Local electron microanalysis in microprobe and electron microscopes

V. Starý

Fac. of Mech. Engn, Czech Technical University in Prague, CZ121 35 Prague 2, Czech Republic

stary@fsik.cvut.cz

Electron microanalysis use the electron beam to excite X-rays and deduce from their intensity the qualitative and quantitative elemental composition of material. As analytical technique, it can be incorporated in electron microscope to utilize the production of X-ray photons in observation both of thin and bulk samples in various types of electron microscopes. The quantitative method for bulk samples was firstly used by R. Castaing in France, the method was improved by K. F. J. Heinrich in USA. The special devices for application of this method are usually called “microprobe”, so the method is often called “Electron probe microanalysis” even though the microprobe is the name for very thin electron beam, used for local microanalysis. After some theoretical information the basic possibilities of the method are reviewed, and some practical examples are given.

The atomic concentration in bulk sample are calculated from the intensity of emitted X-ray radiation using our knowledge of electron-matter interaction, i.e., scattering and deceleration of electron and absorption of X-ray in sample. During the way of electron through the matter, electron interacts with atomic nuclei and with electrons by elastic and inelastic way. Due to the elastic scattering the electron beam mainly increase its width. Simultaneously, the part of electrons can change the direction back to the surface. Inelastic scattering couses mainly the energy losses. The angular deviations are much lover then for elastic scattering, even though they exist. The energy losses are mainly given in materials with „free“ electrons by scattering due to the plasmon excitation, in other way by the interband and intraband excitations. Generaly, the energy decrease is described by „stopping power“. It gives approximately the maximal depth of electron trajectory. In bulk sample, the thin beam creates in material the shape of approximately pear, called “interaction volume”.

For electron microanalysis, we should know:

· For which elements, compounds and alloys does the electron microanalysis give the reasonable results and for which not,

· What is the minimal content (concentration, number of atoms, mass) of element we are able to measure,

· If the bulk material is homogeneous (and we define the mean concentrations) or inhomogeneous (and we define the distribution of elements in sample),

· Especially for film, we should decide, if we are interested on average element content or the dependence of concentration on depth in sample,

· If the method is destructive with disordering of material or non-destructive, etc.

There are two basic systems for detection and analysis of X-ray from sample: the first one separates the X-rays according to energy (EDS system), the second separate the X-rays according the wavelength (WDS system). Both systems are widely used in the practice.

From the peak energy in EDS spectrum or the angle of maxima in WDS spectrum we can define the types of elements in sample and the qualitative analysis can be proceed (with some limitations). For precise knowledge, we must have the quantitative limits of detection, because in the definition of non-presence of sample we should also tell what is the minimal content we are able to proove. For this reason, the critical value, the minimaly detected amount and minimal measurable vales of concentration should be defined for each element. For thin samples, when the number of collision of electrons with atoms in thin sample is low (e.g., one collision per electron) the energy losses are relatively low and electron energy is relatively constant, the semiquantitative analysis can be made directly from the intensities, supposed Sc_i = 1.

At quantitative analysis of bulk material we need to calculate the concentrations (i.e., mass fractions c_x) of presents elements. The physical processes what should be taken into account are as follows:

1. slowing down of electrons;

2. elastic scattering at atoms, connected with change of direction, including the possibility of backscattering;

3. ionisation of atoms with possible emission of X-rays;

4. indirectional ionisation of atoms by both characteristic and bremsstrahlung radiations and fluorescence;

5. absorbtion of generated X-ray radiation in the path through the sample to the detector.

For calculation of concentrations usually two methods are used, ZAF and Phi-Ro-Zet. The base of both methods will be presented.

The depth of information is given by the size of interaction volume. Causing the lateral resolution, the method allows two modifications: local microanalysis and chemical mapping. At local microanalysis, we position the electron beam on defined place of sample and we obtain the information from one “point” of sample. Because we usually see the image of surface as (secondary or bacscattered) electron image, we can position the beam on places, which are interesting for study. At chemical mapping of area, the beam moves in some net of points and matrix of concentrations of individual elements is created. There is the possibility to measure in each point of matrix all the spectrum, which is time spending or to measure only selected energy windows for some elements (or background). The next possibility is the movement of beam along the line, for example to cross the interface. In this case we shall see for example the drop of concentration of element in question at interface.

As conclusion, the X-ray nicroanalysis is very important method for the study of composition of materials as well as of biological objects. The reasonable results are obtained in the range of tenths of percent. Moreover, due to possibility to use very thin electron beam, very good lateral resolution can be obtained.

General references on electron microscopy and electron microanalysis

1. Reimer L.: Transmission Electron Microscopy, Springer, Berlin-Heidelberg 1984, 521 s.

2. Reimer L.: Scanning Electron Microscopy, Springer, Berlin-Heidelberg 1998, 527 s.

3. Hulínský V., Jurek K.: Zkoumání látek elektronovým paprskem, SNTL, Praha 1982, 404 s. (in Czech)

4. Goldstein J.I. v: Introduction to Analytical Electron Microscopy, eds. Hren J.J. et al., Plenum Press, New York 1979, 385 s.

5. Goldstein J.J., Newbury D.E., Echlin P., Joy, D.C., Fiory C.E. Lifshin E., Scanning Electron Microscopy and X-Ray Microanalysis, Plenum Press, New York 1981, 673 s.

6. Newbury D.E., Joy, D.C., Echlin P., Fiory C.E., Goldstein J.I., Advanced Scanning Electron Microscopy and X-Ray Microanalysis, Plenum Press, New York 1986, 454 s.

For a critical reading of this manuscript and for helpful discussions, we would like to thank Dr. R. Kužel. This work was supported by Czech Science Foundation (GA-CR) project No. P108/10/1858 and project No. KAN101120701 of the Grant Agency of the Academy of Science of the Czech Republic. We are grateful to Mr Robin Healey for English language review.