Electron Microscopies

A totally different approach to get information in the nanometer regime is based on the particle-wave dualism of the electron, which was proposed in 1924 by de Broglie [111]. This offers the possibility to use electrons as a probe in a similar way as light with a wavelength A depending on the electron mass m and velocity v:

According to Abbe, the resolution A of an imaging system depends on the wavelength, the diffraction index of the area between sample and objective n, and the angle a under which the light passes through the objective:

The term n ■ sin a is called numerical aperture. This could only be varied to a small extent, so the resolution of light microscopy with typical wavelengths around 500 nm is limited to 200 nm. The much lower wavelength, which electrons can reach, offers therefore a much better resolution. In electron microscopy the glass lenses have to be replaced by electrical and magnetical lenses, which deflect the electron beam in the same way as glass lenses do the light. Because of the aberration of these electromagnetic lenses, the resolution in electron microscopy is limited to 0.2 nm.

In the sample a variety of different interactions between the probe electrons and the sample atoms take place:

• transmission of primary electrons

• secondary electron emission

• Auger electron emission, a special type of secondary electron emission

• electron backscattering

• absorption of electrons

• characteristic X-ray emission

• emission of light

One big disadvantage of electrons as a probe is the charging of the sample during electron bombardment. In principle only conducting samples can be imaged. In a special case nonconducting samples can be imaged if the number of the absorbed electrons is equal to the number of secondary electrons leaving the sample. In all other cases the sample has to be coated by metal or a conducting replica has to be made. Both methods may give artifacts.

There are two approaches to using electron beams for microscopy. The first is the transmission electron microscope realized by Ernst Ruska in 1933 [112, 113]. The TEM can be compared with a transmitted-light microscope: the electrons pass through a thin sample [114] and can be detected at a fluorescence screen. Because of the strong absorption and scattering processes which take place if free electrons interact with matter, only very thin samples in the range of 10 nm to 100 nm thickness can be passed by electrons.

As in a light microscope, there are different mapping modes [111]:

1. Bright-field Image: The image is made of electrons which directly pass through the sample. The image plane of the objective is mapped by the projector lens. The contrast is given by the loss of electrons by scattering and absorption on their way through the sample. It depends on the density of the sample and the atomic number of the sample components. In Figure 7 two examples of this imaging technique are presented.

2. Diffraction Image: This uses also the electrons which directly pass through the sample. But in this mode the focal plane of the objective is imaged by the projector lens. By this one gets a diffraction image which yields information on the crystal structure of the sample in a similar way to an X-ray-Laue-diffraction image (Fig. 8).

3. Dark-field Image: The electrons which scattered inside the sample are imaged on the screen and the electrons which pass the sample in a direct way are filtered out.

The TEM should theoretically reach a resolution which is 1,000,000 times better than light microscopes. The lens aberration, especially the spherical and chromatical aberration, decreases this theoretical value to 1,000 times better resolution in practice. Therefore, points which are separated by 0.2 nm can be imaged.

Furthermore, energy-filtering transmission electron microscopy [115] or TEM with attached energy dispersive X-ray analysis (EDX) is used which gives the elemental composition.

There are three main disadvantages of the TEM:

1. Only very thin samples can be investigated.

2. The samples have to be stable under electron irradiation.

3. The whole TEM has to be held under high vacuum conditions or one has to build a special environmental cell for the sample which is differentially pumped against the other parts.

The second approach to use electrons as a probe for imaging is the scanning electron microscope (SEM) [116],

Figure 7. Two typical TEM images of TiO2 nanoparticles (particle diameter 16 nm, crystal modification anatas). Picture (a) is a low-resolution image where the distribution of the nanoparticles is visible, while the resolution in picture (b) is much higher with clearly visible crystal plates (line-shaped structures) inside the nanoparticles (round structures) [246].

Figure 7. Two typical TEM images of TiO2 nanoparticles (particle diameter 16 nm, crystal modification anatas). Picture (a) is a low-resolution image where the distribution of the nanoparticles is visible, while the resolution in picture (b) is much higher with clearly visible crystal plates (line-shaped structures) inside the nanoparticles (round structures) [246].

Figure 8. Diffraction pattern of the sample in Figure 7. The rings have their origin in the diffraction at different crystal planes. The radii of the rings are correlated with the Miller indices of their diffraction plane [246].

which was invented by Knoll and v. Ardenne [117] in the 1930s. The first realization to image a surface was described by Knoll and Thiele [118]. This microscope scans a very well-focused electron beam over the sample surface. The electrons which leave the sample surface are accelerated by a voltage to the detector, for example, an Everhart-Thornley detector [111], which transforms the electrons into an electrical signal. This is used to control the intensity in a cathode-ray oscillograph while the same signal that scans the electron beam in SEM is used as deflection signal in the oscillograph. So an image of the electrons leaving the sample is obtained.

The resolution in SEM is estimated by the area where the electrons leave the sample. The dimensions of this area depend on the spot radius of the electron beam on the sample surface and the scattering processes inside the sample.

In SEM there are several kinds of electrons which could be detected:

1. Secondary electrons which are produced by primary electrons near the surface with enough energy to pass through the surface barrier. These produce the highest resolution.

2. Secondary electrons which are produced by backscat-tered electrons. These produce the lowest resolution.

3. Secondary electrons which are produced from the backscattered electrons out of the pole shoes of the last lens.

4. Primary electrons which hit the aperture and produce there secondary electrons.

The last two cases increase only the noise without any further information on the sample.

The backscattered electrons themselves are too fast to reach the secondary electron detector. They can, however, be also used for imaging with a different detector. The backscattered electrons give a high material contrast because their number depends on the nuclear charge.

The resolution of the SEM depends on how well the electron beam is focused on the sample. This could give a resolution of 3-6 nm. The depth of focus can reach up to 1 mm, compared to only 100-1000 nm in a light microscope. An example of a SEM image is given in Figure 5b.

A review of spatially resolved spectroscopies in electron microscopy for the study of nanostructures of different metals, semiconductors, and biomaterials is given in [119].

0 0

Post a comment