Determination of Layer Thickness and Composition by WDXRF with the S8 TIGER with MLplus

Today, most sophisticated materials have to be coated with another material to enhance the material’s appearance or properties. These coatings can either be a protective coating or a passivation coating.

The former is used to eliminate the deterioration of a glass or an engine part that is coated to give it supplementary properties such as self-cleaning glass, while the latter is used to avert corrosion.

In many industries, it is important to determine the materials’ layer thickness for a number of reasons. For applications where the coating material needs to impart an extra property, then it has to be ensured that the coating is thick enough to work upon, but it should not be so thick that it changes the material’s fundamental property.

In other applications, coating is expensive. For example, jewelry manufacturers who use precious metals to plate materials have to make sure that the plating is performed as cost-effectively as possible, and at the same time they should be able to estimate the proportion of precious metal utilized in the plating process to price it properly.

While different techniques are available to measure the composition and thickness of layers on substrates, most of these techniques are destructive, for instance incising cross sections and examining them with TEM, SEM, or gravimetric/wet chemical methods.

X-Ray fluorescence spectroscopy,XRF, provides a simple and non-destructive method to measure a layer’s thickness on a specific substrate. In this method when an X-ray beam travels via a material, the said material will have an enhancing or absorbing impact on that beam.

If the composition of the intermediate material is known, it will help to determine and calculate the effect this will have on the X-ray radiation beam. All of the calculations and measurements are based on a complicated mathematical process called Fundamental Parameters (FP).

Software MLplus

MLplus is a user-friendly software for examining multilayer or thin layers samples with the S8 TIGER WDXRF system. In different types of applications, such as painting, coating, surface engineering, oxidation, silicon wafers, and analysis of corrosion products, the MLplus software can be utilized to regulate layer composition and thickness of multilayer and single systems for production control or research purposes. The choice of the calculation model is the vital part of the multiplayer analysis. As such, there are two models:

  • Absorption model – in this model, a fluorescence line from a single element indicating the material below the analyzed layer is quantified, and its absorption is established.
  • Emission model - in this model, the intensity of a powerful fluorescence line from a single element indicating the layer is utilized for measuring the thickness of the layer.
  • Optimize the sum - This model can be utilized for layers that include more than one element.

In all the instances, the MLplus software is based on full fundamental parameter (FP) calculations and S8 TIGER’s built-in standardless software QUANT-EXPRESS can be utilized for all assessments. This helps to determine the composition and thickness of the layer without having to perform prior calibrations and search for costly standards.

Expanding the possibilities of SPECTRAplus, the MLplus is an optional software module designed to study samples of single and multilayer. It is capable of measuring the layer composition and thickness in multilayer samples from several atomic layers to up to µm or mm range. It is not the software features, but the elements existing in layers and bulk that determine the number of layers.

A full FP approach is used by the MLplus software. The measurement of composition and thickness in the MLplus software is based on the SPECTRAplus standardless calibration - there is no need for any particular multilayer standard.

Using the “Interactive mode” provided in the MLplus software, users can characterize the structure of the sample, such as which element is present in which layer, usual values for composition/thickness etc., and establish the most optimum element lines for the given measurement. As soon an interactive assessment has been successfully set up, the parameters can be easily recorded to perform an automatic assessment of analogous samples.

Validating the Choice of Emission Model

Clicking the "Layer" tab, then the "Absorption bar" button will show the Absorption window, containing the quantity of energy taken in by the layer for individual lines, mass absorption coefficients µ, and the d90% limit thickness (90% absorption path) of every line for this layer. Users can subsequently compare this with the layer’s thickness, and see if they are within the precision limit (less than 3.d90% for the determination by absorption and less than d90% for the determination by emission).

The intensity of Zn KA1/ML relies on the thickness of the layer: The software measures a d90% of approximately 70.7 µm. The software is capable of computing 1% transmission value for 3 µm layer thickness, meaning that the layer absorbs the Zn LA1 signal (Figures 1- 5).

For a value of 12 µm, the software calculates a transmission value of around 57% for the Zn KA1/ML line.

