Abstract
Two-dimensional (2D) crystals are attracting growing interest in various research fields such as engineering, physics, chemistry, pharmacy, and biology owing to their low dimensionality and dramatic change of properties compared to the bulk counter parts. Among the various techniques used to manufacture 2D crystals, mechanical exfoliation has been essential to practical applications and fundamental research. However, mechanically exfoliated crystals on substrates contain relatively thick flakes that must be found and removed manually, limiting high-throughput manufacturing of atomic 2D crystals and van der Waals heterostructures. Here, we present a deep-learning-based method to segment and identify the thickness of atomic layer flakes from optical microscopy images. Through carefully designing a neural network based on U-Net, we found that our neural network based on U-net trained only with the data based on realistically small number of images successfully distinguish monolayer and bilayer MoS2 and graphene with a success rate of 70–80%, which is a practical value in the first screening process for choosing monolayer and bilayer flakes of all flakes on substrates without human eye. The remarkable results highlight the possibility that a large fraction of manual laboratory work can be replaced by AI-based systems, boosting productivity.
Highlights
Two-dimensional (2D) crystals,[1] such as graphene, transition metal dichalcogenides, and van der Waals heterostructures[2] have been intensively studied in a wide range of research fields since they show significant properties that has not been observed in their bulk counter parts
Using a deep-neural network architecture, we reproduced the images of 2D crystals from the augmented data based on 24 and 30 optical microscopy (OM) images of MoS2 and graphene, respectively, used as training data and found that both the cross-validation score and accuracy rate for the test images through deep-neural networks is surprisingly around 70–80%, which means that our model can distinguish monolayer and bilayer with a practical success rate for initial screening process
The present results suggest the deep-neural network can become another promising way to quickly identify the thickness of various 2D crystals in an autonomous way, suggesting a large fraction of manual laboratory work can be dramatic decreased by replacing AI-based systems
Summary
Two-dimensional (2D) crystals,[1] such as graphene, transition metal dichalcogenides, and van der Waals heterostructures[2] have been intensively studied in a wide range of research fields since they show significant properties that has not been observed in their bulk counter parts. High-throughput identification of various unexplored 2D materials via machine learning and the development of a machine for mechanically exfoliated 2D atomic crystals to autonomously build van der Waals superlattices[14] have been reported. These advancements are suggestive of a new research direction for 2D materials that is aimed to explore efficiently enormous materials’ properties in large scale using robotics and machine learning. In such a situation, a rapid and versatile method for layer number identification of mechanically exfoliated atomic-layer crystals on substrates is highly desirable in their fundamental research and practical applications
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
