I present a strategy for unsupervised manifold learning on local atomic environments in molecular simulations based on simple rotation- and permutation-invariant three-body features. These features are highly descriptive, generalize to multiple chemical species, and are human-interpretable. The low-dimensional embeddings of each atomic environment can be used to understand and quantify messy crystal structures such as those near interfaces and defects or well-ordered crystal lattices such as in bulk materials without modification. The same method can also yield collective variables describing collections of particles such as for an entire simulation domain. I demonstrate the method on colloidal crystallization, ice crystals, and binary mesophases to illustrate its broad applicability. In each case, the learned latent space yields insights into the details of the observed microstructures. For ices and mesophases, supervised classifiers are trained based on the learned manifolds and directly compared against a recent neural-network-based approach. Notably, while this method provides comparable classification performance, it can also be deployed on even a handful of observed environments without labels or \textit{a priori} knowledge. Thus, the current approach provides an incredibly versatile strategy to characterize and classify local atomic environments, and may unlock insights in a wide variety of molecular simulation contexts.
From structure mining to unsupervised exploration of atomic octahedral networks
Department of Engineering, University of Cambridge(剑桥大学工程系);Department of Chemical and Biological Engineering, University of Colorado Boulder(科罗拉多大学波尔得分校化学与生物工程系);The Institute of Chemistry, Casali Center for Applied Chemistry, and the Center for Nanoscience and Nanotechnology, The Hebrew University of Jerusalem(耶路撒冷希伯来大学化学系、应用化学Casali中心和纳米科学与技术中心);Department of Chemistry and Biochemistry, University of South Carolina(南卡罗来纳大学化学与生物化学系)
Crystal Nucleation in Al-Ni Alloys: an Unsupervised Chemical and Topological Learning Approach
结晶是材料凝固的基本过程,决定了材料的微观结构。在原子尺度上的复杂现象的驱动下,它对合金的理解仍然是难以捉摸的。采用大尺度分子动力学模拟方法研究了典型过冷Al-Ni二元合金的均匀形核途径。一个无监督的拓扑学习分析表明,成核集首先从化学排序,其次是键取向排序的基础晶相。我们的研究结果还表明,不同的多晶型物的选择,这取决于组合物。Al 50 Ni 50的形核过程是一个单一的台阶,B2短程有序的出现,而Al 25 Ni 75的形核过程是一个向L12相的台阶形核过程。进一步讨论了纯Al和纯Ni对应物形核的影响。
Unsupervised landmark analysis for jump detection in molecular dynamics simulations
分子动力学是研究固态离子导体中扩散的一种通用而强大的方法,它只需要系统自由能表面的平衡态或过渡态的最小先验知识。然而,相关的,但罕见的事件,如跳跃的扩散移动离子的轨迹的分析,仍然是相当麻烦的,需要先验知识的扩散过程,以获得有意义的结果。在这项工作中,我们提出了一种新的方法来检测扩散系统中的相关事件,而无需假设有关的基础过程的先验信息。我们开始从一个里程碑的基础上,以确定在移动离子的环境中的主导功能的原子坐标的投影。随后在界标空间中的聚类使得能够将任何轨迹离散化为一系列不同的状态。作为最后一步,使用原子位置描述符的平滑重叠允许以直接的方式区分不同的环境。我们将该算法应用于十个锂离子体系,并对立方Li $_{7}$ La $_{3}$ Zr $_{2}$ O $_{12}$,四元Li $_{10}$ GeP $_{2}$ S $_{12}$,和$β$ -锂霞石LiAlSiO $_{4}$进行了深入分析。我们将我们的结果与现有的方法进行比较,强调优点,缺点,并深入研究材料中离子传导的扩散行为。
LATTE: an atomic environment descriptor based on Cartesian tensor contractions
Scuola Internazionale Superiore di Studi Avanzati(国际高等研究学院);CNR-IOM DEMOCRITOS, Istituto Officina dei Materiali(国家研究委员会材料研究所DEMOCRITOS项目组);NVIDIA Corporation(英伟达公司)