Figure 1. For a value of 12 µm, the software calculates a transmission value of around 57% for the Zn KA1/ML line.

Change the value to 3 µm, the transmission value changes to 87%.

Figure 2. Change the value to 3 µm, the transmission value changes to 87%.

The emission of Zn KA1/ML can be used for calculating the thickness of the layer.

Figure 3. The emission of Zn KA1/ML can be used for calculating the thickness of the layer.

Look at the Zn LA1/ML line. The energy of Zn LA1 is much lower than the energy of Zn KA1: •E(Zn LA1) = 1.0keV •E(Zn KA1) = 8.6keV

Figure 4. Look at the Zn LA1/ML line. The energy of Zn LA1 is much lower than the energy of Zn KA1:
•E(Zn LA1) = 1.0keV
•E(Zn KA1) = 8.6keV

The software provides a value of d90% of about 2.14 µm. For layers whose thickness is higher than 2 µm, the line Zn LA1/ML cannot be used for calculating the layer thickness.

Figure 5. The software provides a value of d90% of about 2.14 µm. For layers whose thickness is higher than 2 µm, the line Zn LA1/ML cannot be used for calculating the layer thickness.

Application Details

The precision of the MLplus software was tested with certain simulated, model samples:

  • Aluminum substrate with organic layers
  • Copper substrate with aluminum layers
  • Titanium substrate with aluminum layers

The actual example of aluminum drinks cans were also tested. To do this, the cans should be coated with an organic polymer to shield it from the drink it will contain. There are many carbonated drinks that have extremely low pH values. Such low pH values can corrode the metal cans in a short period of time, leaving some amounts of metals in the drinks or causing the metal can to fail completely.

Titanium and copper substrate with aluminum layers

  • Layers: Aluminum foil
  • Substrate: Pure Cu or pure Ti

Before measurements, a Vernier gage was used to measure the thickness of the aluminum foil and was found to be 10.6 µm. The initial sample included an aluminum foil sheet with the pure copper or pure titanium substrate behind it. Further samples included more aluminum sheets, with up to six sheets.

Aluminum substrate with organic layers

  • Layers: 2.5 µm MYLAR film
  • Substrate: Pure Al

Just like the way the metal layered samples were built, pure aluminum (up to a maximum of six layers) was added with MYLAR layers to imitate an organic layer on a metal substrate.

Aluminum drink cans

In order to discover the thickness of the protective organic layer on the interior of the can, three examples of cans used to contain carbonated soft drinks were examined in detail. FTIR technique, using a Bruker instrument was applied to find out the composition of this organic layer. In all instances, a wet-and-dry paper was used to clean the second piece of each metal, to remove the organic layer and show that a variation can indeed be seen in the thickness of the organic layer.

Measurement Conditions

Utilizing the Full Analysis measurement technique included with the standardless QUANT-EXPRESS package, all samples were determined on an S8 TIGER (4 kW) instrument, under vacuum. Given the fact that very high signal intensities will be provided by most samples, the full analysis technique was selected because the automatic current reduction was enabled to control the saturation of the detector.

Results

Titanium substrate with aluminum layers

A model containing an aluminum layer and a titanium substrate was built. Using the Ti-Kβ line, an absorbance model was selected to establish the aluminum thickness. This is attributed to the extremely high count rates seen for the Ti-Kα. The quantified layer thickness values for this replicated sample are shown in Table 1. A good agreement can be seen between the computed value and the predicted value from the MLplus software.

Table 1. The calculated layer thickness values for the simulated sample of aluminum layers on titanium substrate.

No. Layers Calc. Thickness (µm) Ti-Kα Intensity (kcps) Ti-Kβ Intensity (kcps)
1 10.4 1164.3 206.19
2 20.1 384.3 84.82
3 29.6 131.2 35.69
4 39.0 45.4 15.07
5 47.8 15.9 6.99
6 57.4 5.7 3.05

Copper substrate with aluminum layers

Aluminum layers and a copper substrate were used to set up a model. Here, the absorption model was again used to establish the aluminum thickness. Also, Cu-Kβ line absorption was again utilized because of the high signal intensities seen for the Cu-Kα (Table 2).

