Abstract
Here we report a facile, prompt protocol based on deep-learning techniques to sort out intricate phase identification and quantification problems in complex multiphase inorganic compounds. We simulate plausible powder X-ray diffraction (XRD) patterns for 170 inorganic compounds in the Sr-Li-Al-O quaternary compositional pool, wherein promising LED phosphors have been recently discovered. Finally, 1,785,405 synthetic XRD patterns are prepared by combinatorically mixing the simulated powder XRD patterns of 170 inorganic compounds. Convolutional neural network (CNN) models are built and eventually trained using this large prepared dataset. The fully trained CNN model promptly and accurately identifies the constituent phases in complex multiphase inorganic compounds. Although the CNN is trained using the simulated XRD data, a test with real experimental XRD data returns an accuracy of nearly 100% for phase identification and 86% for three-step-phase-fraction quantification.
Highlights
We report a facile, prompt protocol based on deep-learning techniques to sort out intricate phase identification and quantification problems in complex multiphase inorganic compounds
Rather than the previous symmetry classification for a single-phase compound, we introduce a more pragmatic Convolutional neural network (CNN) model to sort out phase identification for unknown multiphase mixture powder samples consisting of several phases, which are more frequently confronted by materials scientists and engineers during ordinary materials research activities
We are dealing with large datasets (800,942 or 183,521 X-ray diffraction (XRD) patterns per a dataset), deep architecture, and the prevention of handcrafted data reduction, whereas most of the machine learning (ML) approaches remain restricted within a somewhat vicious circle consisting of small-sized training datasets, shallow artificial neural network architectures, and the excessive feature engineering that is based on human knowledge, all of which impart ML-based analyses with no merit by comparison with rule-based analysis
Summary
We report a facile, prompt protocol based on deep-learning techniques to sort out intricate phase identification and quantification problems in complex multiphase inorganic compounds. One of the most frequently faced situations in the process of materials discovery based on the powder XRD technique involves the identification and quantification of unknown multiphase compounds It would be arduous, for even a well-trained expert with the advantage of wellestablished computational tools to complete both the constituent phase identification and the ensuing phase-fraction estimation for a sample consisting of a grungy, multiphase mixture. Simulated XRD patterns for 150,000 entries registered in the inorganic compound structure database (ICSD) were used to train the CNN model This previous report is appreciated as one of the early-stage deep-learning approaches in the crystallography research field as evidenced by the ensuing reviews on deep learning for crystal structure prediction[11]. We are dealing with large datasets (800,942 or 183,521 XRD patterns per a dataset), deep architecture, and the prevention of handcrafted data reduction, whereas most of the ML approaches (even alleged as deep learning) remain restricted within a somewhat vicious circle consisting of small-sized training datasets, shallow artificial neural network architectures, and the excessive feature engineering that is based on human knowledge, all of which impart ML-based analyses with no merit by comparison with rule-based analysis
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.