Energy-based calibration for quantitative STEM measurements and comparison with 2D-PAD

0568420 - ÚPT 2023 RIV CZ eng A - Abstract
Zouhar, Martin - Paták, Aleš - Konvalina, Ivo - Fořt, Tomáš - Materna Mikmeková, Eliška
Energy-based calibration for quantitative STEM measurements and comparison with 2D-PAD.
16th Multinational Congress on Microscopy, 16MCM, 04-09 September 2022, Brno, Czech Republic. Book of abstracts. Brno: Czechoslovak Microscopy Society, 2022 - (Krzyžánek, V.; Hrubanová, K.; Hozák, P.; Müllerová, I.; Šlouf, M.). s. 450-451. ISBN 978-80-11-02253-2.
[Multinational Congress on Microscopy /16./. 04.09.2022-09.09.2022, Brno]
R&D Projects: GA TA ČR(CZ) TN01000008
Institutional support: RVO:68081731
Keywords : STEM segmented detector * pixelated detector * scanning electron microscopy * Monte Carlo simulations * ray tracing * quantitative imaging
OECD category: Optics (including laser optics and quantum optics)
https://www.16mcm.cz/wp-content/uploads/2022/09/16MCM-abstract-book.pdf

We present a new calibration method of STEM device equipped with a segmented detector consisting of concentric annuli. The core idea of this method is to perform an additional
measurement of gray-scale dependence g(E) on incident energy E for each segment, so called calibration curves (CCs). The resulting profiles, exhibit suppressed response of the HAADF segment. We compare the measured experimental results with Monte-Carlo simulations (MC-sim) by Geant4 and traced using EOD. We utilise 4 different imaging modes, including ultra-high resolution (UHR) mode with beam deceleration (BD). We scale the data to a reference segment in order to avoid the necessity to measure the incident beam current precisely. The MC-sim provide access to detailed information such as electron count, total energy dose and energies Ei of individual electrons per each segment of the detector. Furthermore, we use the CCs to convert the Ei of simulated detected electrons to partial gray-scale values. These values are summed up and they provide a calibrated theoretical gray-scale G. We discuss how these quantities compare with the observed gray-scale intensity. The data are displayed in Fig. 2. The experimental error-bars stem from intensity variations in the region of interest and they are quantified using standard deviation. The errors in theoretical data originate from statistical analysis of an artificial random split into 10 batches. It turns out that the theoretical gray-scale G (obtained using traced MC-sim and experimental CCs) leads to the best agreement with the directly measured gray-scale. We validate this method on several thin foil samples for high and low values of atomic-number Z. We complement the aforementioned results obtained using a 2-dimensional (2D) pixel-array detector (PAD). They are displayed in Fig. 3 with the corresponding MC-sim. The 2D-PAD provides finer angular resolution when compared to the segmented STEM and hence it represents a technological evolution of the segmented detector. On the other hand, data acquisition is faster in the case of the segmented STEM. The calibration of the 2D-PAD turns out to be much simpler. The intensity angular profiles exhibit sensitivity to both Z and sample thickness in the case of the two detectors. This means that if one knows either the composition or the sample thickness, the MC-sim can be used to estimate the other. Thus the MC-sim provide a valuable tool to extract additional information from the measurements.
Permanent Link: https://hdl.handle.net/11104/0339731