编委Ceder G发明：一种新型高效材料相态检测方法

X射线衍射技术是一种常用的材料学特征分析方法，对于材料的相态鉴定十分重要。但是，传统的X射线衍射技术需要手动选取参数并进行扫描，效率较低，而且一些材料可能会因为相变或者形变而改变衍射谱，造成误判。因此，研究如何使得X射线衍射技术更智能化、高效化，对于提高材料相态鉴定的精度和速度具有重要意义。

 

Fig. 1 A schematic for adaptively driven XRD with autonomous phase identification

由美国加州大学伯克利分校材料科学与工程系的Gerbrand Ceder教授（本刊编委）领导的团队，在基于卷积神经网络的ML算法的驱动下，制定了一种用于自主相位识别的自适应引导XRD技术。不确定性量化用于决定何时需要额外的测量，而类激活映射分析则表明在何处执行这些测量。

 

Fig. 2 F1-scores achieved by XRD-AutoAnalyzer when applied to simulated patterns in the (top) Li-La-Zr-O and (bottom) Li-Ti-P-O spaces. 

基于Li-La-Zr-O和Li-Ti-P-O化学空间的材料，该方法在三个复杂程度不断增加的测试案例中得到了验证。这些测试表明，自适应XRD在模拟和实验获得的图像上始终优于传统方法，并可以提供更精确的杂质相检测，同时测量时间更短。作者进一步证明，该ML方法可以有效地指导XRD测量，以Li7La3Zr2O12（LLZO）的合成为例，改善固相反应的原位表征。使用自适应扫描来监测LLZO的合成，成功地识别了一个短寿命的中间相，而传统测量通常漏掉了这种中间相的表征。

  Fig. 3 Detection rates and measurement times required for impurity detection 

这些发现为动态过程的自适应表征提供了一个清晰的概念证明，突出了由ML驱动的自主实验的机会。该文近期发表于npj Computational Materials  9: 31 (2022).

Fig. 4 In situ identification of phases formed during LLZO synthesis. 

Editorial Summary

A novel and efficient method for phase detection of materials

X-ray diffraction technique is a commonly used material science characterization method, which is important for phase identification of materials. However, the traditional X-ray diffraction technique requires manual selection of parameters and scanning, which is inefficient, and some materials may change the diffraction spectrum due to phase change or deformation, resulting in misclassification. Therefore, it is important to study how to make the X-ray diffraction technique more intelligent and efficient to improve the accuracy and speed of material phase identification.  A team lead by Prof. Gerbrand Ceder from Department of Materials Science & Engineering, UC Berkeley, USA, formulated an adaptively steered XRD technique for autonomous phase identification, driven by an ML algorithm based on a convolutional neural network. Uncertainty quantification is used to decide when additional measurements are needed, while class activation map analysis dictates where those measurements are performed. This approach is validated it on three test cases with increasing complexity based on materials from the Li-La-Zr-O and Li-Ti-P-O chemical spaces. These tests reveal that adaptive XRD consistently outperforms conventional methods on both simulated and experimentally acquired patterns, providing more precise detection of impurity phases while requiring shorter measurement times. The authors further demonstrate that the ML approach can effectively guide XRD measurements for improved in situ characterization of solid-state reactions, with the synthesis of Li7La3Zr2O12 (LLZO) considered as an example. The use of adaptive scans to monitor LLZO synthesis led to the successful identification of a short-lived intermediate phase that would otherwise be missed by conventional measurements.  These findings provide a clear proof of concept for adaptive characterization of dynamic processes, highlighting the opportunity for autonomous experiments driven by ML. This article was recently published in npj Computational Materials 9: 31 (2022). 

原文Abstract及其翻译 

Adaptively driven X-ray diffraction guided by machine learning for autonomous phase identification (由机器学习指导的自适应驱动X射线衍射的自主相位识别) Nathan J. Szymanski, Christopher J. Bartel, Yan Zeng, Mouhamad Diallo, Haegyeom Kim & Gerbrand Ceder  Abstract Machine learning (ML) has become a valuable tool to assist and improve materials characterization, enabling automated interpretation of experimental results with techniques such as X-ray diffraction (XRD) and electron microscopy. Because ML models are fast once trained, there is a key opportunity to bring interpretation in-line with experiments and make on-the-fly decisions to achieve optimal measurement effectiveness, which creates broad opportunities for rapid learning and information extraction from experiments. Here, we demonstrate such a capability with the development of autonomous and adaptive XRD. By coupling an ML algorithm with a physical diffractometer, this method integrates diffraction and analysis such that early experimental information is leveraged to steer measurements toward features that improve the confidence of a model trained to identify crystalline phases. We validate the effectiveness of an adaptive approach by showing that ML-driven XRD can accurately detect trace amounts of materials in multi-phase mixtures with short measurement times. The improved speed of phase detection also enables in situ identification of short-lived intermediate phases formed during solid-state reactions using a standard in-house diffractometer. Our findings showcase the advantages of in-line ML for materials characterization and point to the possibility of more general approaches for adaptive experimentation. 

摘要 机器学习（ML）已成为协助和改善材料表征的宝贵工具，能够通过X射线衍射(XRD)和电子显微镜等技术自动解释实验结果。由于ML模型一旦训练好就会很快，因此有一个关键机会可将解释与实验结合起来，并即时做出决策以实现最佳测量效果，这为快速学习和从实验中提取信息创造了广泛的机会。在这里，我们通过开发自主和自适应XRD来证明这种能力。通过将ML算法与物理衍射仪耦合，该方法集成了衍射和分析过程，以便利用早期的实验信息来引导测量，最终提高训练以识别晶体相模型的置信度。我们通过证明ML驱动的XRD可以在较短的测量时间内准确检测出多相混合物中的微量材料，验证了自适应方法的有效性。相检测速度的提高也使我们能够使用标准的内部衍射仪在固态反应中形成的短寿命中间相进行原位识别。我们的研究结果展示了内联ML在材料表征方面的优势，并指出了更通用的自适应实验方法的可能性。