Table 2. The calculated thicknesses observed with increasing number of aluminum foil layers on a copper substrate. The results are in excellent agreement with the expected value of 10.6µm per aluminum foil layer

No. Layers Calc. Thickness (µm) Cu-Kα Intensity (kcps) Cu-Kβ Intensity (kcps)
1 10.9 18134.15 2895.98
2 20.2 14025.61 2460.93
3 30.4 10775.20 2061.32
4 40.1 8541.27 1751.70
5 50.0 6810.24 1479.44
6 60.1 5426.13 1235.79

Aluminum substrate with organic layers

Organic layers (C10H8O4, density 1.39 gcm-1) and an aluminum substrate were used to set up a model. Absorbance of the Al-Kα signal was used to find out the thickness of the organic layer. 2.5 µm is the insignificant layer thickness for every MYLAR layer. Even in the case of organic materials, good agreement was seen between the computed values and the predicted values from MLplus (Table 3).

Table 3. The calculated thicknesses observed with increasing number of aluminum foil layers on copper substrate. The results are in excellent agreement with the expected value of 10.6 µm per aluminum foil layer

No. Layers Calc. Thickness (µm) Expected Thickness (µm) Al-Kα Intensity (kcps)
1 2.88 2.5 1889.19
2 5.23 5.0 1130.56
3 7.63 7.5 670.56
4 10.10 10.0 371.55
5 13.00 12.5 207.97
6 15.20 15.0 129.62

Aluminum drinks cans

The same measuring technique was used to examine the three instances of aluminum drinks cans (Table 4). For an aluminum substrate, a model was built with an organic layer. Using the FTIR technique, the makeup of this model was found to be an epoxy type resin containing small quantities of other polymers. The organic layer was determined to be about 4 µm thick in all cases with the interference bands from the FTIR spectrum. Table 4 shows the results.

Table 4. The calculated thicknesses observed with the examples of aluminum drink cans

Sample Name Can Type Calc. thickness (µm)
Green Non-carbonated, juice based drink 3.86
Green_Cleaned Non-carbonated, juice based drink 0.75
Red Cola type drink 2.93
Red_Cleaned Cola type drink 0.78
Black Carbonated, energy drink 3.69
Black_Cleaned Carbonated, energy drink 0.77

Conclusion

Using the combination of XRF and MLplus, a layer’s thickness on a substrate can be determined in an easy and non-destructive manner. Considering the model samples described above, it is possible to achieve excellent precision if a suitable model is built for the sample types.

The major advantage of the MLplus software is the replication of the multilayer system to assess the optimal evaluation path, all without any measurement. The high precision of the S8 TIGER instrument combined with its improved sensitivity for low energy radiation makes it viable to use L- lines for assessment purposes, enabling the analysis of multiple layers and extremely thin layers.

This information has been sourced, reviewed and adapted from materials provided by Bruker AXS Inc.

For more information on this source, please visit Bruker AXS Inc.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Bruker AXS Inc.. (2019, March 04). Determination of Layer Thickness and Composition by WDXRF with the S8 TIGER with MLplus. AZoM. Retrieved on July 20, 2019 from https://www.azom.com/article.aspx?ArticleID=12953.

  • MLA

    Bruker AXS Inc.. "Determination of Layer Thickness and Composition by WDXRF with the S8 TIGER with MLplus". AZoM. 20 July 2019. <https://www.azom.com/article.aspx?ArticleID=12953>.

  • Chicago

    Bruker AXS Inc.. "Determination of Layer Thickness and Composition by WDXRF with the S8 TIGER with MLplus". AZoM. https://www.azom.com/article.aspx?ArticleID=12953. (accessed July 20, 2019).

  • Harvard

    Bruker AXS Inc.. 2019. Determination of Layer Thickness and Composition by WDXRF with the S8 TIGER with MLplus. AZoM, viewed 20 July 2019, https://www.azom.com/article.aspx?ArticleID=12953.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this article?

Leave your feedback
Submit