Dynamic ferroelectric domain wall topologies overturn the classical idea that our nanoelectronics need to consist of fixed components of hardware. To harness the true potential of domain wall-based electronics, we must take a step back from the device design level, and instead re-look at the subatomic internal properties. With recent advances in experimental characterisation and theoretical calculation approaches, in the last 5 years reports of non-classic internal structures and functionalities within domain walls have become a common occurrence. As the region of interest is at the nanoscale and dynamic, it is essential for the physical characterisation to be at this scale spatially and time resolved.

This presentation focuses on using the applied electric field of aberration corrected scanning transmission electron microscopy (STEM) probes to move domain walls, and thus investigate their dynamics while imaging from the sub-atomic to µm scale. As the STEM probe can be controlled in terms of dose, probe size, direction and speed, a diverse set of experiments is possible without complicated sample preparation. Using STEM, electron diffraction technique known as 4D-STEM and electron energy loss spectroscopy (EELS) we can measure the changing polarisation, strain, band-gap and valence state can be investigated with controlled variants in applied field conditions. By controlling the incoming STEM probe direction, parallel domain walls could be moved around to form stable vertex junctions, thus switching from a neutral to charged state. By combining the local atomic resolution structure, strain, charge density and band structure measurements we can resolve all the measurable parameters of interest within domain walls and thus start unravelling the fundamental physics governing their formation, dynamics and resulting functionality. Finally using a liquid nitrogen cryogenic cooling and biasing STEM holder we probe the lower temperature multiferroic phases and topology dynamics of these room temperature ferroelectric materials.

Over the last 70 years, the observation of ferroelectric domains has been done by different characterization tools such as polarized optical microscopy, piezoresponse force microscopy (PFM), scanning electron microscopy, photoemission electron microscopy, low energy electron microscopy and transmission electron microscopy. In this presentation, we will demonstrate that helium ion microscopy (HIM) can also be added to this ensemble of methods for ferroelectric domain imaging. We will discuss the observation of polar and nonpolar domains in LiNbO₃ and ErMnO₃ single crystals. By comparing with PFM images and conductive atomic force microscopy images, we prove that HIM can reproduce the maps of the domains and of the domain walls with a high spatial resolution in a wide range of acceleration voltages and beam currents of He+ ions. The polar domain (out-of-plane polarization) contrast in the HIM domain images can be explained by the difference of the surface potentials or the work functions of the oppositely poled domains. Surprisingly, HIM also allows to observe nonpolar domains (in-plane polarization) despite their uniform surface potential. The mechanism at play may involve the bulk photovoltaic effect or the asymmetric flux of charge carriers in the ferroelectric material. While the first HIM scans on the nonpolar surface of an ErMnO₃ single crystal revealed only the domain walls, then the consecutive scans make the domain contrast grow along its polar axis. It is believed that the asymmetric flux of charge carriers remains inside the crystal but induces the gradual surface charging and alters the surface potential selectively in different domains.

Domain walls, which separate regions of differing polarisation in ferroic materials, are not only mobile but often exhibit properties different to that of the domains themselves. This includes increased conductivity at the domain walls or displaying polar domain walls in otherwise non-polar materials, which may find applications in nanoelectronics and memristive technologies. Despite being energetically unfavourable in proper ferroelectrics such as BaTiO₃, charged domain walls are abundant in hybrid improper ferroelectric (Ca, Sr)₃Ti₂O₇, as shown by TEM and IP-PFM, which is believed to result in a polarisation switching barrier that is significantly lower than theoretically predicted. Here we utilised scanning 3-dimensional X-ray diffraction (3DXRD) to visualise the domain structure within a single crystal of Ca₂.₁₅Sr₀.₈₅Ti₂O₇. 3DXRD is a versatile, non-destructive tomographic technique that can image through the bulk of a sample, with the image contrast given by diffraction intensities. In these materials, as polarisation arises as a secondary effect driven by the primary order parameters associated with rotations and tilts of the TiO₆ octahedra, which produce superstructure diffraction peaks, 3DXRD provides a particularly powerful tool for probing the domain boundary structures. From our measurements, we can extract the strains and the atomic displacements that underpin the domains to further our understanding of the ferroelastic and ferroelectric domain structure. Understanding the domain structure is fundamental to revealing the polarisation switching mechanism which will, in turn, aid in the development of devices that utilise domain walls as a key component in order to meet the demand for fast high-density data storage.