Elements with atomic number (Z) below 11 are called light elements. Until 1990, beryllium-made entrance windows with a thickness of 8µm were used in most energy dispersive spectroscopy (EDS) detectors.
Transmission of X-rays for elements with Z below 11 through these windows is not possible. Different treatment is required for light elements because measuring the X-rays generated by them reliably is difficult, if at all they have been detected.
Ionizing an atom will be difficult when the atomic number is reduced and the ionized atom generated is less likely to emit an X-ray. The result is a weaker signal from the light elements. Longer wavelength X-rays are generated by light element atoms and are easily absorbable within the sample when compared to shorter wavelength X-rays.
Most of the weak signal generated will be absorbed within the sample itself. This means that much of the X-rays from light elements received by the detector are from near the surface of the sample. As a result, sample contamination or coatings used to avoid charging strongly affect the light element analysis.
These effects make light element measurement impossible and the X-ray intensity measurements from light elements more often encounter systematic errors when compared to the X-ray intensity measurements from heavier elements. The dominance of these effects is more when the atomic number is lower.
A special case is carbon, which often deposits as a contaminant on the sample surface in an electron microscope. As a result, reliable carbon measurement is very difficult because analysts are not able to confirm whether the carbon signal is from the sample itself or from the contaminant accumulated during analysis. The following are the options available to quantify light elements:
- Direct analysis
- Calculation by stoichiometry
- Element by difference
- Fixed concentration
Treating light elements like all other elements without selecting any special options in the software is the simplest approach. Making good measurements on most light elements is now possible with today’s silicon drift detectors and digital pulse processors.
This method is effective for elements as light as carbon or boron. However, it has stringent requirements on the flatness and cleanness of the sample surface.
In addition, the credibility of an analysis needs to be confirmed by comparing with analyses of similar materials. This approach is especially useful in the analysis of oxides of materials due to the possibility of existence of multiple oxide states or incomplete oxidized state. The analysis results of an iron oxide (hematite) are shown in Figure 1.
The analysis involved direct measurement of the iron and oxygen peaks using a scanning electron microscope (SEM) and the Thermo Scientific™ NORAN™ System 7 X-ray microanalysis system (Figure 2).
Figure 1. Quantitative results for iron and oxygen in a hematite sample in which both the iron and oxygen peaks were measured directly
Figure 2. The Thermo Scientific™ NORAN™ System 7 X-ray microanalysis system
For direct measurement of light element peaks, it is necessary to consider the following:
- Sample preparation
- Analysis conditions
- The stability of the sample under the beam
- The quantity of carbon contamination being deposited
A direct analysis may be performed with or without standards.
Calculation by Stoichiometry
Stoichiometry involves reactions of maximum amounts of material reacting to completion such that there is neither any surplus nor shortage of any reagent. Stoichiometry in microanalysis evaluates the light element concentration depending on the measured concentrations of the heavy elements in a sample and their known stoichiometric correlations to the light element.
For instance, in the analysis of albite (NaAlSi3O8) sample, Na, Al and Si are the heavier elements, which have known oxides of Na2O, Al2O3 and SiO2. The amount of oxygen present can be calculated by quantifying the heavier elements based on the assumption that all of the heavier elements are fully oxidized.
Figure 3 illustrates this type of analysis. The symbol "S" next to the weight percent result for oxygen reveals that the oxygen result was based on estimation and not a measurement.
Figure 3. Quantitative result for an albite sample in which oxygen was calculated by stoichiometry
For most elements, multiple oxide states can exist. Hence, for selecting an oxidation state of an element to be used in an analysis, element's symbol in the Element Setup periodic table is clicked and then the Advanced button just below the periodic table is pressed.
In the Advanced Element dialog shown in Figure 4, the desired compound is selected to be used for that element and the window is closed. Options are offered for nitrides, carbides, borides, and oxides when available.
Figure 4. Advanced pane showing how to choose the Compound Formula for an element
Figure 4 illustrates selecting FeO as the desired compound for iron to use in estimating an oxygen result from the iron composition. The checkbox "Use compounds for all elements" under the Analysis Setup tab is selected to use this feature (Figure 5).
Calculating an element by stoichiometry for selected elements may be desirable in special cases. In these cases, the "Use compounds for all elements" option should not be checked, but the "Use Compound" option needs to be checked only for the desired elements as shown in Figure 4.
Figure 5. Analysis Setup pane showing the option to “Use Compounds for All Elements”. In this view the option is selected.
It is possible to use stoichiometry or compound analysis only when there are two or more heavier elements in the sample. For instance, the analysis or iron oxide by this method cannot report the oxidation state, but will only report the oxidation state that is set by the analyst in the periodic table.
The direct method or element by difference is recommended for single element oxides or cases where there are mixed oxide states. It is possible to use stoichiometry both with the standardless quantitative method and in analysis with standards.
Element by Difference
Like stoichiometry, element by difference cannot perform direct measurement of light elements but calculates their concentration from the composition of the other elements present in the sample. Here, the analysis of heavier elements is performed using standards and the missing mass is allotted to the designated light element when the result is below 100%.
To use this method, it is necessary to measure the beam current and use the standard, giving the system the data required to determine the missing mass. This method is selected by clicking on the symbol of the designated light element in the Element Setup periodic table and then clicking on the Advanced button below the periodic table. The "By difference" option in the window popped out is then clicked.
Figure 6 illustrates this option being set for boron. Figure 7 depicts a quantitative result where boron was estimated by difference. The symbol "D" next to the weight percent result represents that this element was estimated by difference.
Figure 6. Advanced pane showing the option By Difference selected
Figure 7. Quantitative result for which the value for boron was determined by the Difference method
This method informs the system how much of the designated element exists, which can be crucial to take into account all elements in an analysis. This is frequently applied when a very light element exists at a low level where the effect of small errors has no impact on the total results. In these cases, the Fixed method may be better than omitting the light element altogether. This option set for boron is illustrated in Figure 8.
Figure 8. Advanced pane showing the value of boron set to the fixed value of 3.00%
Before setting this value, the element's symbol is clicked once and then the Advanced button below the periodic table is clicked. A result using the fixed amount of boron is presented in Figure 9. The symbol "F" next to the weight percent result for boron reveals that this result is a fixed value and not a value obtained from measured data.
Figure 9. Quantitative result in which the value of boron was set to the fixed value of 3.00%
Irrespective of which method is employed, light elements need to be accounted for in an analysis even if they are unobservable to the detector in use. X-ray microanalysis is characterized by various strongly nonlinear physical effects.
To obtain a quantitative result from a spectrum, the NORAN System 7 software performs a calculation by making corrections for all of these effects, most of which involve each element influencing the results for the other elements in the sample.
When the system is informed to incorporate some amount of a light element, although it is not measured directly, the NORAN System 7 software takes into account the effects of the light elements on the other elements in the ZAF or Phi-Rho-Z calculations.
This information has been sourced, reviewed and adapted from materials provided by Thermo Fisher Scientific – Materials & Structural Analysis.
For more information on this source, please visit Thermo Fisher Scientific – Materials & Structural Analysis.