The formation of geological samples is the result of the constant depositing of inorganic and organic materials over millions of years.
As layers of deposit accumulate, the increasing weight creates pressure and increased temperature, resulting in the creation of rocks and a range of other formations.
These formations are eroded and carried away by rain, wind and snow, causing them to be deposited in streams and lake beds over time.
The analysis of these materials provides insight into the mineral bodies, as well as any environmental contaminants that may be present.
Proper analysis of soil and sediment samples must accommodate a diverse array of material and matrix types, featuring elements with concentrations ranging from parts per million (ppm) levels to considerable weight percentages of the composition.
Image Credit: Thermo Fisher Scientific – Materials & Structural Analysis
These factors make wavelength dispersive X-ray fluorescence (WDXRF) an ideal choice for the accurate quantitative analysis of soil and sediment samples.
WDXRF is the most commonly employed technique for the analysis of geological base materials due to its accuracy, precision and wide dynamic range of concentrations (ppm – 100%) that it can accommodate.
This method is also much simpler than other analytical techniques, while its automation possibilities make it ideal for the high throughput of samples exhibiting a wide element range.
Thermo Fisher Scientific has developed a range of factory calibrations to assist in the total analysis of geochemical samples. This involves the analysis of two key groups of elements or oxides: majors and minors and the trace elements.
It is important that the analysis of majors and minors be as accurate as possible because these elements are more often than not the largest influential factor in the analysis of other elements in the sample or the material of interest itself.
Multiple sample-related effects should be considered when working with samples where pressed pellets are utilized; for example, mineralogical factors, particle size and inhomogeneities that could influence the accuracy of the analysis.
One viable method for negating these influences lies in the use of fused bead sample preparation - fusion remains the most accurate means of preparation for XRF samples.
This procedure involves heating a mixture of the sample and a borate flux, which melts and dissolves the sample. This flux is typically lithium tetraborate and/or lithium metaborate, melted at a high temperature (1000 ºC – 1100 ºC).
The heating conditions and overall composition should be set to produce a one phase glass after cooling - the ratio sample:flux is typically 1:11.
However, it should be noted that fusion bead preparation is unsuitable for the analysis of trace elements. Its dilution ratio makes the determination of very low concentrations difficult, for example, 1 – 50 ppm.
A pressed pellet is, therefore, the most appropriate sample preparation method for the analysis of trace elements because the physical effects which impact XRF’s accuracy are not as notable for trace element analysis.
Instrument
The analysis presented here employed a Thermo Scientific™ ARL™ PERFORM’X Series Spectrometer. This 4200-watt system is configured with a gearless goniometer equipped with four collimators, up to nine crystals and two detectors.
A total of six primary beam filters help enhance the peak to background ratio for selected elements.
The instrument’s 5GN+ X-ray tube ensures optimum performance when working with ultra-light to the heaviest elements due to its use of a rhodium anode and 50 micron Be window.
This novel X-ray tube features a low current filament, ensuring unmatched long-term analytical stability.
The ARL™ PERFORM’X Spectrometer has been designed for demanding laboratories. Its rapid goniometer and dual sample loading enable high throughput, while the instrument can cover up to 84 elements of the Mendeleev periodic table.
The ARL™ PERFORM’X Spectrometer provides exceptional performance and sample analysis safety, with its distinct LoadSafe design the key to minimizing risks during sample pumping and loading.
The instrument’s Secutainer system protects the primary chamber, vacuum collecting any loose powders into a specially designed container that can be easily removed to facilitate cleaning.
The ARL™ PERFORM’X Spectrometer also offers the features of small spot sizes and elemental mapping analysis, enabling analysis at 1.5 mm or 0.5 mm.
This additional analytical functionality enhances the capabilities of an XRF system by delivering further screening, inclusion analysis, contamination identification and segregation/inhomogeneity mapping.
