4D STEM

The development of aberration corrected transmission electron microscopes and fast pixelated electron detectors has led to the emergence of four-dimensional STEM (4D STEM) imaging, where a diffraction pattern is captured for every scanning position, as a prominent new method for studying materials down to atomic resolution. Diffraction patterns carry a great wealth of information about the electron beam’s interaction with the sample, much of which is discarded by conventional STEM detectors when the signal is integrated to generate only a single pixel value per position of the probe. However, by capturing the full diffraction pattern, 4D STEM allows us to capture as much information as possible. As high-speed pixelated electron detectors and cameras have proliferated in the electron microscopy community, 4D STEM has garnered significant interest due to its versatility. With a 4D diffraction dataset, any type of STEM image (bright field, annular bright field, or annular dark field) can be generated. Virtual apertures with arbitrary shape applied in post processing can also be used to generate images that would be impossible with a physical aperture plate. But arbitrary image reconstruction is only the beginning of what is possible.

            Our research at UCI has focused on using 4D STEM to image the electronic properties of ferroelectrics, particularly the local electric field and charge distribution. A probe placed near to a nucleus or in a strong electric field will experience a shift in its momentum. In thin samples, the change in momentum of the probe (also called the momentum transfer) is negatively proportional to the average electric field in the interaction volume. The local charge density can then be calculated from the electric field using Gauss’ law from basic electrodynamics. Various other techniques using X-ray diffraction and TEM have been developed to study real-space charge density in the past, but they required significant simulation and analysis of diffraction data from the material under study. 4D STEM allows us to overcome this barrier because each diffraction pattern in the 4D dataset is acquired from an area defined by the size of the electron probe and the highly localized charge distribution can be calculated without the need for simulation. To image the charge density surrounding a defect or interface, we need only to position our scanning probe over the desired region; this opens the door to imaging the charge density in all kinds of nanostructured materials.

Related Publications

• Gao, W., Addiego, C., Wang, H., Yan, X., Hou, Y., Ji, D., Heikes, C., Zhang, Y., Li, L., Huyan, H., Blum, T., Aoki, T., Nie, Y., Schlom, D.G., Wu, R., Pan, X.Q., Real-Space Charge-Density Imaging with Sub-Angstrom Resolution by Four-Dimensional Electron Microscopy. Nature 2019 575, 490-484. https://doi.org/10.1038/s41586-019-1649-6
• Addiego, C., Gao, W., Pan, X.Q., Thickness and Defocus Dependence of Inter-atomic Electric Fields Measured by Scanning Diffraction. Ultramicroscopy 2020 208, 112859. https://doi.org/10.1016/j.ultramic.2019.112850