Addressing the Quantitative Challenges of X-ray Line Interferences in EDS and WDS

Microanalysts face challenges while using Energy-Dispersive X-ray Spectroscopy (EDS) or Wavelength-Dispersive X-ray Spectroscopy (WDS) to perform quantitative analysis of materials containing elements that have interfering X-ray lines.

Rigorous interference corrections are usually a part of the electron-probe microanalysis (EPMA) software for WDS. However, such corrections are often inaccessible to the SEM user. This article discusses the quantitative analysis of a Ti-V-Al-Fe sample comprising two phases with slight variations in iron and vanadium content to demonstrate some of these challenges.

EDS and WDS Quantitative Analysis

After mounted and uncoated in thick section, the Ti-V-Al-Fe metal sample was analyzed in an FESEM. A Thermo Scientific™ UltraDry™ EDS detector (Figure 1) and the Thermo Scientific™ NORAN™ System 7 microanalysis system (NS7) (Figure 2) were used to acquire the EDS spectra and spectral images. The Thermo Scientific™ MagnaRay™ WDS spectrometer (Figure 3) was used for WDS spectral collection and measurements.

UltraDry EDS Detector

Figure 1. UltraDry EDS Detector

NORAN System 7

Figure 2. NORAN System 7

MagnaRay WDS

Figure 3. MagnaRay WDS

The NORAN System 7 software was used to process both EDS and WDS data. EDS and WDS quantitative analysis was carried out at an accelerating voltage of 15kV and the interaction volume was mitigated by performing the EDS spectral imaging at an accelerating voltage of 10kV.

It was not possible to perform elemental mapping due to the slight variations in phase composition. The Thermo Scientific™ COMPASS™ spectral phase mapping software, was used to determine the phases.

Experimental Results

Unique phases were identified based on the principle component analysis of the EDS spectrum at each pixel as depicted in Figures 4 and 5, showing the presence of tiny, normally ~5µm, V-rich (~13 wt% V) grains along the boundaries of bigger, normally ~10µm, V-poor (~3 wt% V) grains. In addition, roughly 1.6wt% Fe was present in the V-rich grains. Table 1 summarizes the EDS (standardless and standards-based) and WDS quantitative results.

Backscattered electron image

Figure 4. Backscattered electron image

COMPASS phase map: V-poor phase is purple; V-rich phase is yellow

Figure 5. COMPASS phase map: V-poor phase is purple; V-rich phase is yellow

Table 1. Quantitative analyses of a Ti-Al-V alloy

Standardless Standards-based Standards-based
Filter Filter VMeas. VDiff.
V-rich Grains Al 3.46 3.63 3.29 3.29
Ti 81.2 83.4 78.3 78.3
V 13.8 13.2 15.8 16.9
Fe 1.61 1.55 1.54 1.54
Total 100.0 101.8 98.9 100.0
V-poor Grains Al 6.42 6.77 5.23 5.22
Ti 90.7 93.4 90.0 90.0
V 2.87 2.75 6.49 4.49
Fe 0.07 0.06 0.26 0.26
Total 100.0 103.0 102.0 100.0

The separation of the Ti Kβ line (4.931 keV) from V Kα line (4.948 keV) is by mere 17eV. EDS cannot differentiate these X-ray lines, whereas WDS can differentiate them but with a poor resolution (Figure 6). V is counted on the Kα line in the WDS quantitative analyses. The over-estimate of the V concentration is the impact of this interference in the WDS quantitative analyses of these phases.

WDS energy scans over V Kα of V-poor and -rich grains overlaid on EDS spectrum of same spectral region

Figure 6. WDS energy scans over V Kα of V-poor and -rich grains overlaid on EDS spectrum of same spectral region

The degree of this impact is higher in the analyses of the V-poor grains owing to the fact V Kα X-rays are represented only by a few counts. Without rigorous interference corrections, it is not possible to get accurate results from WDS analyses of both grain types.

This drawback could be addressed with three methods. First, the V Kβ line is an appealing peak on which to count due to the absence of interfering energy line in this sample. Nevertheless, very long acquisition times (>10x) are needed to count the V Kβ line in the V-rich grains. The very low V concentrations (~3 wt%) in the V-poor grains makes it impossible to differentiate the V Kβ line from the background.

Second method is a difference method, wherein the wt% of the other elements is subtracted from 100% and the remainder represents the V concentration. This approach needs that only one line is complicated but with accurate measurements of the remaining elements.

Performing EDS quantitative analysis with standards is the third method, wherein the assumption is the accuracy of WDS is better than that of EDS. Nevertheless, in this case, well-developed peak deconvolution methodologies with both standards based and standardless EDS quantitative analysis help EDS provide more accurate results than WDS.


WDS is an essential tool to corroborate the presence or absence of interfering elements. Nevertheless, the ability to completely resolve interfering X-ray lines is essential for the WDS spectrometer, without which it is not possible to use the instrument for accurate quantitative analysis with no interference corrections.

Moreover, the concept of phase mapping with WDS element maps is confounded by these interfering energy lines. In cases where WDS is incapable, accurate results can be obtained from modern EDS quantitative analysis, thanks to the peak deconvolution methodologies.

In addition, COMPASS software discerns between phases with only slight variations in compositions, thus yielding a comprehensive phase map with accurate quantitative analysis of each phase.

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.

Ask A Question

Do you have a question you'd like to ask regarding this article?

Leave your feedback