Charging of a SiO2 particle induced by electron illumination was investigated by changing the illuminated area of the particle and its support film through control of the position of the mask plate inserted in a transmission electron microscope illumination system. The electric fields around the charged SiO2 particle were analyzed using electron holography. The amount of charge was evaluated quantitatively by comparing the reconstructed phase images with the simulated phase images. When the support film was not covered against the incident electron beam, secondary electrons emitted from the conductive support film were attracted to the charged particle, resulting in particle discharge. In contrast, when the support film was completely covered, secondary electrons were not emitted from the film, so that the particle remained positively charged.
When non-conductive specimens such as metal oxides are observed using transmission electron microscopy (TEM), secondary electrons are emitted from the specimens, leaving them positively charged [1–4]. Excess charging can result in morphology changes and specimen drifts, which makes it difficult to conduct accurate TEM observations and analyses. It is, therefore, important to understand the mechanism of charging.
Several methods have been proposed to suppress charging effect, such as coating the specimens with conductive materials [5,6] and electron irradiation to conductive objects near the specimen [7,8]. With a conductive coating, electrons are supplied to the specimen through the coating, so that charging is suppressed. In reference , significant modifications of the object wave were achieved by controlling the size and position of the condenser aperture.
In this study, we discuss illumination-induced charging of a SiO2 particle in detail using a mask plate inserted in the electron microscope illumination system to precisely control the illuminated area. The amounts of charge on the SiO2 particle were quantitatively evaluated by comparing observed and simulated data, and the changes are discussed as a function of the mask position.
Amorphous SiO2 particles were synthesized by the seed growth method . Synthesized particles mixed with ethanol were dropped onto a micro-grid support film. Figure 1 shows the region of interest of the TEM observation specimen, a SiO2 particle (ca. 320 nm in diameter) attached to a curled surface of the support film. The support film was ripped off using a glass probe. A suitable SiO2 particle on the curled surface was searched under the electron microscope view.
Figure 2a shows a schematic illustration of electron optics for electron holography. A mask plate was inserted in the illumination system and the illuminated area was controlled by moving its focused shadow . Figure 2b shows an example of the TEM image when the mask is inserted. Using shadows of the mask plate and condenser aperture, the illuminated area was selected. We note that the shadow of the mask plate is almost focused on the specimen height, so that the Fresnel diffraction at the image plane from the mask plate was made much smaller than that from the condenser aperture. Both of shadows were moved precisely through a beam-shift function using double deflectors.
Electron holography observation of the charging effect was performed using a transmission electron microscope (HF-3000X, Hitachi High-Technologies Co.) equipped with a cold field-emission gun and three biprisms in the imaging system. Two biprisms were used simultaneously to obtain holograms without Fresnel fringes .
The TEM operation conditions were as follows: The voltages of the upper and lower biprisms were set to −35 V and −50 V, respectively, to produce a 15 nm fringe pitch of the electron hologram. The accelerating voltage was 300 kV and a charged coupled device camera (4080 × 4080 pixels) was used to continuously acquire holograms by moving the mask shadow. The hologram acquisition interval was 6 s and the exposure time was 2 s.
Computer simulations of the reconstructed phase images around the SiO2 particle were carried out to evaluate the amount of charges associated with the SiO2 particle. The commercially available code ELFIN (ELF Corporation) was used, in particular, the ELF Integral Element Method was employed to obtain electric charges on the particle and electric fields around the particle. To calculate the electric field generated by the charged specimen, a relatively simple model was used: The SiO2 particle and support film were assumed to be conductive and the potential of the support film was set to 0 V, while that of the particle was varied. Based on the calculated data, reconstructed phase images were simulated by considering electron phase modulations in the reference wave region .
Result and discussion
Figure 3 shows results of electron holography observation for the SiO2 particle. Figure 3a is an electron hologram obtained when the mask was inserted; the SiO2 particle and a part of the support film were illuminated by the incident electron beam. Figure 3b is the reconstructed phase image obtained from the hologram of Fig. 3a, where the contour lines indicate a positively charged SiO2 particle. Figures 3c–h show examples of reconstructed phase images obtained by gradually moving the mask shadow from left to right with each moving step of ~27 nm. When the support film was partially illuminated (Fig. 3c), the contour lines around the particle did not change from those of Fig. 3b. However, when the support film was completely shielded (Fig. 3d), the density of the contour lines around the particle increased, which indicate an increase in the electric potential projected along the direction of the incident electrons, i.e. an increase in particle charging due to support film shielding. When the particle was completely shielded, charging of the particle decreased (Figs. 3e and 3f). When the mask shadow movement was reversed so that the particle was again illuminated (Fig. 3g), the particle charging increased again showing the behavior similar to that of Fig. 3d. Finally, when the support film was partially illuminated again (Fig. 3h), the particle charging was suppressed to the level almost the same as that shown in Fig. 3b. These results indicate that secondary electrons emitted from the illuminated support film suppress charging of the SiO2 particle.
Simulation of the reconstructed phase images were carried out to evaluate the amount of charges on the SiO2 particle. Figure 4a shows a three dimensional configuration of the specimen model. Figures 4b–h show the simulated phase images reproducing Figs. 3b–h: estimated electric potentials and corresponding charge amounts are shown at the lower left corner of each figure.
More detailed analyses of the amount of charge on the SiO2 particle were conducted as a function of the mask shadow position. Figure 5a shows a schematic illustration of the experimental arrangement; the direction of mask movement for specimen shielding is indicated by the notation of ‘forward’ and ‘backward’. Figure 5b shows the charge amount as a function of the mask shadow position, where hysteresis behavior was observed. This behavior can be explained as follow: The incident electrons illuminated on the particle induce the emission of secondary electrons so that the particle becomes positively charged. Part of these secondary electrons emitted from the illuminated area of the particle and the conducting support film are attracted to the positively charged particle, resulting in discharge from the particle. Secondary electrons emitted from the support film significantly discharge the particle. After the mask shadow completely covered the support film during its forward movement, the amount of charge increased gradually. This means that the discharging by the secondary electrons emitted from the support film still continued, though weak, probably because a part of incident electrons causing Fresnel diffraction hit the support film or secondary electrons emitted from the particle hit the support film causing electron emission. When the particle was completely shielded, the particle charging became smaller. This time-dependent discharging process is probably due to a small charge flow through the interface between the particle and the conducting support film. During the backward movement, at first the charge amount increased with increasing the illumination area on the particle. When the position of the mask shadow became <170 nm, it reduced slowly. The reason for this behavior is due to a start of secondary electron emission from the support film.
Electron holography was used to investigate charging of a SiO2 particle attached to a support film by changing the electron beam illumination conditions of the particle with a mask shadow movement. The reconstructed phase images were compared with simulated phase images to estimate charge amounts. When the support film was shielded by the mask shadow, the electric field around the SiO2 particle increased because the supply of secondary electrons from the conductive support film decreased.
The authors would like to thank the late Shinji Aizawa of the RIKEN Center for Emergent Matter Science (CEMS) for valuable discussions and technical support. We are also grateful to Dr Y. A. Ono for critical reading of the manuscript and for comments and suggestions.
A Grant-in-Aid from the Japan Society for the Promotion of Science (JSPS) through the ‘Funding Program for World-Leading innovative R&D on Science and Technology (FIRST program)’ initiated by the Council for Science and Technology Policy (CSTP); Kakenhi Grant-in-Aid (No. 25249093) from JSPS.