Berlin 2024 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

HL: Fachverband Halbleiterphysik

HL 36: Poster III

HL 36.8: Poster

Wednesday, March 20, 2024, 18:00–20:30, Poster E

Machine Learning based Monolayer Classification using a Convolutional Neural Network and Characterization of Transition Metal Dichalcogenides Heterostructures — •Maximilian Nagel, Chirag Palekar, Bárbara Rosa, Ching-Wen Shih, and Stephan Reitzenstein — Institut für Festkörperphysik, Technische Universität Berlin, Hardenbergstraße 36, 10623 Berlin, Germany

This research presents an advancement in the mass production of transition metal dichalcogenides (TMDCs) with potential benefits for the semiconductor industry. We introduce an automated system using convolutional neural networks (CNNs) for identification, classification and streamlining the monolayer search. Additionally, our study utilizes photoluminescence and second harmonic generation measurements for precise twist angle determination of heterostructures which are prepared using dry transfer method. By integrating data acquisition, image analysis, and machine learning, this innovative approach aids the TMDC heterostructure fabrication and enhances our understanding of TMDC heterostructures' properties. Collectively, this work represents a significant step forward in achieving cost-effective and efficient thin film manufacturing, with substantial implications for the semiconductor industry's future competitiveness and innovation.

Keywords: Transition Metal Dichalcogenide (TMDC); Heterostructure; convolutional neural network (CNN); Automated System; Twist Angle