Shedding Light on the Electronic Structure and Local Configuration of Fe/N/C Active Sites – an in Situ X-Ray Absorption Spectroscopy Study
Kathrin Ebner,Viktoriia A. Saveleva,Adam H. Clark,Grigory Smolentsev,Jingfeng Chen,Lingmei Ni,Jingkun Li,Andrea Zitolo,Frederic Jaouen,Ulrike Ingrid Kramm,Thomas J. Schmidt,Juan Herranz
DOI: https://doi.org/10.1149/ma2020-02623156mtgabs
2020-01-01
ECS Meeting Abstracts
Abstract:In recent years, the search for platinum group metal-free oxygen reduction reaction (ORR) catalysts as an inexpensive and sustainable replacement for state-of-the-art Pt-based materials has inspired a lot of work in the field of single-atom Fe-based catalysts (often referred to as Fe/N/C). While promising initial ORR-activities were achieved with different synthesis approaches,[1,2] tackling the lack of durability[3] will require a strategic optimization of the employed preparation routes. A comprehensive fundamental understanding of the catalysts active sites’ electronic structure and local configuration under operating conditions lies at the heart of this process and remains a subject of vivid debate. Several studies employing both ex and in situ Mössbauer and X-ray absorption spectroscopies (XAS)[4–7] have brought insights regarding these questions, but a full unambiguous conclusion has not been reached and the community is seeking for novel approaches to unravel such structure-activity and structure-stability relations. With this motivation, we have used in situ XAS to study two catalysts prepared with distinctively different synthesis approaches and widely free of detectable inorganic side phases,[5,8] and that therefore constitute ideal model systems to draw conclusions applicable to this catalyst family. We focused our attention on the spectral 1s → 3d pre-edge feature of the X-ray absorption near edge structure (XANES) spectrum that is sensitive to any changes in the local site symmetry, orbital occupancy, and spin state of the absorbing atom,[9,10] but that to the best of our knowledge has so far not been looked into in the field. In order to achieve adequate energy resolution to discern the relevant features, the spectra were acquired with a Si(311) monochromator at the SuperXAS beamline of the Swiss Light Source employing our group’s spectroelectrochemical flow cell.[11] As illustrated in Fig. 1a, pseudo-Voigt component fits allow tracking the spectral changes as a function of the applied polarization conditions, as qualitatively indicated with green arrows. When using bulk-sensitive techniques like XAS, one must bear in mind that while the catalytic process and expected site changes of interest take place only on the surface sites in contact with the electrolyte, the observed signal also contains the spectral contributions of those sites that remain unaffected by the potential. This is particularly relevant when studying such single-atom Fe-based catalysts, since it is expected that a relatively large (and hard to quantify) fraction of their sites are buried in the catalysts’ bulk. Herein we overcome this intrinsic challenge by performing modulation excitation (ME) experiments using the applied potential as a stimulus (illustrated in Fig. 1b). The technique allows to greatly enhance the signal of the species that changes with said stimulus,[12] and consequently to effectively surmount the lack of surface sensitivity described above. Furthermore, these ME experiments acquired with fluorescence-detected quick-scanning XAS[13] complemented with a multivariate curve resolution (MCR) analysis of the data provide quantitative information about the kinetics of the structural changes undergone by the sites in these Fe/N/C catalysts upon modifying the potential (Fig. 1c). In summary, this work will provide a novel perspective by exploiting the sensitivity of pre-edge XANES features and ME-experiments for the first time for the analysis of Fe/N/C-catalysts allowing insights that will broaden the greatly-needed understanding of the potential-dependent structure of the active sites. [1] E. Proietti et al., Nat. Commun. 2011, 2:416. [2] J. Shui et al., Proc. Natl. Acad. Sci. 2015, 112, 10629–10634. [3] D. Banham et al., J. Power Sources 2015, 285, 334–348. [4] U. I. Kramm et al., Phys. Chem. Chem. Phys. 2012, 14, 11673–88. [5] A. Zitolo et al., Nat. Mater. 20 15, 14, 937–42. [6] Q. Jia et al., Nano Energy 2016, 29, 65–82. [7] S. Wagner et al., Angew. Chemie Int. Ed. 2019, 58, 10486–10492. [8] S. Wagner et al., Hyperfine Interact. 2018, 239:10. [9] T. E. Westre et al., J. Am. Chem. Soc. 1997, 119, 6297–6314. [10] M. Wilke et al., Am. Mineral. 2001, 86, 714–730. [11] T. Binninger et al., J. Electrochem. Soc. 2016, 163, H913–H920. [12] A. Urakawa et al., Chem. Eng. Sci. 2008, 63, 4902–4909. [13] A. H. Clark et al., J. Synchrotron Radiat. 2020, 27, 1–8. Figure 1