XRF (X-Ray Fluorescence) Analysis - Total Oxide X-Ray Analysis

There are several techniques for elemental analysis. However, none are as flexible in the analytical range and material type as X-ray fluorescence. Wavelength dispersive X-ray fluorescence (WDXRF) enables measurement of up to 83 elements of the periodic table in samples of various forms and nature: solids or liquids, conductive or nonconductive.

The benefits of XRF over other techniques include:

• Speed of analysis.
• Easy sample preparation.
• High-stability and precision.
• Wide dynamic range (from ppm levels to 100 %).

The two largest factors affecting the precision of the analysis in WDXRF are accuracy of standards and sample preparation. Standard accuracies are typically derived from multiple analyses performed on the material by several different laboratories. The most common form of this type of analysis is done using round robin testing. A statistical computation of all results is performed to achieve a certified reference value for the measured elements. These materials are commercially available through several government agencies and private companies.

As for the second issue, with respect to powder and non-homogenous solids, fusion is the most accurate method of sample preparation for XRF. Fusing oxide materials is an optimal way to completely remove both grain size and mineralogical effects. Essentially, the technique comprises heating a mixture of the sample and a borate flux, namely lithium tetraborate and/or lithium metaborate at high temperature (1000 °C- 1200 °C) so that the flux melts and dissolves the sample. The overall composition and cooling conditions must be such that the end product is a one-phase glass after cooling. The Thermo Scientific Total Oxide X-ray Analyzer (TOXA) is a complete analytical package, which enables analysis by WDXRF of a broad range of minerals, using the General Oxide calibration based on a sample preparation by fusion.

Instrument

For demonstrative purposes, an elemental analysis was performed with a 4200 W system from the Thermo Scientific ARL PERFORM’X series spectrometer (Fig. 1). The features of the system are:

• The system has 6 primary beam filters, 4 collimators, up to nine crystals, two detectors, helium purge and 5GN+ Rh X-ray tube for superior performance from ultra-light to heavy elements as it is equipped with a 50 µm Be window.
• This new X-ray tube fitted with a low current filament ensures an unequalled analytical stability month after month.
• The ARL PERFORM'X provides excellent performance and sample analysis safety.
• Its unique LoadSafe design includes a series of features that prevent any trouble during sample pumping and loading.
• The liquid cassette recognition guarantees that the liquid sample is not exposed to vacuum even by mistake. In case the X-ray exposure time is too long, the liquid sample is automatically ejected.
• Loose powders are collected in a specially designed container by the Secutainer system in order to protect the primary chamber. The container can be easily removed and cleaned.
• For spectral chamber protection, the ARL PERFORM'X uses a helium shutter designed for absolute protection of the goniometer during liquid analysis under helium operation.
• In the "LoadSafe Ultra" configuration, a special X-ray tube shield offers total protection against sample breakage or liquid cell rupture. The ARL PERFORM’X analyzer also features small spot and elemental mapping analysis allowing for 1.5 mm and 0.5 mm areas. These features improve XRF system capabilities by offering additional screening, contamination identification, inclusion analysis and segregation/ non-homogeneity mapping.

Figure 1. Thermo Scientific ARL PERFORM’X series spectrometer

Calibration Ranges and Results

The types of oxides that can be addressed and their concentration ranges are shown in Table 1.

Table 1. Concentration ranges of the various oxide types with the Standard Errors of Estimate achieved

A working curve is drawn for each element using the multivariable-regression incorporated in the sophisticated Thermo Scientific OXSAS software package. Theoretical alpha factors are used for all matrix corrections. Loss on ignition values, which spread up to 47 %, can be used for correction purposes in the multi-variable regression. The Standard Error of Estimate (SEE) is a measure of the accuracy of analysis. It is the average error between the certified concentrations of the standard samples and the calibration curve for a given oxide.

Table 2. Typical limits of detection in 40 s obtained on various oxides (fusions with 1:12 dilution)

The limits of detection (LOD) determined with precision tests at low concentrations are listed in Table 2 for the range of oxides. The analysis time per element can range from 4 to 40 s depending on the element and the precision required. Finally, a 12-day stability test was performed. One sample was run sixty times over twelve days without any drift correction during this period. The results show that the ARL PERFORM’X spectrometer has high-stability ensuring precise analytical results. These results are summarised in Table 3.

Table 3. 12-day stability test without any drift correction

Conclusion

An analysis of fused beads can be performed with ease on the ARL PERFORM’X sequential XRF spectrometer. High-precision and accuracy is obtained in both short term and long term analyses. All calibration ranges can be extended with the simple addition of more certified reference standards. Furthermore, operation is made easy through the state-of-the-art Thermo Scientific OXSAS software which operates with the latest Microsoft Windows 7 packages.

This information has been sourced, reviewed and adapted from materials provided by Thermo Fisher Scientific - Elemental Analyzers.

For more information on this source, please visit Thermo Fisher Scientific - Elemental Analyzers.