Aberration-corrected STEM, combined with synchronization of fast spectrometers and cameras, is able to study a wide range of materials properties with unprecedented spatial localization. Properties of interest can include a material’s optical response (valence EELS), chemical bond hybridization and magnetic moments (core-loss EELS), composition (EDX), local crystal structure and strain (4D STEM diffraction), and more. These properties are determined by analyzing independently-acquired datasets captured from disparate scattering cross sections that are acquired either in parallel or in series from either completely different detectors or similar detectors operating in different recording modes. Consequently, while it is possible to study singular properties with high spatial resolution, at present, there is no widely accepted way to combine these different datasets such that the property interactions become evident.

In this presentation, we outline a data acquisition and analysis workflow called hypermodal data fusion that we have specifically designed to tackle this challenge. By exploiting the spatial coordinates shared by multiple experiments, we show how an arbitrarily large number of datasets can be directly connected (or "fused") and jointly analyzed for variations. While we have previously demonstrated this approach by combining spectroscopy datasets [1–3], here we illustrate its agnosticism by expanding it to also include 4D STEM. We demonstrate the power of this approach by using it to first identify different TiO$_2$ polymorphs (anatase and rutile) in a random assortment of overlapping nanoparticles, but then also extract bandgaps (monochromated EELS), composition (EDX), and even diffraction patterns (4D STEM) that are unique to each polymorph. We conclude with examples of how this workflow can address a huge range of relevant materials ranging from lithium ion batteries to heterogeneous photocatalysts.

[1] T. Thersleff, S. Budnyk, L. Drangai, A. Slabon, Dissecting complex nanoparticle heterostructures via multimodal data fusion with aberration-corrected STEM spectroscopy, Ultramicroscopy. 219 (2020) 113116. https://doi.org/10.1016/j.ultramic.2020.113116.

[2] T. Thersleff, I.Z. Jenei, S. Budnyk, N. Dörr, A. Slabon, Soot Nanoparticles Generated from Tribofilm Decomposition under Real Engine Conditions for Identifying Lubricant Hazards, ACS Appl. Nano Mater. 4 (2021) 220–228. https://doi.org/10.1021/acsanm.0c02536.

[3] P. Merkl, S. Zhou, A. Zaganiaris, M. Shahata, A. Eleftheraki, T. Thersleff, G.A. Sotiriou, Plasmonic Coupling in Silver Nanoparticle Aggregates and Their Polymer Composite Films for Near-Infrared Photothermal Biofilm Eradication, ACS Appl. Nano Mater. (2021). https://doi.org/10.1021/acsanm.1c00668.