New Methods for Imaging Polar Topological Textures in Oxide Heterostructures

Polar topological textures such as polar skyrmions and merons in ferroelectric heterostructures emerge resulting from the interplay of elastic, electrostatic and gradient energies. These dipolar textures not only have mathematical beauty but also exhibit exotic functionalities such as emergent chirality and local negative capacitance for potential applications in next generation nanodevices. To understand and explore the microscopic details of such topological textures requires a characterization method which can track their detailed 3D structures at the atomic scale.<br/><br/>A popular approach for mapping polar distortions in ferroelectrics has been to use high-angle annular dark field (HAADF) scanning transmission electron microscopy (STEM), which directly image the atomic displacements with a precision of a few picometers. Although conceptually straightforward, because of the small depth-of-field and strong dechanneling of the probe scattered by heavy cations, HAADF-STEM only samples the top surface of the sample, which can have a very different polar order from the middle of a 3D structure such as skyrmion, for example Néel (top) versus Bloch (middle) characters.<br/><br/>Here, by utilizing the full 4-dimensional phase space information collected by a new design of direct electron detector that can measure the momentum transfer distributions at every probe position (4D-STEM), we can solve the inverse multiple scattering problem and retrieve the 3-dimensional electrostatic potential of the sample, yielding a lateral resolution of &lt;20 pm. This allows us to determine both the cation displacements and octahedral tilts. To further improve the sensitivity to polar order, we combine scanning diffraction with dynamical diffraction simulations to robustly image the polarization and chirality in each polar texture. I will show how 4D-STEM-based approach allows us to image dipolar waves, polar skyrmions and merons in ferroelectric thin films that are unable to measure by conventional methods.<br/><br/>Research supported by AFOSR Hybrid Materials MURI (FA9550-18-1-0480) and ARO ETHOS MURI (W911NF-21-2-0162), facilities support from the NSF (DMR-1429155, DMR-1719875, DMR-2039380). Researchers at the University of Arkansas also thank the Vannevar Bush Faculty Fellowship Grant No. N00014-20-1-2834 from the Department of Defense.