Autonomous synthesis and characterization of inorganic materials requires the automatic and accurate analysis of X-ray diffraction spectra. For this task, we designed a probabilistic deep learning algorithm to identify complex multi-phase mixtures. At the core of this algorithm lies an ensemble convolutional neural network trained on simulated diffraction spectra, which are systematically augmented with physics-informed perturbations to account for artifacts that can arise during experimental sample preparation and synthesis. Larger perturbations associated with off-stoichiometry are also captured by supplementing the training set with hypothetical solid solutions. Spectra containing mixtures of materials are analyzed with a newly developed branching algorithm that utilizes the probabilistic nature of the neural network to explore suspected mixtures and identify the set of phases that maximize confidence in the prediction. Our model is benchmarked on simulated and experimentally measured diffraction spectra, showing exceptional performance with accuracies exceeding those given by previously reported methods based on profile matching and deep learning. We envision that the algorithm presented here may be integrated in experimental workflows to facilitate the high-throughput and autonomous discovery of inorganic materials.
MatDiffract: A Material-Informed Automated Analysis Platform for X-ray Powder Diffraction
Institute of High Energy Physics, Chinese Academy of Sciences（中国科学院高能物理研究所）；Spallation Neutron Source Science Center（散裂中子源科学中心）；College of Computer and Information Engineering, Henan Normal University（河南师范大学计算机与信息工程学院）；Dongguan Institute of Materials Science and Technology, Chinese Academy of Sciences（中国科学院东莞材料科学与技术研究所）；Songshan Lake Material Laboratory（松山湖材料实验室）；School of Computer Science, Sun Yat-sen University（中山大学计算机学院）
opXRD: Open Experimental Powder X-ray Diffraction Database
Karlsruhe Institute of Technology (KIT)（卡尔斯鲁厄理工学院）；Hong Kong University of Science and Technology (Guangzhou) (HKUST)（香港科技大学（广州））；University of Southern California (USC)（南加州大学）；Lawrence Berkeley National Laboratory (LBNL)（劳伦斯伯克利国家实验室）；Shenzhen Polytechnic（深圳职业技术大学）；King Fahd University of Petroleum and Minerals (KFUPM)（法赫德国王石油矿产大学）；Chimie ParisTech（巴黎化学学院）；PSL University（巴黎文理大学）；CNRS（法国国家科学研究中心）；Institut de Recherche de Chimie Paris（巴黎化学研究所）；Empa–Swiss Federal Laboratories for Materials Science and Technology (EMPA)（瑞士联邦材料科学与技术实验室）；Forschungszentrum Juelich GmbH（于利希研究中心）；Engler-Bunte-Institut & Institute of Catalysis Research and Technology（恩格勒-本生研究所及催化研究与技术研究所）
4DMulti: automated multicomponent identification at complex material interfaces
Global Institute of Future Technology, Shanghai Jiao Tong University（上海交通大学未来技术学院）；Global College, Shanghai Jiao Tong University（上海交通大学国际学院）；School of Materials Science and Engineering, Shanghai Jiao Tong University（上海交通大学材料科学与工程学院）
Automated Phase Mapping with AgileFD and its Application to Light Absorber Discovery in the V-Mn-Nb Oxide System
快速构建相图是组合材料科学的核心原则，加速材料发现的努力往往受到解释组合X射线衍射数据集的挑战的阻碍，我们通过开发AgileFD来解决这个问题，AgileFD是一种人工智能算法，可以从X射线衍射图案的组合库中快速绘制相位图。AgileFD通过卷积非负矩阵分解的新扩展对基于合金的峰位移进行建模，这不仅改善了组成相的识别，而且还将其浓度和晶格参数映射为组成的函数。通过将Gibbs相位规则引入算法，在无监督操作的情况下得到有物理意义的相位图，并通过注入系统的专家知识得到更精细的解。该算法是证明通过调查的V-Mn-Nb氧化物系统，其中分解的8个氧化物相，包括两个大量的合金化，提供了第一个相图为这个伪三元系统。该相图能够解释高通量带隙数据，从而发现新的太阳光吸收剂和基于合金的MnV 2 O 6直接允许带隙能量的调谐。AgileFD算法的开源系列可以在广泛的高通量工作流程中实施，以加速材料发现。
PowderLine: a programmatic powder diffraction analysis application
National Synchrotron Light Source II, Brookhaven National Laboratory（美国国家同步光源二号，布鲁海文国家实验室）；Materials Science and Engineering, University of Florida（佛罗里达大学材料科学与工程）
Disentangling multiple scattering with deep learning: application to strain mapping from electron diffraction patterns
Center for Nanoscale Materials, Argonne National Laboratory（阿贡国家实验室纳米尺度材料中心）；National Center for Electron Microscopy, Molecular Foundry, Lawrence Berkeley National Laboratory（劳伦斯伯克利国家实验室国家电子显微镜中心（分子工厂））；Department of Materials Science and Engineering, University of California, Berkeley（加利福尼亚大学伯克利分校材料科学与工程系）；Scientific Data Division, Lawrence Berkeley National Laboratory（劳伦斯伯克利国家实验室科学数据部）
AI-Driven Structure Refinement of X-ray Diffraction
The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；Shanghai University（上海大学）；University of Science and Technology of China（中国科学技术大学）
