2021 Volume 62 Issue 9 Pages 1420-1423
“The 12th Japanese-Polish Joint Seminar on Micro and Nano Analysis” was held in Fukuoka, Japan from August 29 to September 1, 2018, and the proceedings were published in May, 2019, as a special issue of Materials Transactions (Vol. 60, No. 5). The main purpose of this seminar is to discuss the structural analysis of materials by electron microscopy techniques. Among the papers presented at the seminar, this article briefly reviews the following topics: observations of dislocations in a thick specimen by ultra-high voltage electron microscopy, suppression of geometric phase shift due to antiphase boundaries in dark-field electron holography, and structural characterization of amorphous materials by electron diffraction techniques.
Since the mechanical, physical, and chemical properties of materials strongly depend on the atomic arrangements and the slightly added functional elements, obtaining structural and chemical information is of technological importance for developing new structural and functional materials. In recent material developments, hybridization of various materials and introduction of nanostructures into conventional materials are being attempted in order to achieve high strength and high functionality. As a result, it is difficult to evaluate materials with nanoscale heterogeneous structures by using “average structure information” taken from a wide area. Electron microscopy can obtain “local structure information” by illuminating a nanometer-sized electron beam onto materials, and structural information, such as precipitates and defects, as well as chemical information, such as compositions and bonding states, is measurable from the same location of an object with the high accuracy at the atomic scale. Because of their excellent spatial resolution, transmission electron microscopy (TEM) and scanning transmission electron microscopy (STEM) are one of the important techniques indispensable for promoting materials science and engineering.
“The Japanese-Polish/Polish-Japanese Joint Seminar on Micro and Nano Analysis” was established for discussing progresses in structural and chemical analyses using microscopy and microanalysis techniques and their application to the field of materials science. This seminar is a biennial event and alternately organized in Japan and Poland since 1997, as summarized in Table 1. The twelfth joint seminar was held at Kyushu University Nishijin Plaza in Fukuoka, Japan from August 29 to September 1, 2018,1) and consisted of the following sessions: “Severe Deformation”, “Deformation, Stress, and Dislocation”, “Relationship between Structure and Functionality”, “Nanowire, Interface, and Low-dimensional Materials”, “Spectroscopy”, and “Processing”. Eight selected papers2–9) were published in a special issue of Materials Transactions (Vol. 60, No. 5) under the title of “New Trends for Structural and Chemical Analyses by Transmission Electron Microscopy”. Here, we review three topics presented in the seminar.
Ultra-high voltage electron microscopy (UHVEM) is one of the world’s leading technologies in Japan. Because of the high transmission power, it is possible to observe micrometer-thick specimens using UHVEM. However, the number of inelastically-scattered electrons increases with the specimen thickness along the electron beam direction, and the quality of images is degraded by chromatic aberration associated with the energy loss of the electrons. The UHVEM installed at the Ultramicroscopy Research Center, Kyushu University is equipped with an energy filter,10) which enables imaging using an electron beam with a specific energy. Sadamatsu et al.11) examined a single crystalline Si including artificially introduced high-density dislocations using energy-filtered UHVEM in combination with electron energy-loss spectroscopy, and demonstrated that the dislocations are clearly visible even in bulk specimens over 10 µm-thick. On the other hand, Sato et al.2,12) observed the dislocations in thick specimens by STEM using UHVEM installed at the Research Center for Ultra-High Voltage Electron Microscopy, Osaka University. Figure 1 shows a bright-field STEM image of a wedge-shaped Si specimen with a thickness from 1 µm (left-side) to 9 µm (right-side).2) The width of the dislocation lines is almost constant at 13–16 nm in the region with a thickness of 1 to 7.5 µm, suggesting that a wedge-shaped specimen with a thickness up to 7.5 µm can be observed in focus using STEM operated at an acceleration voltage of 1 MV. This is because STEM is less susceptible to the effect of resolution reduction due to chromatic aberration than TEM. In addition, the STEM technique is less affected by diffraction contrasts, such as thickness contours and bend contours, than TEM imaging. Because of these advantages, the UHVEM-STEM is a powerful technique to observe the defects in bulk specimens.
Bright-field STEM image of a wedge-shaped single crystalline Si (Ref. 2). The thickness changes from 1 µm (left side) to 9 µm (right side). Artificially-generated dislocations are clearly visible up to a thickness of 7.5 µm in the specimen.
