What remains to be done to allow quantitative X-ray microanalysis performed with EDS to become a true characterization technique?

Electron Probe Microanalysis of Transition Metals using L lines: The Effect of Self-absorption

Electron microprobe-based quantitative compositional measurement of first-row transition metals using their L$\alpha$ X-ray lines is hampered by, among other effects, self-absorption. This effect, which occurs when a broad X-ray line is located close to a broad absorption edge, is not accounted for by matrix corrections. To assess the error due to neglecting self-absorption, we calculate the L$\alpha$ X-ray intensity emitted from metallic Fe, Ni, Cu, and Zn targets, assuming a Lorentzian profile for the X-ray line and taking into account the energy dependence of the mass absorption coefficient near the absorption edge. We find that calculated X-ray intensities depart increasingly, for increasing electron beam energy, from those obtained assuming a narrow X-ray line and a single fixed absorption coefficient (conventional approach), with a maximum deviation of $\sim$15% for Ni and of $\sim$10% for Fe. In contrast, X-ray intensities calculated for metallic Zn and Cu do not differ significantly from those obtained using the conventional approach. The implications of these results for the analysis of transition-metal compounds by electron probe microanalysis as well as strategies to account for self-absorption effects are discussed.

The f-ratio model for quantitative X-ray microanalysis

The f-ratio method is a new quantitative X-ray microanalysis method developed based on a cold field emission scanning electron microscope/energy dispersive spectroscopy system. The f-ratio is calculated with the characteristic X-ray intensities, and the Monte Carlo simulation is employed to build the theoretical relation between the system composition and the f-ratio. In this study, the f-ratio model is formulated with the elemental concentrations and the f-ratio coefficients. The f-ratio models in the binary S-Fe system and the ternary O-Al-Si system were studied, and the beam energy effects were investigated. The quantitative analyses were performed on the standard pyrite (FeS₂) and kyanite (Al₂SiO₅) specimens, and the results show that the f-ratio model is able to achieve a satisfying accuracy.

The Impact of Chemical Bonding on Mass Absorption Coefficients of Soft X-rays

Accurate elemental quantification of materials by X-ray detection techniques in electron microscopes or microprobes can only be carried out if the appropriate mass absorption coefficients (MACs) are known. With continuous advancements in experimental techniques, databases of MACs must be expanded in order to account for new detection limits. Soft X-ray emission spectroscopy (SXES) is a characterization technique that can detect emitted X-rays whose energies are in the range of 10 eV to 2 keV by using a varied-line-spaced grating. Transitions producing soft X-rays can be detected and accurate MACs are required for use in quantification. This work uses Monte Carlo modeling coupled with multivoltage SXES measurements in an electron probe micro-analyzer (EPMA) to compute MACs for the L2,3-M and Li Kα transitions in a variety of aluminum alloys. Electron depth distribution curves obtained by the software MC X-ray are used in a parametrized fitting equation. The MACs are calculated using a least-squares regression analysis. It is shown that X-ray distribution cross-sections at such low energies need to take into account additional contributions, such as Coster–Kronig transitions, Auger yields, and wave function effects in order to be accurate.

Improvement of the energy resolution of energy dispersive spectrometers (EDS) using Richardson-Lucy deconvolution

A serious limitation in the use of energy dispersive spectrometers (EDS) for materials characterization arises from the fact that these x-ray detectors provide a poor energy resolution compared to other x-ray techniques. This is mainly due to the broadening function generated by the electronics of the detector. However, new windowless detectors are now capable of resolving low energy peaks like Li Kα with modified electronics and show a system peak full width at half maximum (FWHM) of around 30 eV. In this paper, we investigated how the Richardson-Lucy algorithm can be used to remove, or at least attenuate, the contribution of the broadening function to experimental spectra. This method proved to be efficient in improving the energy resolution at any energy in the EDS spectrum. The resulting system peak FWHM was reduced to as low as 7-8 eV and the separation of low energy x-ray peaks were demonstrated in the low energy range of the spectra. The method was also efficient in reducing the peak broadening at higher energies (Cu Kα) and the broadening function of the detector could be experimentally determined to provide higher accuracy in predicting peak broadening. Although some critical artefacts were noted on the restored spectra, like energy shifts and intensity variations, the method explored in this work is worth to be considered for further quantification tests.

