Crystallographic prediction from diffraction and chemistry data for higher throughput classification using machine learning

Aguiar, Jeffery A.; Gong, Matthew L.; Taşdizen, Tolga

doi:10.1016/j.commatsci.2019.109409

Cited by 36 publications

(29 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This idea to combine multiple ML models is called stacking 46 . Aguiar et al 32 , 33 reports that stacking improves the prediction accuracy for space group determination by electron diffraction. However, for our case, stacking did not improve the performance, i.e., the prediction performance was deteriorated.…”

Section: Resultsmentioning

confidence: 99%

“…Application of ML and related techniques for diffraction data analysis is a hot research topic in recent times 13 , 14 . Among various subtopics such as pattern decomposition and phase identification 15 – 18 cluster analysis and phase mapping 19 – 23 , similarity metrics for comparison of diffraction data 24 – 26 classification of a crystal symmetry 27 – 33 , a paper by Park et al 34 is relevant to this work. Park et al classified crystal systems and space groups by applying a convolutional neural network (CNN) to simulated powder XRD patterns.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Symmetry prediction and knowledge discovery from X-ray diffraction patterns using an interpretable machine learning approach

Suzuki

Hino

Hawai

et al. 2020

Sci Rep

View full text Add to dashboard Cite

Determination of crystal system and space group in the initial stages of crystal structure analysis forms a bottleneck in material science workflow that often requires manual tuning. Herein we propose a machine-learning (ML)-based approach for crystal system and space group classification based on powder X-ray diffraction (XRD) patterns as a proof of concept using simulated patterns. Our tree-ensemble-based ML model works with nearly or over 90% accuracy for crystal system classification, except for triclinic cases, and with 88% accuracy for space group classification with five candidates. We also succeeded in quantifying empirical knowledge vaguely shared among experts, showing the possibility for data-driven discovery of unrecognised characteristics embedded in experimental data by using an interpretable ML approach.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Symmetry prediction and knowledge discovery from X-ray diffraction patterns using an interpretable machine learning approach

Suzuki

Hino

Hawai

et al. 2020

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Because the data were largely inconclusive when compared to known phases, a peak-to-peak correspondence with all known materials was computed using the latest trained neural network model for crystallography, reported by Aguiar et al in Figure 34. 4,5 Based on the results of combining multiple peaks and combinations across all known crystallographic chemistries, the cubic structure is indexed at the highest significance amongst the crystal families. Within cubic, there is a high probability reported in Figure 34c to classify within cubic to space group Fm3m, #225.…”

Section: Crystal Chemistrymentioning

confidence: 99%

Aluminum Spent Fuel Performance in Dry Storage Task 4 Aluminum Oxide Sampling of ATR Dry Stored Fuel

Winston

Middlemas

Winston

et al. 2020

View full text Add to dashboard Cite

and SEM Analysis Jatuporn Burns, TEM preparation Kevin Tolman, X-ray Diffraction AnalysisXiang Liu, TEM Data Collection v non-native oxide is somewhat porous, meaning that if the fuel storage environment changes, these locations may be points where corrosion would be initiated.Video images from the sampling operation indicate that the cut ends and locations where the elements were handled using a "friction tool" have retained the large white oxide blooms. Otherwise, the fuel elements have a neutral medium gray external appearance with little evidence of corrosive attack or surface alteration. The individual fuel plates are clearly visible through the side plate openings and show no sign of oxide alteration or damage. vi vii ACKNOWLEDGEMENTAssistance from Andy Gray, Alan Olaveson, Craig Carlson, Caleb Maynard, and Gary Landon for help in validation of the sampling tools and acquiring the samples from the ATR fuel elements, Claude Kimball for contract logistics and video review support, Marcie Anderson and Shawn Tigue for coordinating shipment, Al Carlson for retrieving video footage, Mark Taylor for Electron Microscopy Laboratory receipt scheduling, and Joanne Grimmett for receiving and completing the inter-building laboratory transfers is gratefully acknowledged.

show abstract

“…Liu et al 21 refined atomic pair distribution functions in a convolutional neural network (CNN) to classify SGs. For similar purposes, Park et al 22 , Vecsei et al 23 , Wang et al 24 , Oviedo et al 25 , and Aguiar et al 26 used powder X-ray diffraction (XRD) 1D curves, for which information such as peak positions, intensities, and fullwidths at half-maximum are mainly treated as the key input descriptors. In addition, Ziletti et al 27 (in a parent work of this study), Aguiar et al 28 , Kaufmann et al 29 , and Ziatdinov et al 30 developed DL models by extracting features from electron-beam based 2D DPs.…”

Section: Introductionmentioning

confidence: 99%

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

et al. 2020

View full text Add to dashboard Cite

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.

show abstract

Crystallographic prediction from diffraction and chemistry data for higher throughput classification using machine learning

Cited by 36 publications

References 42 publications

Symmetry prediction and knowledge discovery from X-ray diffraction patterns using an interpretable machine learning approach

Symmetry prediction and knowledge discovery from X-ray diffraction patterns using an interpretable machine learning approach

Aluminum Spent Fuel Performance in Dry Storage Task 4 Aluminum Oxide Sampling of ATR Dry Stored Fuel

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

Contact Info

Product

Resources

About