Transmission electron microscopy (TEM) has proven itself an invaluable tool for measuring composition, chemistry, and internal structure at the nanoscale and below. TEM micrographs, however, are inherently two-dimensional projections of the real three-dimensional object; important structural information can be lost and ambiguities introduced. Nanoparticles (NP) of different structures – which can possess dramatically different catalytic activity and effectiveness – can yield similar apparent projected sizes in micrographs . The ability to distinguish between them – and, ideally, recover the 3D atomic structure – is necessary for continued improvement in catalyst design.
Quantitative STEM offers an approach for extracting this information contained in a single high-angle annular dark-field scanning TEM (HAADF-STEM) micrograph. First demonstrated nearly two decades ago by Singhal et al. , the recent advancements in aberration-correction for STEM spurred a resurgence in interest in quantitative STEM when it was shown the technique could be performed with atomic resolution by LeBeau et al. . Through careful calibration of the microscope and image, the contrast (scattered intensity) in the micrographs can be explicitly related back to the number of atoms involved in the scattering through comparison with image simulation, yielding information about the 3D structure  and composition  of the specimens.
The current quantitative STEM work has been performed using microscopes equipped with Schottky field-emission gun (FEG) electron sources. The highest-performance analytical STEMs, however, are equipped with cold FEG (CFEG) sources, since CFEGs offer superior spatial and temporal coherences, yielding higher spatial and energy resolutions for imaging and spectroscopy. This performance gain comes at the cost of reduced stability of the emission current, which, in a CFEG, decays in a continuous, non-linear fashion even image-to-image. This poses a challenge for quantitative STEM, as one of the key calibrations requires knowledge of the incident beam current to normalize image intensities into units of fractional beam current for comparison with image simulations [2,5]. In this presentation we will discuss our method to overcome this issue by adapting the condenser aperture of a Hitachi HD-2700C STEM to act as a beam monitor to measure the incident probe current in real-time concurrent with STEM image acquisition. This method enables a more accurate calibration of intensity to be achieved on microscopes with CFEGs.
We are also adapting quantitative STEM to enable its use on a conventional, non-aberration-corrected TEM/STEM – a JEOL JEM2100F – without the need for any special modifications or attachments, as most university facilities do not possess such aberration-corrected instruments with corresponding expensive service contracts. The lower magnifications (2-4 MX) used in this approach means that a much larger number of nanoparticles can be present in each micrograph, such as in Figure 1a, enabling robust statistics about particle size, shape, and monodispersity to be gathered from hundreds to thousands of nanoparticles. We have developed a program for MATLAB to automatically perform the quantitative STEM analysis on batches of micrographs. For each nanoparticle, the optimal intensity integration cut-off radius is calculated via an iterative process that determines when the scattered intensity from the NP is fully enclosed, Figure 1b . It should be noted that this approach can be paired with analysis of selected NP at atomic-resolution studies, to gain the benefit of both robust statistics and atom-level characterization.
 A Singhal, et al., Ultramicroscopy 67 (1997), p. 191-206
 JM LeBeau, et al., Physical Review Letters 100 (2008), p. 20601
 L Jones, et al., Nano Letters 14 (2014), p. 6336-6341
 A Rosenauer, et al., Ultramicroscopy 109 (2009), p. 1171-1182
 H E, et al., Ultramicroscopy 133 (2013), p. 109-119
 This work was supported by DOE BES through grant DE FG02-03ER15476, and performed using the facilities at the Center for Functional Nanomaterials at Brookhaven National Laboratory, which is supported by DOE BES through contract DE-SC0012704.
To cite this abstract:Judith Yang, Stephen House, Yuxiang Chen, Dong Su, Tom Schamp, Russell Henry, Eric Stach, Rongchao Jin; Quantitative STEM Atom Counting in Supported Metal Nanoparticles. The 16th European Microscopy Congress, Lyon, France. https://emc-proceedings.com/abstract/quantitative-stem-atom-counting-in-supported-metal-nanoparticles/. Accessed: January 21, 2022
EMC Abstracts - https://emc-proceedings.com/abstract/quantitative-stem-atom-counting-in-supported-metal-nanoparticles/