AI 中文总结
研究如何将漫散射与原子位点相关信息联系起来，核心方法是引入MOSAIC计算框架，通过对含相位散射振幅应用傅里叶滤波器获取相关图谱，主要贡献是能有效探究大规模结构模型中原子间相关性。
AI 中文摘要
局部结构相关性编码在漫散射中，但识别产生特定漫散射特征的原子基序具有挑战性。我们引入了MOSAIC，这是一个在有原子构型且能计算含相位散射振幅时用于此任务的计算框架。我们的方法是在包含感兴趣散射特征的倒易空间区域对该振幅应用傅里叶滤波器，以获得负责这些特征的原子位移和位点占有率图。该方法在探究大规模结构模型中原子间相关性的性质和空间分布方面很有效，如通过实验数据的反向蒙特卡罗精修、分子动力学或蒙特卡罗模拟得到的模型，或原子分辨率电子显微镜图像的二维结构投影。
英文摘要
Local structural correlations are encoded in diffuse scattering, but identifying atomic motifs that produce specific diffuse features can be challenging. We introduce MOSAIC, a computational framework for this task when an atomistic configuration is available and a phase-bearing scattering amplitude can be calculated. Our approach relies on applying the Fourier filter to this amplitude over the reciprocal-space regions encompassing the scattering features of interest to obtain maps of atomic displacements and site occupancies responsible for those features. The method is effective in interrogating the nature and spatial distributions of interatomic correlations in large-scale structural models, such as obtained using Reverse Monte Carlo refinements from experimental data, molecular dynamics, or Monte Carlo simulations, or 2D structural projections derived from atomic-resolution electron microscopy images.
Comments19 pages, 7 figures
