是非局域且多体的，即使对于先进的密度泛函近似方法也仍然具有挑战性。在此，我们利用扩散蒙特卡洛（DMC）方法，为多种双层材料建立了一个系统可控的多体基准，涵盖单层材料、过渡金属二硫属化物和磁性过渡金属卤化物。我们获得了平衡间距、结合能学、层间振动模式和电荷再分布，与现有实验数据高度吻合，同时揭示了广泛使用的半局域、meta-GGA 和色散校正密度泛函中系统性且依赖性质的失效。超越能量学，DMC 解析了细微的层间电荷重排，特别是在磁性 Cr 三卤化物中产生长程偶极尾，这些尾在远超半局域 DFT 预测显著层间极化的区域仍然存在，为理解层间耦合铁电和磁现象提供了多体基础。由此产生的能量、响应特性和高精度电子密度构成了开发下一代密度泛函的可迁移基准。最后，我们提供了一个可扩展的高性能计算工作流程，使科学界能够系统性地扩展新兴层状量子材料家族的多体基准数据集。

英文摘要

Interlayer interactions in two-dimensional materials can generate emergent phenomena absent in their constituent monolayers, including unconventional magnetic order, multiferroicity, and topological magnetic phases. Predicting such emergent behavior requires simultaneously resolving long-range dispersion, short-range orbital hybridization, and electronic correlation interactions that are intrinsically nonlocal and many-body and remain challenging even for advanced density-functional approximations. Here we establish a systematically controlled many-body benchmark for diverse bilayer materials using diffusion Monte Carlo (DMC), spanning single-sheet materials, transition-metal dichalcogenides, and magnetic transition-metal halides. We obtain equilibrium separations, binding energetics, interlayer vibrational modes, and charge redistribution, finding excellent agreement with available experiments while revealing systematic and property-dependent failures across widely used semilocal, meta-GGA, and dispersion-corrected density functionals. Beyond energetics, DMC resolves subtle interlayer charge rearrangements, particularly leading to long-ranged dipolar tails in magnetic Cr trihalides that survive well beyond the regime where semilocal DFT predicts appreciable interlayer polarization, providing a many-body basis for understanding interlayer-coupled ferroic and magnetic phenomena. The resulting energies, response properties, and high-accuracy electron densities constitute a transferable benchmark for developing next-generation density functionals. Finally, we provide a scalable high-performance-computing workflow that enables systematic expansion of many-body benchmark datasets across emerging families of layered quantum materials by the scientific community.

发表机构

• Oak Ridge National Laboratory（橡树岭国家实验室）
• Argonne National Laboratory（阿贡国家实验室）

机构由 AI 辅助整理，请以论文原文为准。