skip to main content

Title: Physically informed machine-learning algorithms for the identification of two-dimensional atomic crystals

After graphene was first exfoliated in 2004, research worldwide has focused on discovering and exploiting its distinctive electronic, mechanical, and structural properties. Application of the efficacious methodology used to fabricate graphene, mechanical exfoliation followed by optical microscopy inspection, to other analogous bulk materials has resulted in many more two-dimensional (2D) atomic crystals. Despite their fascinating physical properties, manual identification of 2D atomic crystals has the clear drawback of low-throughput and hence is impractical for any scale-up applications of 2D samples. To combat this, recent integration of high-performance machine-learning techniques, usually deep learning algorithms because of their impressive object recognition abilities, with optical microscopy have been used to accelerate and automate this traditional flake identification process. However, deep learning methods require immense datasets and rely on uninterpretable and complicated algorithms for predictions. Conversely, tree-based machine-learning algorithms represent highly transparent and accessible models. We investigate these tree-based algorithms, with features that mimic color contrast, for automating the manual inspection process of exfoliated 2D materials (e.g., MoSe2). We examine their performance in comparison to ResNet, a famous Convolutional Neural Network (CNN), in terms of accuracy and the physical nature of their decision-making process. We find that the decision trees, gradient boosted decision trees, and random forests utilize physical aspects of the images to successfully identify 2D atomic crystals without suffering from extreme overfitting and high training dataset demands. We also employ a post-hoc study that identifies the sub-regions CNNs rely on for classification and find that they regularly utilize physically insignificant image attributes when correctly identifying thin materials.

more » « less
Award ID(s):
1846747 1749774
Author(s) / Creator(s):
; ; ; ;
Publisher / Repository:
Nature Publishing Group
Date Published:
Journal Name:
Scientific Reports
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Two-dimensional (2D) materials assembled into van der Waals (vdW) heterostructures contain unlimited combinations of mechanical, optical, and electrical properties that can be harnessed for potential device applications. Critically, these structures require control over interfacial adhesion for enabling their construction and have enough integrity to survive industrial fabrication processes upon their integration. Here, we promptly determine the adhesion quality of various exfoliated 2D materials on conventional SiO 2 /Si substrates using ultrasonic delamination threshold testing. This test allows us to quickly infer relative substrate adhesion based on the percent area of 2D flakes that survive a fixed time in an ultrasonic bath, allowing for control over process parameters that yield high or poor adhesion. We leverage this control of adhesion to optimize the vdW heterostructure assembly process, where we show that samples with high or low substrate adhesion relative to each other can be used selectively to construct high-throughput vdW stacks. Instead of tuning the adhesion of polymer stamps to 2D materials with constant 2D-substrate adhesion, we tune the 2D-substrate adhesion with constant stamp adhesion to 2D materials. The polymer stamps may be reused without any polymer melting steps, thus avoiding high temperatures (<120 °C) and allowing for high-throughput production. We show that this procedure can be used to create high-quality 2D twisted bilayer graphene on SiO 2 /Si, characterized with atomic force microscopy and Raman spectroscopic mapping, as well as low-angle twisted bilayer WSe 2 on h-BN/SiO 2 /Si, where we show direct real-space visualization of moiré reconstruction with tilt-angle dependent scanning electron microscopy. 
    more » « less
  2. Abstract

    Advancements in low‐dimensional functional device technology heavily rely on the discovery of suitable materials which have interesting physical properties as well as can be exfoliated down to the 2D limit. Exfoliable high‐mobility magnets are one such class of materials that, not due to lack of effort, has been limited to only a handful of options. So far, most of the attention has been focused on the van der Waals (vdW) systems. However, even within the non‐vdW, layered materials, it is possible to find all these desirable features. Using chemical reasoning, it is found that NdSb2is an ideal example. Even with a relatively small interlayer distance, this material can be exfoliated down to few layers. NdSb2has an antiferromagnetic ground state with a quasi 2D spin arrangement. The bulk crystals show a very large, non‐saturating magnetoresistance along with highly anisotropic electronic transport properties. It is confirmed that this anisotropy originates from the 2D Fermi pockets which also imply a rather quasi 2D confinement of the charge carrier density. Both electron and hole‐type carriers show very high mobilities. The possible non‐collinear spin arrangement also results in an anomalous Hall effect.

    more » « less
  3. Abstract

    Mechanical exfoliation yields high‐quality 2D materials but is challenging to scale up due to the small lateral size and low yield of the exfoliated crystals. Gold‐mediated exfoliation of macroscale monolayer MoS2and related crystals addresses this problem. However, it remains unclear whether this method can be extended to other metals. Herein, mechanical exfoliation of MoS2on a range of metallic substrates is studied. It is found that Au outperforms all the other metals in their ability to exfoliate macroscale monolayer MoS2. This is rationalized by gold's ability to resist oxidation, which is compromised on other metals and leads to a weakened binding with MoS2. An anomalously high monolayer yield found for Ag suggests that the large interfacial strain in the metal–MoS2heterostructures measured by Raman spectroscopy also is a critical factor facilitating the exfoliation, while the relative differences in the metal–MoS2binding play only a minor role. These results provide a new incentive for investigations of 2D material‐substrate combinations applicable where high‐quality 2D crystals of macroscopic dimensions are of importance.

    more » « less
  4. Abstract

    Advances in nanoscience have enabled the synthesis of nanomaterials, such as graphene, from low‐value or waste materials through flash Joule heating. Though this capability is promising, the complex and entangled variables that govern nanocrystal formation in the Joule heating process remain poorly understood. In this work, machine learning (ML) models are constructed to explore the factors that drive the transformation of amorphous carbon into graphene nanocrystals during flash Joule heating. An XGBoost regression model of crystallinity achieves anr2score of 0.8051 ± 0.054. Feature importance assays and decision trees extracted from these models reveal key considerations in the selection of starting materials and the role of stochastic current fluctuations in flash Joule heating synthesis. Furthermore, partial dependence analyses demonstrate the importance of charge and current density as predictors of crystallinity, implying a progression from reaction‐limited to diffusion‐limited kinetics as flash Joule heating parameters change. Finally, a practical application of the ML models is shown by using Bayesian meta‐learning algorithms to automatically improve bulk crystallinity over many Joule heating reactions. These results illustrate the power of ML as a tool to analyze complex nanomanufacturing processes and enable the synthesis of 2D crystals with desirable properties by flash Joule heating.

    more » « less
  5. As they hold extraordinary mechanical and physical properties, two-dimensional (2D) atomic layer materials, including graphene, transition metal dichalcogenides, and MXenes, have attracted a great deal of attention. The characterization of energy and charge transport in these materials is particularly crucial for their applications. As noncontact methods, Raman-based techniques are widely used in exploring the energy and charge transport in 2D materials. In this review, we explain the principle of Raman-based thermometry in detail. We critically review different Raman-based techniques, which include steady state Raman, time-domain differential Raman, frequency-resolved Raman, and energy transport state-resolved Raman techniques constructed in the frequency domain, space domain, and time domain. Detailed outlooks are provided about Raman-based energy and charge transport in 2D materials and issues that need special attention. 
    more » « less