Skip to content Skip to main navigation Report an accessibility issue

Dr. Kevin Roccapriore, November 18th

Discovery and control of emergent quantum phenomena in the STEM


Understanding and controlling phenomena on the nanometer and atomic scales is both a long-term dream of mankind, and is the key to quantum computing, beyond-Moore nanoelectronics, plasmonic and magneto-plasmonic devices, and other serendipitous areas. Over the last decade, Scanning Transmission Electron Microscopy (STEM) has become the go-to tool for exploring the physics of materials and devices on the atomic level. A number of cutting-edge machines now exist worldwide, and new technique advances are reported monthly. In this presentation, I will discuss a variety of phenomena observed through STEM and EELS ranging from nanoparticle plasmonics and new effects in less-studied 2D material systems, to beam-induced transformations in these. Many of the electronic effects captured in (3D) EELS hyperspectral data can be difficult to deconvolve or extract physically meaningful interpretations. To this end, machine learning (ML) methods are discussed that aim to help understanding the observed effects as well as pinpointing the relationship between the electronic response and structural image or geometry. I further aim to extend STEM from an imaging and spectroscopy tool towards an atomic-scale fabrication tool. In combining knowledge of the learned relationships from ML with the ability to sculpt materials using the electron beam, systems can be designed to exhibit, for example, specific plasmon resonances and other desired functionality. Finally, I demonstrate that automated experiments enabled by deep learning can be used as a tool for discovery of physics by augmenting the microscope’s ability in its decision-making criteria for automatic probing.


Kevin is a Research Associate in the Center for Nanophase Materials at Oak Ridge National Laboratory (ORNL). He received his B.S. from the University of Florida in 2011 and Ph.D. in Physics from the University of North Texas in 2018, and completed his post-doc with ORNL in 2021 before joining as staff. His research interests include both the understanding of fundamental physics in areas such as plasmonics, nanophotonics, metamaterials, and two-dimensional materials, as well as device physics using these unique materials and properties. To better understand the rich physics, Kevin uses advanced electron microscopy techniques (STEM), including both optical and vibrational spectroscopy on the nanometer scale. He uses machine learning techniques and artificial neural networks to realize the physical behavior in a variety of material systems. Finally, Kevin has enabled automated experiments in the STEM where active learning of structure property relationships is now possible, allowing autonomous exploration and discovery of physical behavior.