Image Credit: Thermo Fisher Scientific – Materials & Structural Analysis
Table 1. Concentration ranges of the various oxide types with the standard errors of estimate achieved. Source: Thermo Fisher Scientific – Materials & Structural Analysis
| Elements |
Range % Ignited Samples |
Typical SEE (%) Ignited Samples |
| Na2O |
0.4 – 10.4 |
0.1 |
| MgO |
0.2 – 97.3 |
0.22 |
| Al2O3 |
0.2 – 89.2 |
0.16 |
| SiO2 |
0.3 – 99.7 |
0.23 |
| P2O5 |
0.06 – 40.0 |
0.11 |
| SO3 |
0.05 – 3.7 |
0.05 |
| K2O |
0.03 – 15.4 |
0.03 |
| CaO |
0.03 – 94.4 |
0.32 |
| TiO2 |
0.02 – 3.8 |
0.03 |
| Cr2O3 |
0.02 – 17.4 |
0.03 |
| MnO |
0.02 – 8.0 |
0.01 |
| Fe2O3 |
0.03 – 94 |
0.15 |
Table 2. Typical precision and limits of detection in 40s obtained on vario. Source: Thermo Fisher Scientific – Materials & Structural Analysis
| Elements |
Average Conc (%) |
Precision St. Dev. at the Avg Conc (%) |
LoD ppm |
| Na2O |
0.053 |
0.005 |
120 |
| MgO |
0.014 |
0.0021 |
63 |
| Al2O3 |
0.015 |
0.0019 |
60 |
| SiO2 |
0.003 |
0.0017 |
50 |
| P2O5 |
0.005 |
0.0005 |
15 |
| SO3 |
0.271 |
0.0009 |
27 |
| K2O |
0.002 |
0.0007 |
21 |
| CaO |
0.002 |
0.006 |
18 |
| TiO2 |
0.005 |
0.003 |
10 |
| Cr2O3 |
0.001 |
0.0004 |
12 |
| MnO |
0.0003 |
0.0003 |
10 |
| Fe2O3 |
0.003 |
0.0003 |
10 |
Calibration Ranges and Results
Table 1 shows the range of elements and working concentration ranges employed in the fused bead analysis.
It was possible to establish a working curve for each element using the Multivariable-Regression incorporated in the state-of-the-art Thermo Scientific™ OXSAS software. Theoretical alpha factors have been employed for all matrix corrections.
Trace element analysis was performed to enable the measurement of 30 additional elements for quantification. Table 3 summarizes the complete set of elements, and analytical ranges, as well as standard errors of estimate (SEE), and typical limits of detection (LoD).
These results were achieved in 100 seconds. Table 3 also highlights the precision and analysis times required to achieve these results using the Thermo Scientific™ Geo-Chemical calibration.
XRF instruments must offer sufficient sensitivity, resolution and background/overlap correction functions to accurately analyze trace elements in geological samples. The ARL™ PERFORM’X Spectrometer is the most sophisticated spectrometer on the market, making it ideally suited to this task.
Graphs 1 and 2 provide insight into the analysis strategy for background correction. This was made possible thanks to the instrument’s fully digital goniometer. Its unique optical design is able to lower the background signal due to its polarization effect.
Calibrations were performed based on internationally certified reference materials.
Table 3. LoD–limit of detection and precision. Source: Thermo Fisher Scientific – Materials & Structural Analysis
| Elements |
Line |
Analytical Range
[ppm] |
Number of Samples |
SEE
[ppm] |
LoD
[ppm]
100s |
Precision
[ppm] |
At a Concentration of..
