Background
Windowless energy dispersive spectroscopy (EDS) detectors are renowned for their high sensitivity and ability to detect low energy x-rays, making them ideal for analyzing light elements such as boron, nitrogen, and oxygen. However, many models that are commercially available cannot be used as general-purpose EDS detectors, as they suffer from an inability to perform quantitative analysis at high accelerating voltages, and therefore cannot accurately determine the concentrations of heavy elements.
Figure 1. The EDAX Octane Elite Ultra EDS detector.
Standardless quantitative analysis of heavy elements requires the scanning electron microscope (SEM) to be operated at high accelerating voltages (typically >20 kV). Therefore, an electron trap that prevents (high energy) backscattered electrons from reaching the x-ray sensor is required (thereby preventing pollution of the x-ray signal by electrons) and an analytical model that can correct for generation and absorption of x-rays from a large sample volume.
Materials and methods
Many windowless EDS detectors forego quantitative analysis of elements at accelerating voltages higher than 10 kV due to the complexity in designing an electron trap suitable for a large area detector that must also offer a short distance-to-sample to maximize the solid angle of detection. However, leveraging Gatan’s electron optics expertise which has been established over decades of development of electron energy loss spectrometers (EELS), a novel electron trap was developed for the EDAX Octane Elite Ultra EDS system that permits operation of the 160 mm2 windowless detector at all SEM accelerating voltages, while maintaining an outstanding solid angle for detecting x-rays. The performance of the new electron trap was benchmarked against other EDS detectors using a bismuth specimen (atomic number 81) and an accelerating voltage of 30 kV to maximize the backscattered electron signal. Compared to existing conventional (windowed) detectors, the EDAX Octane Elite Ultra delivered EDS spectra with a lower background than all other detectors evaluated at x-ray energies >15 keV (Table 1). The electron trap of the EDAX Octane Elite Ultra outperformed the standard models despite offering a solid angle of x-ray detection more than 6x larger.
| Model |
Window |
Active area |
Ratio Bi Lα to background* |
| Standard EDS detector |
Windowed |
Medium |
238 |
| Other |
Windowless |
Large |
N/A |
| EDAX Octane Elite Ultra |
Windowless |
Largest |
400 |
Table 1. Comparison of the electron trap performance. EDS spectra were normalized to the Bi Lα to account for differences in solid angle and detector geometry. *The background was measured at an x-ray energy of 25 keV. EDS spectra could not be captured at an accelerating voltage of 30 kV for other windowless EDS detectors.
Performing accurate standardless quantitative analysis of a sample that contains light and heavy elements at high accelerating voltages can be particularly challenging. Standardless quantitative analysis uses matrix correction factors to account for changes in x-ray excitation, absorption, and secondary fluorescence processes as a function of accelerating voltage and sample composition (e.g., EDAX’s eZAF correction). For specimens that contain samples of large difference in atomic number, Z, the matrix corrections can be extremely large, particularly the absorption correction (Mass Absorption Coefficient (MAC)). In general, MACs increase as the energy of the absorbed x-ray decreases, so corrections for low Z elements are large while those for high Z elements are smaller. Moreover, high-Z elements tend to be strong absorbers, so large corrections are required for low Z elements in a matrix containing high-Z elements.
When performing quantitative analysis under these circumstances, a small error in the measured net counts, or matrix corrections and analytical model results in a large deviation in the quantitative results. Therefore, it is important that the analytical model can 1) separate the Bremsstrahlung background from characteristic x-rays accurately, 2) separate overlapping peaks effectively, 3) correct for changes in the detector efficiency with x-ray energy and 4) have accurate matrix corrections for the sample and conditions used.
Using the EDAX Octane Elite Ultra and EDAX APEX™ EDS Advanced 3.0 software, quantitative analysis was performed on five samples of certified composition at high accelerating voltage. The samples contained two elements that differed in atomic number by >50. Standardless normalized analysis using eZAF matrix corrections and a calculated Bremsstrahlung background fit was used to measure the elemental composition (Table 2). In all cases, including tungsten silicide, which suffers from a significant overlap of the tungsten Mα and silicon Kα peaks, the measured compositional analysis was acceptable and an average absolute deviation between measured and certified composition of only 2.1% was observed.
| Standard compound |
Element |
Certified composition (wt.) |
Measured composition (wt.) |
Absolute error per element (wt.) |
| 1 |
2 |
1 |
2 |
1 |
2 |
| Lead fluoride |
Pb |
F |
84.95% |
15.05% |
86.4% |
13.6% |
1.4% |
| Tungsten silicide |
W |
Si |
77.9% |
22.1% |
74.4% |
25.6% |
3.5% |
| Europium (III) oxide |
Eu |
O |
81.1% |
18.9% |
84.5% |
15.5% |
3.4% |
| Erbium fluoride |
Er |
F |
74.1% |
25.9% |
73.6% |
26.4% |
0.4% |
| Ytterbium fluoride |
Yb |
F |
75.2% |
24.8% |
76.8% |
23.2% |
1.6% |
Table 2. Standardless normalized analysis of elemental standards measured at 25 kV for a live time of 20 s and analyzed by normalized standardless analysis using eZAF matrix and carbon coat corrections and background fitting using the modeled Bremsstrahlung radiation.
Summary
The EDAX Octane Elite Ultra has been used to determine the composition of heavy metal compounds with very acceptable accuracy at high accelerating voltages. Leveraging the power of a newly designed electron trap, the EDAX Octane Elite Ultra has been shown to be suitable for quantitative analysis of heavy elements making it the first large area windowless EDS detector suitable for routine microanalysis in addition to having high sensitivity and the ability to detect low-energy x-rays that large area windowless EDS detectors are renowned for.