Atomic-Scale Insights into Matter via Multi-Modal STEM: From Structure to Function

The development of next-generation technologies, from energy conversion to quantum information, is increasingly limited by our ability to study and control matter at the atomic scale. Bridging the gap between synthesis and application requires a detailed understanding of how local structural motifs and electronic states govern macroscopic functionality. Scanning transmission electron microscopy (STEM) has emerged as an indispensable approach for this purpose. By leveraging a rich diversity of simultaneous signals, it provides a unique multi-modal platform to correlate structural, chemical, and physical properties. While scanning probe techniques offer comparable resolution for surface analysis, STEM remains essentially unmatched for accessing information from within the bulk. In this talk, I will present research highlights from our group demonstrating the versatility of modern STEM across a wide range of material systems and research questions. These include the investigation of single dopant atoms in complex materials [1] and the exploration of the interplay between structural and electronic properties in oxide eterostructures. Utilizing atomic-resolution electron energy-loss spectroscopy (EELS) and integrated Differential Phase Contrast (iDPC) imaging, we study two-dimensional electron gas (2DEG) formation, correlating oxygen defect distribution with metallic states at the interface [2]. Furthermore, we demonstrate the quantitative mapping of lithium distribution and structural evolution in battery materials by harnessing low-dose imaging and diffraction techniques [3]. Our multi-modal approach extends to beam-sensitive systems, where Cryo-STEM is utilised to reveal elusive structure-property relationships, such as the dielectric response in organic photovoltaics (OPV) and the structure of cellulose nanocrystals [4]. Possibilities enabled by novel direct electron detectors for in situ EELS elemental mapping will also be discussed, illustrated by phase evolution studies in aluminium alloys [5]. Finally, I will provide an outlook on how simulation-informed ptychography and machinelearning-assisted analysis may circumvent.