A deep convolutional neural network for real-time analysis of big powder diffraction data Hongyang Dong 1 , K. Butler 2 , R. Khatry 3 , S. D. M. Jacques 4 , A. M. Beale 1,4 , A. Vamvakeros 1,4 1 University College London, UK, 2 Rutherford Appleton Laboratory, UK 3 National Physical Laboratory, UK, 4 Finden Limited, UK We present the first regression deep convolutional neural network, termed PQ-Net, 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 X-ray diffraction computed tomography dataset of a multi-phase Ni-Pd/CeO2-ZrO2/Al2O3 catalytic material system consisting of ca. 50,000 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 using full profile analysis using the Rietveld method. The main advantage of PQ-Netis its ability to yield these results orders of magnitude faster, showing its potential as a new tool for real-time diffraction data analysis during in situ/operando experiments. References 1. Dong, H., Butler, K.T., Matras, D.et al.A deep convolutional neural network for real-time full profile analysis of big powder diffraction data.npj Comput Mater7,74 (2021). https://doi.org/10.1038/s41524-021-00542-4
P09
© The Author(s), 2022
Made with FlippingBook Learn more on our blog