S Motoki

Transmission electron microtomography in soft materials

This review summarizes the recent advances in three-dimensional (3D) imaging techniques and their application to polymer nanostructures, for example, microphase-separated structures of block copolymers. We place particular emphasis on the method of transmission electron microtomography (electron tomography for short; hereafter abbreviated as ET). As a result of recent developments in ET, truly quantitative 3D images of polymer nanostructures can now be obtained with subnanometer resolution. The introduction of scanning optics in ET has made it possible to obtain large amounts of 3D data from micrometer-thick polymer specimens by using conventional electron microscopes at a relatively low accelerating voltage, 200 kV. Thus, ET covers structures over a wide range of thicknesses, from a few nanometers to several hundred nanometers, which corresponds to quite an important spatial range for hierarchical polymer nanostructures. ET provides clear 3D images and a wide range of new structural information that cannot be obtained using other methods. Information traditionally derived from conventional microscopy or scattering methods can be directly acquired from 3D volume data. ET is a versatile technique that is not restricted to only polymer applications; it can also be used as a powerful characterization tool in energy applications such as fuel cells.

Dependence of beam broadening on detection angle in scanning transmission electron microtomography

It has been shown that scanning transmission electron microtomography (STEMT) is quite effective for observing specimens with thicknesses on the order of micrometers in three dimensions (3D). In STEMT, the specimen is scanned using a focused electron beam, and the electrons from the convergence point are detected at the detector placed at a certain detection angle. Until recently, a wide detection angle corresponding to the mode often called the dark-field (DF) mode was mainly used. Although the detection angle can vary and is one of the crucial experimental factors in STEMT, its effect on 3D reconstruction has never been discussed from either an experimental or a theoretical viewpoint. Moreover, the effectiveness of another mode of electron tomography, transmission electron microtomography (TEMT), is not clear. In the present study, a polymeric specimen, an acrylonitrile butadiene styrene resin, with a thickness of ~1 mum and a fixed volume was observed using three different modes, namely, TEMT, small detection-angle STEMT referred to as bright-field STEMT, and DF-STEMT, in order to examine their advantages and disadvantages by observing multiple scattering of electrons inside the specimen.

Hole Free Phase Plate Tomography for Materials Sciences Samples

We report, for the first time, the three dimensional reconstruction (3D) of a transistor from a microprocessor chip and roughness of molecular electronic junction obtained by electron tomography with Hole Free Phase Plate (HFPP) imaging. The HFPP appears to enhance contrast between inorganic materials and also increase the visibility of interfaces between different materials. We demonstrate that the degree of enhancement varies depending on material and thickness of the samples using experimental and simulation data.

Hole-Free Phase Plate Energy Filtering Imaging of Graphene: Toward Quantitative Hole-Free Phase Plate Imaging in a TEM

Hole free phase plate (HFPP) imaging has been developed to increase TEM phase contrast for qualitative imaging purposes [1] in the sense that the image intensity is not related to sample properties in a simple manner. Here we report simultaneous utilization of HFPP and energy filtered TEM (EFTEM) thickness mapping as a first step toward quantitative HFPP imaging, using a contaminated graphene film as test specimen. The data were collected in a JEOL JEM 2200 FS TEM operated at 200 kV with a Schottky electron source and an in-column omega filter. The HFPP is placed at the in-gap objective aperture plane, enabling us to utilize microscope magnification between 1500x and 1Mx. The carbon phase plate was heated during operation to about 250 °C and data were collected after the HFPP was allowed to settle for about 100 s. Figure 1 shows conventional a) and HFPP b) images of the graphene film at 50kx magnification. The contrast in a) is higher in b). A thickness map obtained from logarithm of the ratio between elastic and inelastic electrons is shown in c) for data obtained without HFPP [2]. A HFPP-EFTEM thickness map obtained by the same log ratio method of zero-loss filtered and unfiltered HFPP images is shown in d). e) shows a HFPP-EFTEM thickness map obtained from the area marked by the red rectangle in d) at 300 kx magnification. The maximum thickness variation in e) is about 0.06 inelastic mean free path λ1. A EFTEM thickness map without HFPP over the same are as in e) did not yield interpretable data and was difficult to align due to presence of fringes arising from spherical aberration. Profiles along the lines indicated in c) and d), are shown in f). The upper profile in e) corresponds to the HFPP thickness map (red, dashed) from d) and the lower (blue, solid) profile to a thickness map without HFPP, derived from c). The HFPP-EFTEM thickness map appears to provide more spatial resolution that the standard EFTEM thickness map and thinner contamination islands can be observed in the HFPP-EFTEM map more clearly than in the standard EFTEM map. The profiles in f) indicate that the addition of the HFPP in the beam for the log-ratio EFTEM thickness mapping results in addition of ~ 0.04 inelastic mean free paths λ across the field of view. The features in the HFPP-EFTEM thickness map appears to be more localized than in EFTEM thickness map without HFPP. Taking into account the known thickness of the carbon HFPP of ~13 nm, the 0.04 λ2 implies ~0.4 mrad collection angle for an object placed in the objective aperture plane rather than in sample plane. Assuming an inelastic mean free path λ1 = 114 nm for a 200 keV beam in carbon at the sample plane (100 mrad collection angle adequate for no objective aperture), the estimated thickness in the sample region is 17.6±0.5 nm in HFPP-EFTEM (marked with yellow band in f) and 13.5±0.5 nm from the standard EFTEM thickness map. 842 doi:10.1017/S1431927617004871 Microsc. Microanal. 23 (Suppl 1), 2017

Optimum optical condition of Tomography for thick samples

In the soft materials science, it is known that structures of the order of 10 ∼ 100 nm significantly influence various properties of materials. Transmission Electron Microtomography is one of the most powerful tools for structural analysis of threedimensional entities in such materials. Structural analysis of those materials requires thicker samples than those examined so far due to the larger internal structures.