Identification of crystal symmetry from noisy diffraction patterns by a shape analysis and deep learning

Leslie Ching Ow Tiong,Donghun Kim,Jeongrae Kim,Sang Soo Han

doi:10.1038/s41524-020-00466-5

Leslie Ching Ow Tiong, Donghun Kim + Show 2 more

Open Access

https://doi.org/10.1038/s41524-020-00466-5

Copy DOI

Abstract

The robust and automated determination of crystal symmetry is of utmost importance in material characterization and analysis. Recent studies have shown that deep learning (DL) methods can effectively reveal the correlations between X-ray or electron-beam diffraction patterns and crystal symmetry. Despite their promise, most of these studies have been limited to identifying relatively few classes into which a target material may be grouped. On the other hand, the DL-based identification of crystal symmetry suffers from a drastic drop in accuracy for problems involving classification into tens or hundreds of symmetry classes (e.g., up to 230 space groups), severely limiting its practical usage. Here, we demonstrate that a combined approach of shaping diffraction patterns and implementing them in a multistream DenseNet (MSDN) substantially improves the accuracy of classification. Even with an imbalanced dataset of 108,658 individual crystals sampled from 72 space groups, our model achieves 80.12 ± 0.09% space group classification accuracy, outperforming conventional benchmark models by 17–27 percentage points (%p). The enhancement can be largely attributed to the pattern shaping strategy, through which the subtle changes in patterns between symmetrically close crystal systems (e.g., monoclinic vs. orthorhombic or trigonal vs. hexagonal) are well differentiated. We additionally find that the MSDN architecture is advantageous for capturing patterns in a richer but less redundant manner relative to conventional convolutional neural networks. The proposed protocols in regard to both input descriptor processing and DL architecture enable accurate space group classification and thus improve the practical usage of the DL approach in crystal symmetry identification.

Highlights

High-throughput material synthesis and characterization have been popular topics of research during the past few decades and have accelerated the discovery of novel materials[1,2,3,4,5]
For the further processing of multiple inputs (DPs collected from the three-zone axes), we propose a multistream network, Raw diffraction patterns (DPs) are spotty and noisy and, difficult to learn from
To enhance the capabilities of deep learning (DL), we propose two ideas: one is to shape the DPs, and the other is to implement them in a multistream DL network (Fig. 1)

Summary

Introduction

High-throughput material synthesis and characterization have been popular topics of research during the past few decades and have accelerated the discovery of novel materials[1,2,3,4,5]. Various characterization methods exist, identifying the crystal symmetry, that is, the way the atoms are arranged in space, is inarguably the first and most important process in material characterization. This is because the crystallographic structure of a material plays an important role in determining the material properties (structure–property relationship)[6,7]. Consider the magnetism of iron: body centred cubic Fe is ferromagnetic, while face centred cubic Fe shows paramagnetic behaviors[8]. There are 230 distinct types of SGs when chiral copies are considered[9,10,11]; these SGs are formed from the combinations of the 32 point groups with the 14 Bravais lattices[12]

Methods

Results

Conclusion