edfas.org ELECTRONIC DEVICE FAILURE ANALYSIS | VOLUME 26 NO. 1 4 EDFAAO (2024) 1:4-13 1537-0755/$19.00 ©ASM International® FOUR-DIMENSIONAL SCANNING TRANSMISSION ELECTRON MICROSCOPY: PART II, CRYSTAL ORIENTATION AND PHASE, SHORT AND MEDIUM RANGE ORDER, AND ELECTROMAGNETIC FIELDS Aaron C. Johnston-Peck and Andrew A. Herzing National Institute of Standards and Technology, Gaithersburg, Maryland aaron.johnston-peck@nist.gov INTRODUCTION The second part of this series presents applications of four-dimensional scanning transmission electron microscopy (4D-STEM). The first article of this series, published in the August 2023 issue of EDFA, described a multitude of ways to characterize a sample’s structure. As a refresher, 4D-STEM is a spatially resolved electron diffraction technique that records the electron scattering distribution (kx, ky) at each point of the electron beam raster (rx, ry), thereby producing a four-dimensional dataset (rx, ry, kx, ky). This article covers measurements of crystal orientation and phase, short- and medium-range order (SRO/MRO), and internal electromagnetic fields. CRYSTAL ORIENTATION AND PHASE MAPPING Orientation and phase mapping of materials is commonly carried out using electron backscatter diffraction (EBSD) in the scanning electron microscope (SEM). This involves measurement of the position and angles between lines in the Kikuchi patterns formed at each position in an image scan. More recently, the spatial resolution of this method has been improved by collecting the forward- scattered electrons in a transmission Kikuchi diffraction (TKD) experiment.[1] For example, TKD was recently used to map the phases present in Zr-doped HfO2 films used in nonvolatile memory applications.[2] The development of high-speed direct electron detection cameras has enabled the acquisition of such data over very large fields of view, which would have been prohibitively slow using CCDbased cameras. The utility of this increased analytical area was recently demonstrated by using TKD to characterize a variety of 2D materials such as graphene and MoS2. [3] The resulting maps showed the orientation and phase over several millimeters, in some cases, while retaining spatial resolution below 10 nanometers. The spatial resolution of this type of measurement can be improved still further using a 4D-STEM approach. A reduction in interaction volume, due to the higher beam energies available in STEM combined with the thin specimens employed, further improves the spatial resolution and signal above background for the diffraction peaks. As in EBSD/TKD, STEM-based orientation mapping can be performed by analyzing the Kikuchi lines present in the patterns. This approach is typically only an option for thicker specimens since the Kikuchi lines are formed by diffracted electrons that had previously undergone diffuse scattering. In thinner samples, Kikuchi lines are much weaker or often invisible and therefore cannot be used for phase and orientation mapping by 4D-STEM. In this case, the measurement is based on indexing individual diffraction peaks.[4] Both commercial[5] and open-source software packages (references 6 and 7, for example,) are available for this task. Several approaches have been demonstrated to improve the fidelity of analyzing the Bragg reflections to determine sample orientation. As discussed in Part I, the ability to precess the incident beam about the optic axis during 4D-STEM data acquisition can be invaluable. In the case of orientation and phase mapping, the primary benefit of precession is that the resulting patterns can be more readily fit to models due to the reduction in dynamic scattering effects.[5] However, precession requires additional specialized hardware and poses some experimental constraints. A recently proposed alternative is multibeam diffraction. This technique utilizes a diaphragm with multiple beam forming apertures, thereby generating multiple electron probes on the sample and hence multiple diffraction patterns are simultaneously
RkJQdWJsaXNoZXIy MTYyMzk3NQ==