A deep convolutional neural network for real-time full profile analysis of big powder diffraction data

We present Parameter Quantification Network (PQ-Net), a regression deep convolutional neural network providing quantitative analysis of powder X-ray diffraction patterns from multi-phase systems. The network is tested against simulated and experimental datasets of increasing complexity with the last...

Full description

Saved in:
Bibliographic Details
Published in:npj computational materials 2021-05, Vol.7 (1), p.1-9, Article 74
Main Authors: Dong, Hongyang, Butler, Keith T., Matras, Dorota, Price, Stephen W. T., Odarchenko, Yaroslav, Khatry, Rahul, Thompson, Andrew, Middelkoop, Vesna, Jacques, Simon D. M., Beale, Andrew M., Vamvakeros, Antonis
Format: Article
Language:English
Subjects:
639/301
639/638
Aluminum oxide
Artificial neural networks
Cerium oxides
Characterization and Evaluation of Materials
Chemistry and Materials Science
Computational Intelligence
Computed tomography
Crystallites
Crystals
Data analysis
Datasets
Diffraction patterns
Materials Science
Mathematical and Computational Engineering
Mathematical and Computational Physics
Mathematical Modeling and Industrial Mathematics
Multiphase
Neural networks
Parameters
Real time
Rietveld method
Theoretical
X ray powder diffraction
X-ray diffraction
Zirconium dioxide
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present Parameter Quantification Network (PQ-Net), a regression deep convolutional neural network providing quantitative analysis of powder X-ray diffraction patterns from multi-phase systems. The network is tested against simulated and experimental datasets of increasing complexity with the last one being an X-ray diffraction computed tomography dataset of a multi-phase Ni-Pd/CeO 2 -ZrO 2 /Al 2 O 3 catalytic material system consisting of ca. 20,000 diffraction patterns. It is shown that the network predicts accurate scale factor, lattice parameter and crystallite size maps for all phases, which are comparable to those obtained through full profile analysis using the Rietveld method, also providing a reliable uncertainty measure on the results. The main advantage of PQ-Net is its ability to yield these results orders of magnitude faster showing its potential as a tool for real-time diffraction data analysis during in situ/operando experiments.
ISSN:2057-3960
2057-3960
DOI:10.1038/s41524-021-00542-4