Electron holography can quantitatively analyze electric and magnetic field with high spatial resolution.13–20) In ordinary electron holography, a hologram is constructed by interfering electrons transmitted through a thin film sample (object wave) with a reference wave in a vacuum. The phase changes due to magnetic field and electric field are recorded in the hologram, and they are experimentally separable. On the other hand, an extra geometric phase shift occurs in the vicinity of lattice defects, which is a critical issue in data acquisition and analysis. While ordinary holography uses a transmitted beam (the origin “000” of the reciprocal lattice space) as the object wave, dark-field electron holography uses a diffracted beam that contains distortion information. By interfering the object wave transmitted through the distorted region with the reference wave transmitted through the undistorted region, the extra geometric phase shift can be recorded in the hologram. For non-magnetic materials, a two-dimensional strain map can be extracted by analyzing the dark-field electron hologram.21)
Cho et al.6) investigated the effects of antiphase boundaries, where the geometric phase shift due to strain is negligibly small, on the dark-field electron holography using a B2-type Fe70Al30 alloy. Figure 2 shows the change of the phase shift across the antiphase boundary obtained from reconstructed phase images of the holograms. The hologram was taken using (a) the 000 transmitted beam, (b) 100 superlattice reflection, and (c) 200 fundamental lattice reflection of the electron diffraction pattern. A significant phase shift of ∼2.3 rad was observed at the antiphase boundary in Fig. 2(b), whereas no phase shift occurs in Fig. 2(c). This suggests that the undesired geometric phase shift can be suppressed in the dark-field electron holography using the fundamental lattice reflections. This is useful for analyzing magnetic information at the antiphase boundaries in ferromagnetic materials.
Phase shift at the antiphase boundary obtained from reconstructed phase images of the holograms (Ref. 6). The holograms were taken using (a) the 000 transmitted beam, (b) 100 superlattice reflection, and (c) 200 fundamental lattice reflection of the electron diffraction pattern. It is apparent that the phase shift is suppressed in (c).
Conventional electron microscopy techniques, such as bright- and dark-field TEM and high-resolution TEM observations, are still useful for analyzing the materials, and several researchers reported the results of the structural characterization of metals,3,5,8) semiconductors,4) and composite materials7,9) in the seminar. Electron diffraction can provide scattering information over a wide range of the reciprocal lattice space, due to the short wavelength of high-energy electrons. This is useful for obtaining radial distribution functions and atomic pair-distribution functions which express amorphous structures as the existence probability of atoms as a function of a distance from the center of an arbitrary origin atom. Higashiyama et al.22) performed the structural analysis of amorphous Ge1−xSnx via electron diffraction radial distribution analysis. Figure 3(a) shows atomic pair-distribution functions, g(r), of sputtered amorphous thin films with different Sn concentrations. Prominent peaks appear at ∼0.25 and ∼0.40 nm, which correspond to the first and second coordination shells, respectively, of the amorphous network. On the other hand, the g(r) converges to unity at the longer distance side, indicating the lack of long-range order. A closer examination of Fig. 3(a) reveals that the location of the peaks moves to the longer distance side with increasing the Sn concentration. Figure 3(b) shows the magnified first peak in the g(r) of Fig. 3(a). The peak location of the specimen with a concentration of 9.1 at%Sn is almost the same as the bond length of Ge–Ge atomic pair. The peak height due to the Ge–Ge atomic pair decreases with increasing Sn concentration, while the number of Sn-related atomic pairs increases. On the basis of the radial distribution function analysis and high-resolution TEM observations, it was suggested that Ge and Sn are mixed within the first coordination shell and no remarkable phase separation occurs. Knowledge of the amorphous structures as well as their structural changes23–25) is of technological important for fabricating polycrystalline Ge1−xSnx thin films which are anticipated as a channel material for high performance thin film transistors.26,27) Selected-area electron diffraction taken from a relatively wide area gives average structural information of amorphous materials. On the other hand, atomic clusters embedded in the amorphous matrix are detectable by focusing an electron beam on a material. Indeed, nanoscale inhomogeneities in amorphous materials have been successfully detected by using a highly parallel, high-brightness angstrom beam.28,29)
(a) Atomic pair-distribution functions and (b) magnified first peak of amorphous Ge1−xSnx thin films with different Sn concentrations (Ref. 22). The first peak consists of Ge–Ge, Ge–Sn, and Sn–Sn atomic pairs, and its position moves to the longer distance side with increasing the Sn concentration. (Reprinted from J. Appl. Phys. 125 (2019) 175703, with the permission of AIP Publishing.)
In summary, this article briefly reviewed the papers presented in the 12th Japanese-Polish Joint Seminar on Micro and Nano Analysis. In addition to the topics covered here, the following impressive presentations were given at the seminar: in situ plastic deformation studied by time-resolved three-dimensional electron tomography,30) atomic-resolution two-dimensional elemental mapping by electron energy-loss spectroscopy,31,32) and determination of lattice strain of gold nanoparticles by high-angle annular dark-field observations.33) The improvement of the spatial resolution of (S)TEM and the development of highly efficient analytical instruments are still ongoing, and there is no doubt that electron microscopy techniques will become more important in the research field of materials science and engineering.
This work was supported by JSPS Bilateral Joint Research Projects/Seminars.