The qualitative f-ratio method applied to electron channelling-induced x-ray imaging with an annular silicon drift detector in a scanning electron microscope in the transmission mode

Electron channelling is known to affect the x-ray production when an accelerated electron beam is applied to a crystalline material and is highly dependent on the local crystal orientation. This effect, unless very long counting time are used, is barely noticeable on x-ray energy spectra recorded with conventional silicon drift detectors (SDD) located at a small elevation angle. However, the very high count rates provided by the new commercially available annular SDDs permit now to observe this effect routinely and may, in some circumstances, hide the true elemental x-ray variations due to the local true specimen composition. To circumvent this issue, the recently developed f-ratio method was applied to display qualitatively the true net intensity x-ray variations in a thin specimen of a Ti-6Al-4V alloy in a scanning electron microscope in transmission mode. The diffraction contrast observed in the x-ray images was successfully cancelled through the use of f-ratios and the true composition variations at the grain boundaries could be observed in relation to the dislocation alignment prior to the β-phase nucleation. The qualitative effectiveness in removing channelling effects demonstrated in this work makes the f-ratio, in its quantitative form, a possible alternative to the ZAF method in channelling conditions.

Electron Probe Microanalysis of Ni Silicides Using Ni-L X-Ray Lines

We report electron probe microanalysis measurements on nickel silicides, Ni5Si2, Ni2Si, Ni3Si2, and NiSi, which were done in order to investigate anomalies that affect the analysis of such materials by using the Ni L3-M4,5 line (Lα). Possible sources of systematic discrepancies between experimental data and theoretical predictions of Ni L3-M4,5 k-ratios are examined, and special attention is paid to dependence of the Ni L3-M4,5 k-ratios on mass-attenuation coefficients and partial fluorescence yields. Self-absorption X-ray spectra and empirical mass-attenuation coefficients were obtained for the considered materials from X-ray emission spectra and relative X-ray intensity measurements, respectively. It is shown that calculated k-ratios with empirical mass attenuation coefficients and modified partial fluorescence yields give better agreement with experimental data, except at very low accelerating voltages. Alternatively, satisfactory agreement is also achieved by using the Ni L3-M1 line (Lℓ) instead of the Ni L3-M4,5 line.

A method to test the performance of an energy-dispersive X-ray spectrometer (EDS)

A test material for routine performance evaluation of energy-dispersive X-ray spectrometers (EDS) is presented. It consists of a synthetic, thick coating of C, Al, Mn, Cu, and Zr, in an elemental composition that provides interference-free characteristic X-ray lines of similar intensities at 10 kV scanning electron microscope voltage. The EDS energy resolution at the C-K, Mn-Lα, Cu-Lα, Al-K, Zr-Lα, and Mn-Kα lines, the calibration state of the energy scale, and the Mn-Lα/Mn-Kα intensity ratio as a measure for the low-energy detection efficiency are calculated by a dedicated software package from the 10 kV spectrum. Measurements at various input count rates and processor shaping times enable an estimation of the operation conditions for which the X-ray spectrum is not yet corrupted by pile-up events. Representative examples of EDS systems characterized with the test material and the related software are presented and discussed.

The secondary X-ray fluorescence and absorption near the interface of multi-material: case of EDS microanalysis

A simple model is proposed to take into account secondary X-ray fluorescence and absorption effects near the interface. This model is based on the investigation of the shape change of the first derivative equation that can fit the sigmoidal EDS profile obtained when a high vacuum electron beam passes through the interface of two adjacent materials. The contribution of the photoelectric absorption of primary X-rays (characteristic and Bremsstrahlung) and the secondary fluorescence on the degradation of the X-ray spatial resolution can be easily quantified. The close agreement between the simulated (Monte Carlo simulation using the Casino software) and the experimental data serves to assess the reliability of this developed model.