ISIJ International
Online ISSN : 1347-5460
Print ISSN : 0915-1559
ISSN-L : 0915-1559
Note
SEM Observation of Inclusions in Steel Samples Using Fast Cleaning and Modification of the Surface by Glow Discharge
Kouichi Tsuji Kenichi Shimizu
Author information
JOURNAL OPEN ACCESS FULL-TEXT HTML

2013 Volume 53 Issue 11 Pages 1936-1938

Details
Abstract

SEM (scanning electron microscopy) is a useful technique for the observation of the surface morphology of various materials. Compared to TEM (transmission electron microscope), one of the advantages of SEM is easy sample preparation, although the spatial resolution of SEM is normally less than that of TEM. To improve the spatial resolution of SEM observation, it is well known that a low accelerating voltage SEM is an effective technique. We have proposed the application of glow discharge surface treatment before high-resolution SEM observation. An rf-glow discharge was applied with Ar gas for a steel sample just for 6 sec., leading to surface cleaning, that is, removing the surface oxidation or contamination layer. Besides the surface cleaning, a glow discharge sputtering modified the surface of the steel sample depending on crystal orientation. This surface modification was useful for high-resolution SEM observation. The surface of the steel sample was observed by FE (field emission)-SEM with a low accelerating voltage. A fine structure of grains and inclusions in the sample was clearly observed. The density of the inclusions was roughly determined as being 4 × 104/cm2.

1. Introduction

SEM (scanning electron microscopy) has been used in various fields because of its easy operation. It should be also mentioned that an electron beam produces characteristic x-rays that are useful for elemental analysis. Thus, SEM observation in combination with an EDS (energy dispersive x-ray spectrometer) is a powerful tool as well as optical microscope observation. Compared to SEM, TEM (transmission electron microscopy) gives a high-resolution image. However, sample preparation for thin layers is a complicated, time-consuming process. Lateral resolution in SEM images can be improved by reducing the accelerating voltage because the secondary electron emission region is localized on the top surface. In addition, the escape depth of secondary electrons are more reduced by applying a low-voltage SEM, leading to improvement of surface sensitivity of SEM observation. Since FE (field emission)-SEM is useful for obtaining an intense electron beam with a small diameter, FE-SEM with a low accelerating voltage (low-voltage FE-SEM) is a powerful tool for observing surface morphology. Here, we have to carefully consider the surface treatment, because low accelerating voltage FE-SEM is a very surface-sensitive technique. The process requires a new, simple technique for cleaning the surface of the sample, which should fit well with FE-SEM.

It is well known that inclusions in the steel product influence the physical and chemical properties of the material. Therefore, analysis of the inclusions themselves, their distribution in the material, and control of the size and numbers of inclusions are required. Usually, chemical etching is applied to the steel samples subjected to SEM observation. However, such chemical etching is a time-consuming process and requires special skills. In addition, the emission of liquid waste after the chemical etching poses a serious problem, as does contamination originating from the process. Thus, a simpler technique for surface cleaning is required.

Co-author, Shimizu et al. have proposed a glow discharge process for pre-treatment of the steel materials for SEM observation.1,2,3,4) They reported many useful results of combining glow discharge and FE-SEM, which were especially applied because of the corrosion chemistry of the metals. A dc glow discharge with Ar gas causes Ar sputtering on the surface of the cathode.5,6) Usually, this cathode sputtering is strong enough to mix the atoms near the surface, leading to modified surface composition and structure. However, it has been reported that in the case of rf glow discharge, more “moderate” or “mild” sputtering is possible.1,7) In this case, the damage on the surface of the sample induced by Ar sputtering is drastically reduced. This discharge condition is suitable for low-voltage FE-SEM observation. The sputtering speed is quite high compared to that of an ordinary Ar ion beam gun. Another advantage of glow discharge sputtering is a large sputtering area of about 4 mm in diameter, depending on the glow discharge device.

2. Experiment

A Fe-0.1% Ti sample (10 mm × 10 mm, 5 mm thick) was prepared at Tohoku Univ. The surface of the sample was mirror polished with a diamond paste. A GD Profiler 2 (Horiba Ltd.) was applied for surface cleaning of the Fe–Ti sample. This instrument is designed for use with a GD optical emission spectrometer, but we only used it as a tool for glow discharge sputtering. A Fe–Ti disk sample was attached to the cylindrical discharge tube with an O-ring. After evacuation, Ar gas was introduced at a pressure of 650 Pa. An rf glow discharge was operated at an applied power of 20 W. It has been reported that this sputtering condition is suitable to have modified surface of the steel sample depending on crystal orientation.1,2)

A FE-SEM (Ultra55, SII Nano Tech., Carl Zeiss Co.) was used for surface observation. This instrument has a built-in lens-type secondary electron detector and an energy selective backscattered electron detector. The accelerating voltage of electrons was 1.5 kV during SEM observation. This FE-SEM has an energy filter and enables to detect high-angle backscattered electrons (BSE) in scattered angles less than 15 degrees from the incident electron beam. Although the glow discharge device and FE-SEM are installed in different rooms in the same building, the sample subjected to the glow discharge was placed in the vacuum chamber of the FE-SEM for one minute.

