• April 21, 2026

  • ‍ ‍4:00 PM (CST): Progress Toward Automated Van der Waals Heterostructure Stacking

    • Speaker: Timmy Rosno (Department of Physics, University of Arkansas)

    • Abstract

      • The manual production of heterostructure devices from two-dimensional van der Waals materials such as graphene and hexagonal boron nitride is time-consuming, inconsistent, and inconvenient to perform in inert environments that limit oxidation and hydrocarbon residue. The 2D Quantum Materials Automated Pipelines (2D-QMaPs) address these challenges through automated exfoliation, cataloging, and stacking of bulk-derived materials inside a glovebox. Metrology is performed by loading a tray of chips with exfoliated flakes onto a motorized stage with a camera and autofocus system to rapidly raster scan and process images from each location via machine learning for flake detection and classification, reducing human error and bias. This approach accelerates flake identification by up to 20x, improves consistency, and removes most false positives. Detected flakes are used to design heterostructure stacks and can be assembled by an automated stacker using thermal manipulation with a hemispherical polymethyl disulfide (PDMS) stamp and a sacrificial polycarbonate (PC) layer. Successful stacking requires near-perfect translational and rotational alignment, challenged by chip rotation within trays, thermal expansion, and the limits of conventional image processing in dynamic optical conditions. These constraints motivate machine learning approaches for image matching and real-time segmentation to generalize automated stacking.

  • 4:30 PM (CST): Robotic exfoliation and heterostructure assembly using the 2D-Quantum Materials Pipelines

    • Speaker: Jane Peabody (Department of Physics and Astronomy, The University of Utah)