Theoretical introduction to scanning electron microscopy (SEM)

A. Electron source and column

  • The source of the electrons used in SEM is a tungsten (W) filament. The filament emits high amount of electrons (electron beam), when it is under voltage.
  • The electron beam is focused by electromagnetic lenses, which can control the beam of electrons in a similar way like the optical lenses are controlling the light.
Electron column

Fig. 2/1.  The electron column

B. Scanning and imaging

  • The imaging of a SEM is quite different than a optical microscope. In optical microscopes the entire image is formed at once, all part of the image is created simultaneously. The SEM's computer controls a scanning coil, which manage the focused electron beam to scan step by step the area of interest on the specimen (Fig. 2/2).
scangif1

Fig. 2/2. Scanning

  • When the magnification is changed, in fact, the size of the scanned area was changed (Fig. 2/3).
magnification_1

Fig. 2/3.  Low and high magnification

  • Seeking on the specimen is done by physical moving of the specimen holder, the scanned area is not moving (Fig. 2/4).
stagegif_1

Fig. 2/4.  Moving the specimen instead of moving the scanned area (and the entire electron column)

C. Electron beam - matter (specimen) interaction

There are different products of the electron beam - specimen matter interactions, the most important:

  • Secondary electrons (SE) are emitted when a beam electron interacts with the electric field of an electron from the specimen's atom. The energy of beam electron can extrude an electron by inelastic scatter from the atom as a secondary electron (Fig. 2/5). Secondary electrons by definition are less than 50 eV, and they are emitted from very shallow depth (10-50 nm, Fig. 2/8) which means these will only provide surface information.
se_1

Fig. 2/5. SE

  • Back scattered electrons (BSE) are emitted when a beam electron interacts with the electric field of the nucleus of a specimen's atom; this will result a change in direction (Fig. 2/6) of the beam electron. During this elastic scatter event there is no significant change in the energy of the beam electron, most of the back scattered electrons retain at least 50% of the beam energy, therefore they can be emitted from higher depth (1-5 μm, Fig. 2/8) of the specimen.
bse_1

Fig. 2/6. BSE

  • If an electron extruded from a lower level (inner) orbit of a specimen atom by a beam electron (Fig. 2/7/I), the vacancy is filled by an electron from higher level orbit (Fig. 2/7/II). The electron from higher level orbit will occupy a lower energy state position, and its surplus energy will be emitted as a characteristic X-ray photon. The energy and the wave length of the emitted X-ray photon is highly Z (atomic number) dependent, therefore it is characteristic to the atom.
x-ray_1

Fig. 2/7. Characteristic X-rays

The beam-matter interaction is characterized via the volume of interaction (Fig. 2/8). The size (depth) of the volume of interaction is determined by the energy of the electron beam and the average atomic number of the specimen material. Higher acceleration voltage cause deeper interaction volume, while higher atomic number of specimen material cause shallower interaction volume.

The shape of the interaction volume is determine the spatial resolution of the SEM images created by using different detectors (smaller is better):

  • Due to their low energy (<50 eV), the secondary electrons has the most shallower escape depth (Fig. 2/8), therefore the width of excited volume SEs is the smallest.
  • The back scattered electrons escape depth is significantly bigger (Fig. 2/8) because of their high energy, thus the excited width of BSEs is much higher.
  • By reason of the highest escape depth, the characteristic X-ray's excited width is the largest.
interaction-volume_1

Fig. 2/8. Electron beam - specimen reaction: the interaction volume

D. Relation between noise and acceleration voltage, beam current & scanning speed

In order to reach the best image quality, must avoid the noise as much as possible. The amount of noise is inversely proportional to amount of signal (SE, BSE and characteristic X-ray) collected by detectors (SED, BSED, EDS), which is depends on the acceleration voltage & beam current, scanning speed (~exposure time), and the material of specimen. Since the material of the specimen is given, acceleration voltage, beam current and scanning speed can be modified to reach the best image quality. Higher acceleration voltage, higher beam current and slow scan speed cause more signal and vice versa.

E. Brightness and contrast

  • Brightness: overall lightness or darkness of the image
  • Contrast: difference between maximum and minimum pixel intensity of the image

F. Conductor and insulator samples

  • Electric conductor samples (e.g. metals) are examined without any special treatment (carbon coating) or special settings (low vacuum mode).
  • Insulator samples usually charges (see point G.), therefore they need carbon/gold coating or low vacuum mode (see details in module 3).

G. Charging, vacuum modes and coating

  • The electrons lose their energy when the beam enters the specimen and if the sample is conductor, the electrons will flow through the specimen holder to the ground, however if the specimen is an insulator, the electrons will remain in there and a negative charge accumulation can be observed. Accumulated charges will a create an electric field, therefore the image will be distorted, shifted and other artefacts will appear. Charging can prevent the proper operation of detectors, hence the result of imaging or analysis can be false.
  • Charging effect can be reduced or avoid with two method:
    • Using low vacuum mode during the SEM analysis
      • In low vacuum mode the specimen chamber is filled with air or water steam at 30-200 Pa pressure. Collision between the electrons from beam and molecules of air or water steam produce cations which are neutralizing the negative charging of the specimen. EDS measurements are not recommended under low vacuum mode because of the disturbing effect of air or water steam molecules in the chamber.
    • Coating the sample with a thin layer of conductive material (most commonly with carbon or gold)
      • Gold coating is used when the aim is to create detailed SE (rarely BSE) images of the specimen.
      • Carbon coating is used when the goal is to analyze composition of the specimen with EDS.

H. Few important characteristics of the different detectors

Secondary electron detector (SED):

  • provides surface information of the specimen
  • best spatial resolution due to shallow escape depth of SEs
  • sensitive to charging effects
  • edge effect (edges are emitting extra SEs, therefore the edges are brighter than the surroundings)

Back scattered electron detector (BSED):

  • provides relative composition information (Z contrast image) of the specimen
  • worse spatial resolution due to deeper escape depth of BSEs

Energy dispersive X-ray spectrometer (EDS):

  • provides composition information
  • worst spatial resolution due to deepest escape depth of characteristic X-rays
  • best results can be obtained:
    • proper working distance is set (see module 3 and 4)
    • properly prepared sample surface (polished, smooth, perpendicular to electron beam)
    • carbon coated surface in case of insulator sample
    • using high vacuum mode

I. Summary

The electron beam formed through the process of emitting is directed, corrected and focused before interacts with the specimen. These complex processes are done by using electromagnetic lenses, scan coils and correction coils. The image formation is done through a scanning process, which has a signal feedback. In other words, electron beam interacts with the matter, the signal arrives after amplification to the computer, which sends another signal to the scan generator to move on, and so the scanning is done in small steps, very differently from the usual optical video recording.

© Ábel Szabó & Sándor Lipcsei

Please publish modules in offcanvas position.

Back To Top