3. Results and Discussion

It is reported that the energy of Ar+ ion bombardment in RF glow discharge is estimated to be about 50 eV at the surface of the sample. Therefore, if we could apply such mild sputtering conditions, the surface of the sample would be cleaned without any serious damage caused by ion bombardment, leading to clear SEM observation. Figure 1(a) shows a typical SEM image of a Fe–Ti sample just after it was mirror polished. As shown in Fig. 1(a), an inclusion was observed. Although the surface was highly polished, like a mirror, the SEM image was not clear. This lack of clarity was probably caused by a thin layer of metal oxide. Therefore, an rf-glow discharge was applied to this sample. The time for the discharge treatment was just 6 sec., leading to removal of a thin layer with a thickness of about 10 nm. After glow discharge cleaning of the sample, a SEM image was taken. A very clear image of three inclusions (size: about 500 nm) was observed with grain and sub-grain boundaries, as shown in Fig. 1(b).

Fig. 1.

FE-SEM images of Fe-0.1Ti sample before (a) and after (b) glow discharge treatment of 6 sec.

Figures 2(a) and 2(b) show enlarged SEM images of the complex inclusion, which consists of TiN and TiO2. Since the material is sputtered by glow discharges in different rates depending on the composition and atomic scale structure of the inclusions,8) a different surface morphology was observed with different brightness in the BSE image, as shown in Fig. 2. After detailed analysis, TiN and TiO2 were recognized from the special pattern induced by glow discharge sputtering,9) as shown in Fig. 2(b).

Fig. 2.

FE-SEM images of Fe-0.1Ti sample after GD treatment. A complex inclusion is observed with high resolution.

Figure 3 shows FE-SEM images of the same sample. The image in Fig. 3(a) is a normal SEM image (secondary electron image), where surface morphology was clearly observed. The image in Fig. 3(b) is a high-angle BSE image, which has information on Z (atomic number) and density (ρ), leading to compositional imaging. We applied an electron accelerating voltage of 1.5 keV, where a Z-dependent contrast in images is still significant. Therefore, the BSE image in Fig. 3(b) is a flat image without showing surface morphology of the inclusions. The position and shape of inclusions are clearly shown in a high-angle BSE image. In Fig. 3(c), the information of elemental composition is indicated by color using a commercial available software such as Gatan Digital Micorgraph. Compositional information is highlighted in the high-angle BSE images, since other BSEs with morphological information are totally excluded here by energy-filtering. The image obtained by this technique was named as “Z and ρ image”.1) Therefore, it is easy to recognize the inclusions in the surface of the steel sample. Figure 4 shows a similar BSE image of a large area (140 μm × 100 μm), where five inclusions are easily seen. From this result, the density of the inclusions in this sample was roughly estimated to be 4 × 104/cm2.

Fig. 3.

FE-SEM images of Fe-0.1Ti sample after GD treatment. (a) Built-in lens-type secondary electron image, (b) high-angle backscattered electron image, and (c) color image including Z and density.

Fig. 4.

FE-SEM image of Fe-0.1Ti sample after GD treatment in a large area (140 μm × 100 μm).

4. Conclusions

Glow discharge treatment enabled the sputtering of the large area (140 μm × 100 μm) in a short time. In addition, FE-SEM observation also enables fast surface observation of such a large area with high resolution that is sufficient for observation of inclusions in an area as large as several 100 nm or several 10 nm. Therefore, the combination of FE-SEM and rf-glow discharge surface treatment is a powerful tool for a fast observation of inclusions in steel samples.

Acknowledgment

The Fe-0.5%Ti sample was prepared by Prof. Ryo Inoue at the Institute of Multidisciplinary Research for Advanced Materials (IMRAM), Tohoku University.

References
  • 1)   K.  Shimizu and  T.  Mitani: New Horizons of Applied Scanning Electron Microscopy, Springer, Heidelberg, (2010).
  • 2)   K.  Shimizu and  T.  Mitani: A New World of Scanning Electron Microscopy, Nikkan Kogyo Publishing Production, Tokyo, (2008).
  • 3)   K.  Shimizu,  R.  Payling,  H.  Habazaki,  P.  Skeldon and  G. E.  Thompson: JAAS, 19 (2004), 692.
  • 4)   K.  Shimizu,  G. M.  Brown,  H.  Habazaki,  K.  Kobayashi,  P.  Skeldon,  G. E.  Thompson and  G. C.  Wood: SIA, 27 (1999), 24.
  • 5)   R. K.  Marcus and  J. A. C.  Broekaert: Glow Discharge Plasmas in Analytical Spectroscopy, Wiley, Chichester, (2003).
  • 6)   R.  Payling,  D. G.  Jones and  A.  Bengtson: Glow Discharge Optical Emission Spectrometry, Wiley, Chichester, (1997).
  • 7)   K.  Shimizu,  H.  Habazaki,  P.  Skeldon and  G. E.  Thompson: SIA, 35 (2003), 564.
  • 8)   R.  Behrisch, ed.: Sputtering by Particle Bombardment I; Topics in Applied Physics, Springer, Heidelberg, (1981).
  • 9)   K.  Shimizu: Electrochemistry, 79 (2011), 947.
 
© 2013 by The Iron and Steel Institute of Japan
feedback
Top