[ppm] |
Analysis
[s] |
| Ag |
Kα |
LoQ-35 |
5 |
5 |
2.5 |
2.1 |
8 |
20 |
| As |
Kβ |
LoQ-600 |
13 |
7 |
2 |
2.5 |
116 |
10 |
| Ba |
Lα |
LoQ-1000 |
12 |
6 |
2 |
4.2 |
252 |
10 |
| Bi |
Lα |
LoQ-50 |
7 |
2 |
0.8 |
0.6 |
10 |
20 |
| Ce |
Lβ |
LoQ-200 |
10 |
10 |
3.8 |
4 |
57 |
30 |
| Cd |
Kα |
LoQ-42 |
9 |
3 |
2.5 |
3.1 |
8 |
20 |
| Cr |
Kα |
LoQ-1300 |
12 |
6 |
0.6 |
1.1 |
35 |
6 |
| Co |
Kα |
LoQ-80 |
13 |
2 |
0.6 |
1.2 |
9 |
20 |
| Cu |
Kα |
LoQ-3000 |
13 |
12 |
0.5 |
0.6 |
80 |
6 |
| Se |
Kα |
LoQ-600 |
2 |
- |
- |
0.3 |
2 |
6 |
| Ti |
Kα |
LoQ-8000 |
13 |
98 |
0.6 |
8.6 |
1457 |
6 |
| Pb |
Lβ |
LoQ-5200 |
13 |
46 |
1.2 |
2 |
308 |
10 |
| Zn |
Kα |
LoQ-1000 |
12 |
25 |
0.5 |
1.5 |
362 |
6 |
| Sn |
Kα |
LoQ-400 |
7 |
1 |
3.5 |
3.9 |
40 |
20 |
| Sb |
Kα |
LoQ-40 |
9 |
3 |
3.5 |
3 |
30 |
20 |
| W |
Lα |
LoQ-120 |
7 |
2 |
1.2 |
2.2 |
36 |
6 |
| Mo |
Kα |
LoQ-90 |
4 |
2 |
0.4 |
0.9 |
8 |
6 |
| Mn |
Kα |
LoQ-10000 |
11 |
16 |
0.7 |
4.7 |
1383 |
6 |
| Nb |
Kα |
LoQ-100 |
7 |
1 |
0.3 |
0.7 |
15 |
6 |
| Ni |
Kα |
LoQ-180 |
12 |
3 |
0.7 |
0.7 |
16 |
6 |
| V |
Kα |
LoQ-170 |
13 |
9 |
0.5 |
1.9 |
42 |
6 |
| Y |
Kα |
LoQ-40 |
6 |
3 |
0.5 |
0.3 |
32 |
6 |
| S |
Kα |
LoQ-4000 |
10 |
250 |
1.2 |
3.6 |
96 |
6 |
| Sr |
Kα |
LoQ-340 |
10 |
6 |
0.5 |
1 |
23 |
6 |
| Te |
Kα |
LoQ-500 |
2 |
- |
- |
- |
- |
20 |
| Zr |
Kβ |
LoQ-230 |
8 |
6 |
0.4 |
2.4 |
188 |
20 |
| Th |
Lα |
LoQ-70 |
9 |
1 |
0.6 |
1.1 |
20 |
20 |
| U |
Lα |
LoQ-17 |
5 |
1 |
1 |
0.8 |
8 |
10 |
| Rb |
Kα |
LoQ-500 |
14 |
6 |
0.3 |
1.3 |
265 |
6 |
| Hg |
Lα |
LoQ-35 |
5 |
3 |
1.4 |
2.1 |
30 |
100 |
Conclusion
The ARL™ PERFORM’X Sequential XRF Spectrometer is ideally suited for the high-performance analysis of geological samples.
The fused bead method ensures the highest accuracy for the detection of major and minor elements, while trace element analysis is ideally performed using a simple pressed pellet method which enables comprehensive elemental analysis of an array of material types and matrices.
It should be noted that the addition of certified reference standards can be used to extend all calibration ranges and elements analyzed.
The instrument’s operation is straightforward and user-friendly thanks to the state-of-the-art OXSAS software, compatible with the latest Microsoft Windows® 10 package.
3
Graph 1. 2-Theta scans from 25 degrees to 38 degree using LiF220 crystal and scintillation detector. Image Credit: Thermo Fisher Scientific – Materials & Structural Analysis
Graph 2. Zoomed region of graph 1. Image Credit: Thermo Fisher Scientific – Materials & Structural Analysis
Graph 3. 2-Theta scan from 55 degrees to 90 degree using LiF220 crystal and scintillation detector showing the excellent spectral resolution. Image Credit: Thermo Fisher Scientific – Materials & Structural Analysis
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.