Data cluster analysis and machine learning for classification of twisted bilayer graphene

Vincent, Tom; Kawahara, Kenji; Antonov, Vladimir; Ago, Hiroki; Kazakova, Olga

doi:10.1016/j.carbon.2022.09.021

Cited by 7 publications

(5 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, another possibility to enhance the precision can be the use of whole (or parts) of the Raman spectra. This has already been proved with spectra simulated by first principles, and more recently with trainsets limited to a few selected angles of artificially stacked tBLG . However, the method employed here using some selected features of the spectra results in much faster and simpler calculations.…”

Section: Resultsmentioning

confidence: 84%

“…Once the spectra are classified, a few measurable features were extracted from each spectrum (Figure ). , The selected features include the shift (ω) and full width at half-maximum (FWHM) of each of the main graphene Raman bands (G and 2D) and the relative intensities of the 2D, D, R, and R′ bands with respect to the G band ( I 2D/G , I D/G , I R/G , and I R′/G ). The relative band intensities were used instead of the actual intensities, as they are less prone to variations from different measurements.…”

Section: Resultsmentioning

confidence: 99%

“…Using clustering analysis, we first tried to look in our training set for hidden information that can arise from different twist angles within each of the classes. Clustering comprises a range of unsupervised learning techniques that allow finding groups (clusters) of instances in the dataset without prior knowledge of the classes. ,, Instances pertaining to the same cluster are expected to be more similar to each other than to those in other clusters. We chose to use spectral clustering, an algorithm that allows us to decide the number of clusters, hence allowing us to start from the same number of classes that we have determined.…”

Section: Resultsmentioning

confidence: 99%

“…The structural relaxation at low twist angles to minimize the AA-stacking area leads to phononic reconstructions, expected to produce evident changes in the Raman G band that include splitting and shift. , However, real Raman spectra can differ significantly from the simulated ones, with the positions, widths, and relative intensities of the most characteristic Raman peaks being heavily affected by things like strain and doping . ML has also been used to identify selected twist angles of BLG made by artificial stacks of single-layer graphene (SLG) . Clustering, an unsupervised ML paradigm, has also been used to assemble uncategorized Raman data into groups roughly corresponding to different kinds of BLG, but without sensitivity toward the twist angle …”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Machine Learning Determination of the Twist Angle of Bilayer Graphene by Raman Spectroscopy: Implications for van der Waals Heterostructures

Solís‐Fernández

Ago

2022

ACS Appl. Nano Mater.

Self Cite

View full text Add to dashboard Cite

With the increasing interest in twisted bilayer graphene (tBLG) of the past years, fast, reliable, and nondestructive methods to precisely determine the twist angle are required. Raman spectroscopy potentially provides such a method, given the large amount of information about the state of the graphene that is encoded in its Raman spectrum. However, changes in the Raman spectra induced by the stacking order can be very subtle, thus making the angle identification tedious. In this work, we propose the use of machine learning (ML) analysis techniques for the automated classification of the Raman spectrum of tBLG into a selected range of twist angles. The ML classification proposed here is low computationally demanding, providing fast and accurate results with a ∼99% agreement with the manual labeling of the spectra. The flexibility and noninvasive nature of the Raman measurements, paired with the predictive accuracy of the ML, is expected to facilitate the exploration of the emerging research field of twisted van der Waals heterostructures. Moreover, the present work showcases how the currently available open-source tools facilitate the study and integration of ML-based techniques.

show abstract

Section: Resultsmentioning

confidence: 84%

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Machine Learning Determination of the Twist Angle of Bilayer Graphene by Raman Spectroscopy: Implications for van der Waals Heterostructures

Solís‐Fernández

Ago

2022

ACS Appl. Nano Mater.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Similarly, based on the Gaussian mixture model data clustering algorithm and PCA, it is also proved experimentally that the algorithm can help distinguish the number of layers and torsion angle. [ 287 ]…”

Section: Rising Machine Learning Algorithms Enabled Thickness Determi...mentioning

confidence: 99%

Thickness Determination of Ultrathin 2D Materials Empowered by Machine Learning Algorithms

Mao

Wang

Chen

et al. 2023

Laser & Photonics Reviews

View full text Add to dashboard Cite

The number of layers of 2D materials is of great significance for regulating the performance of nanoelectronic devices and optoelectronic devices, where the optimal thickness of the target sample should be determined before further physical research or device manufacturing steps. At present, a variety of different optical technologies have been proposed to determine the thickness of samples by using the relationship between the number of layers and optical properties, including optical contrast, optical imaging, Raman spectra, photoluminescence spectra, nonlinear spectra, near‐field optical imaging, ellipsometry spectra, and hyperspectral imaging. In the past decade, the rapidly growing number of 2D materials and their heterostructures has exceeded the capacity of traditional experimental and computational methods. In recent years, machine learning (ML) is emerging as a powerful tool to support such traditional methods, which brings new opportunities to tap the potential of optical technology in a more perspicacious way. The application of optical technology has greatly facilitated the acquisition of optical information of materials, while ML algorithms provide a fast, high‐throughput, and intelligent way to complete back‐end data processing and inference. A profound integration of conventional optical technologies and ML algorithms significantly helps 2D materials progress in basic studies and practical applications and further promotes industrial manufacturing.

show abstract

MEBA: AI-powered precise building monthly energy benchmarking approach

Li,

Bie,

et al. 2024

Applied Energy

View full text Add to dashboard Cite

Data cluster analysis and machine learning for classification of twisted bilayer graphene

Cited by 7 publications

References 31 publications

Machine Learning Determination of the Twist Angle of Bilayer Graphene by Raman Spectroscopy: Implications for van der Waals Heterostructures

Machine Learning Determination of the Twist Angle of Bilayer Graphene by Raman Spectroscopy: Implications for van der Waals Heterostructures

Thickness Determination of Ultrathin 2D Materials Empowered by Machine Learning Algorithms

MEBA: AI-powered precise building monthly energy benchmarking approach

Contact Info

Product

Resources

About