Elemental Quantification of a STEM Sample of Varying Thickness

Scanning transmission electron microscopy (STEM) combined with the the XFlash^® 7T30 detector for Energy-Dispersive Spectroscopy (EDS) is a powerful approach for nanoscale compositional analysis. Even with thin electron-transparent specimens, the accuracy of quantitative analysis can be affected by X-ray absorption and other thickness-dependent effects. This is a critical aspect to consider in samples comprising both heavy and light elements like zinc oxides, where absorption of low-energy X-rays results in a systematic underestimation of light-element content with increasing specimen thickness.

To assess these effects, an electron-transparent lamella with increasing thickness (wedge-shaped) was prepared from a CIGS solar cell using Ga⁺ focused ion beam (FIB) milling (Fig. 1). Its ZnO layer, identified as the region of interest, was analyzed by STEM EDS using 200 kV acceleration voltage and 440 pA electron beam current.

Using Bruker’s ESPRIT software a line profile was extracted from the ESPRIT HyperMap (in which a spectrum is saved for each mapped point) (Fig. 2) and quantified.

Three methods were compared: standardless Cliff–Lorimer, standard-based Cliff-Lorimer and Zeta-factor. The latter explicitly accounts for X-ray absorption effects due to specimen thickness.

The comparison of EDS line profiles across the ZnO layer (Fig. 3) highlights differences between elemental quantification approaches. Both the standardless and standard-based Cliff-Lorimer methods show thickness-dependent deviations, with oxygen appearing deficient in thicker regions.

In contrast, the Zeta-factor method gives a constant elemental ratio corresponding to the expected Zn:O stoichiometry of 1:1, independent of local thickness variations. By correcting for absorption and thickness-dependent effects, this specific elemental quantification method ensures accurate and reliable quantitative EDS results under realistic experimental conditions.

It is particularly useful for the analysis of samples with light and heavy element combinations, or where variations in sample thickness significantly affect the elemental quantification accuracy. By accounting for these variations, the Zeta-factor method provides thickness-independent quantification results and ensures accurate analysis of electron-transparent materials.

Discover ESPRIT Software Contact an expert

Fig. 1. Side‑view diagram of the CIGS solar cell sample with a 25° wedge. The EDS analysis is performed from the top side, where the electron beam penetrates progressively thicker regions of the material as the distance from the edge increases.

Fig. 2. High‑Angle Annular Dark‑Field (HAADF) micrograph of the CIGS solar cell sample with elemental maps of Pt, Zn, O, Mg, S and Cu as overlay. A line profile was extracted from the ESPRIT HyperMap within the ZnO layer. The statistics in the data were improved by integrating the data perpendicular to the line profile direction, as shown in the image.

Fig. 3. Atomic concentration along a line profile acquired over a Zn₀.₅O₀.₅ electron‑transparent sample of increasing thickness, using (a) the standardless Cliff‑Lorimer, (b) the standard‑based Cliff‑Lorimer and (c) the Zeta‑factor quantification methods. The standardless Cliff‑Lorimer method only shows accurate atomic concentrations in the thin specimen part, where absorption of the oxygen signal occurs to a negligible extent. The standard‑based Cliff‑Lorimer method only gives proper content at the specimen thickness of choice used to define the Cliff‑Lorimer factors (here at the 100 nm thickness mark of the wedge‑shaped sample). Only the Zeta‑factor quantification method provides accurate atomic concentrations for the entire line profile by considering thickness‑dependent absorption effects.

Further Information

Data courtesy of Karlsruher Institut für Technologie (KIT) and Zentrum für Sonnenenergie- Und Wasserstoff-Forschung Baden-Württemberg (ZSW).

Learn More:

X. Jin et al., “Improvement of Quantitative STEM/EDXS Analyses for Chemical Analysis of Cu(In,Ga)Se2 Solar Cells with Zn(O,S) Buffer Layers”, Microscopy and Microanalysis, vol. 29, no. 1, pp. 69–77, Feb. 2023.
X. Jin et al., “Improved quantitative chemical analyses of Cu(In,Ga)Se2 solar cells performed by STEM/EDXS”, Microscopy and Microanalysis, vol. 27, no. S1, pp. 2060–2063, Aug. 2021.