Insights into surface chemistry down to nanoscale: An accessible colour hyperspectral imaging approach for scanning electron microscopy

0575337 - ÚPT 2024 RIV NL eng J - Článek v odborném periodiku
Nohl, J. F. - Farr, N. T. H. - Sun, Y. - Hughes, G. M. - Stehling, N. - Zhang, J. - Longman, F. - Ives, G. - Pokorná, Zuzana - Mika, Filip - Kumar, V. - Mihaylova, L. - Holland, C. - Cussen, S. A. - Rodenburg, C.
Insights into surface chemistry down to nanoscale: An accessible colour hyperspectral imaging approach for scanning electron microscopy.
Materials Today Advances. Roč. 19, August (2023), č. článku 100413. ISSN 2590-0498. E-ISSN 2590-0498
Grant ostatní: AV ČR(CZ) StrategieAV21/26
Program: StrategieAV
Institucionální podpora: RVO:68081731
Klíčová slova: chemical imaging * secondary electron hyperspectral imaging * scanning electron microscopy * secondary electrons * machine learning
Obor OECD: Coating and films
Impakt faktor: 10, rok: 2022
Způsob publikování: Open access
https://www.sciencedirect.com/science/article/pii/S2590049823000735

Chemical imaging (CI) is the spatial identification of molecular chemical composition and is critical to characterising the (in-) homogeneity of functional material surfaces. Nanoscale CI on bulk functional material surfaces is a longstanding challenge in materials science and is addressed here. We demonstrate the feasibility of surface sensitive CI in the scanning electron microscope (SEM) using colour enriched secondary electron hyperspectral imaging (CSEHI). CSEHI is a new concept in the SEM, where secondary electron emissions in up to three energy ranges are assigned to RGB (red, green, blue) image colour channels. The energy ranges are applied to a hyperspectral image volume which is collected in as little as 50 s. The energy ranges can be defined manually or automatically. Manual application requires additional information from the user as first explained and demonstrated for a lithium metal anode (LMA) material, followed by manual CSEHI for a range of materials from art history to zoology. We introduce automated CSEHI, eliminating the need for additional user information, by finding energy ranges using a non-negative matrix factorization (NNMF) based method. Automated CSEHI is evaluated threefold: (1) benchmarking to manual CSEHI on LMA, (2) tracking controlled changes to LMA surfaces, (3) comparing automated CSEHI and manual CI results published in the past to reveal nanostructures in peacock feather and spider silk. Based on the evaluation, CSEHI is well placed to differentiate/track several lithium compounds formed through a solution reaction mechanism on a LMA surface (eg. lithium carbonate, lithium hydroxide and lithium nitride). CSEHI was used to provide insights into the surface chemical distribution on the surface of samples from art history (mineral phases) to zoology (di-sulphide bridge localisation) that are hidden from existing surface analysis techniques. Hence, the CSEHI approach has the potential to impact the way materials are analysed for scientific and industrial purposes.
Trvalý link: https://hdl.handle.net/11104/0345124