Links in this section:

Introduction
Fundamentals of the TEM technique
Beam-sample interaction
The Analytical TEM
Detector Protection
Qualitative Analysis
Quantitative Analysis
Microanalysis Examples (1)
Microanalysis Examples (2)
Microanalysis Examples (3)
Summary

Examples of Microanalysis in the TEM

Quantitative Analysis of areas of varying film thickness

When analysing samples, a number of practical issues arise.

• Samples are often inhomogeneous and it is advisable to take measurements at more than one point.

• Sample self absorption can usually be neglected for X-rays greater than 1.5keV in energy but may be significant for low energy peaks from B,C,O,N,F,Na.

• Therefore, it is expected that if the sample is of a thickness where matrix corrections need to be applied, it is important to enter thickness and density values into the INCA TEM quant set up. Thicknesses may be calculated by a number of imaging, diffraction or EELS techniques in the AEM (see Williams and Carter).

• Some materials are beam sensitive. In these cases, it is necessary to adjust beam current and count times accordingly.

• Any spectral artefacts caused by the microscope, sample holder and sample grid should be taken into consideration.

As an example, Figure 6 shows data collected from an aluminium oxide film of varying thickness. When a spectrum taken from a thin area  is analyzed, it gives a concentration ratio close to that expected from the stoichiometry of the compound. Data collected from a thicker area shows the effect of increased absorption of the O Kα peak. The analysis result would suggest that the composition has changed, whereas this is purely a consequence of not making a correction for specimen thickness. If a sample does have variable thickness then it is important to determine the likely effects of absorption by trying different thickness and densities in the correction program.

Table 1 shows quantitative analysis from an aluminium alloy (Ti-47% Al “A”) analysis using an INCAEnergy TEM system. It can be seen that the results from different points are consistent and the mean value of several point analyses can be used to improve the precision if the material is expected to be homogeneous. A single point would not show whether the analysis was representative of the sample.

Table 1

Back | Next