Deep-learning-based quality filtering of mechanically exfoliated 2D crystals

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 counterparts. 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 24 images successfully distinguish monolayer and bilayer MoS2 with a success rate of 70%, which is a practical value in the first screening process for choosing monolayer and bilayer flakes of MoS2 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.

Facile and Reliable Thickness Identification of Atomically Thin Dichalcogenide Semiconductors Using Hyperspectral Microscopy

“…Besides, it is also difficult for human eyes to determine the flake thickness precisely. To this end, using machine learning or AI-based techniques to automatically search for flakes with the desired size and thickness has been developed recently, as described in the introduction [25][26][27]. However, the AI-based techniques are not smart enough to explore thousands of possible novel layered 2D materials for future electronics.…”

“…On the contrary, OM is an easy-accessible, efficient, and nondestructive technique that enables rapid characterization of layered 2D materials. Nowadays, due to the layer-dependent optical contrast between the atomically thin 2D crystals and the substrate, optical microscopic techniques have been widely employed to identify the flakes and to determine their thickness [24,25].…”

“…Besides, it is also difficult for human eyes to determine the flake thickness precisely. To this end, using machine learning to automatically search for flakes with the desired size and thickness has been demonstrated recently [25][26][27]. The artificial-intelligence (AI)-based techniques have enabled high-throughput autonomous assembly of van der Waals superlattices and heterostructures of 2D atomic crystals [28,29].…”

Deep learning enabled inorganic material generator

“…[12][13][14][15][16][17][18] In the past decade, machine learning techniques have been widely used by researchers to address many challenges in the field of material science and engineering. 2,[19][20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35] Ward et al. 27 proposed a set of 145 hand engineered features based on stoichiometric attributes, elemental property statistics, electronic structure attributes, ionic compound attributes which can be used for a broad range of datasets.…”

Fully Automated Identification of Two-Dimensional Material Samples

“…However, the observed difference in contrast and color of a flake with respect to the background not only depends on its thickness and material but also on the substrate that is used and on the settings of the microscope. This large parameter space makes the identification of usable flakes tedious and, while there exist proposed algorithmic solutions [14][15][16][17][18][19][20][21], a sufficiently general and fast algorithm is difficult to formulate. So far, many existing algorithmic approaches have concentrated on rule-based image processing [16] and a combination of the latter with machine learning [15,17].…”