Iron-exchanged zeolites are redox-active materials able to activate N₂O, a potent greenhouse gas and environmental pollutant, making them highly relevant in environmental catalytic processes, such as N₂O decomposition [1] and NO-N₂O-SCR [2], and in direct chemical oxidations, such as methane to methanol [3]. The sites capable of N₂O activation in CHA and FER topologies have been identified as isolated iron ions stabilized by six-membered rings of the zeolite framework [1], [2]. However, controlling the iron speciation and distribution by synthesis has proven challenging, as isolated iron ions often coexist with oligomers and particle-like species.

In this work, we studied the influence of the preparation method, post-treatment, and the nature of the iron precursor on the iron speciation and distribution in ZSM-5. Standard characterization and catalytic testing were combined with in situ XAS and EPR spectroscopy to monitor the iron species during pulsed NO-assisted N₂O decomposition experiments.

Iron-exchanged ZSM-5 samples were prepared by incipient wetness impregnation (IWI) with iron acetylacetonate (Fe(acac)₃), and by hydrothermal synthesis (HTS) using Fe(NO₃)₃ and Fe(acac)₃. All calcined materials were also exposed to high‑temperature treatment in Ar flow to promote the redispersion of clustered species into isolated iron ions [3]. This was confirmed by ex-situ DRUV spectroscopy, through a decrease of the oligomer-specific bands. Catalytic testing indicated higher activity of Ar-treated catalysts, with N₂O conversion of ca. 50% at 400 °C, compared to only ca. 23% N₂O conversion for the non-Ar-treated counterparts. This distinct behavior pointed to an increase in active site population for the Ar-treated materials.

Fe K-edge quick-XANES experiments indicated that only a minor fraction of Fe species can accomplish a complete redox cycle Fe³⁺ ↔ Fe²⁺. The reversible changes in the pre-edge and rising edge were revealed exclusively through pulsed experiments. EPR spectroscopy experiments further emphasized the differences in the Fe speciation across synthesis procedures and post-treatment. In particular, the IWI-synthesized and Ar‑treated material exhibited a strong signal at g’ = 2, which was previously assigned to contributions from isolated Fe in the α-exchange cationic position and clustered Fe species [4]. Our pulsed EPR experiments (N₂O in NO/Ar at 400 °C) showed the reversible change of the g’ = 2 signal (Fe³⁺ ↔ Fe²), alongside the g’ = 6 signal characteristic of isolated Fe sites in six-membered rings (β-exchange cationic position), suggesting that two distinct isolated Fe species interact reversibly with the N₂O/NO system [4]. The absence of the g’ = 2 signal in the non-Ar-treated counterpart pointed to the role of the post-treatment in fixing the Fe ions in precise cationic positions in ZSM-5. The nature of the iron precursor affected the speciation similarly. By coupling catalytic testing with pulsed in situ spectroscopic experiments, we show that post-treatment and iron precursor type influence Fe speciation in Fe-ZSM-5. These findings contribute to the improved design of catalysts for NO-assisted N₂O-decomposition.

[1] D.C. Cano-Blanco, J.W.A. Fischer, F. Buttignol, I. Alxneit, G. Jeschke, O. Kröcher, D. Ferri, ACS Catalysis, 2025, 15, 17.[2] F. Buttignol, A. Garbujo, P. Biasi, O. Kröcher, D. Ferri, Journal of the American Chemical Society, 2025, 147, 10.[3] M.L. Bols, J. Devos, H.M. Rhoda, D. Plessers, E.I. Solomon, R.A. Schoonheydt, B.F. Sels, M. Dusselier, Journal of the American Chemical Society, 2025, 143, 1624.[4] E. Berrier, O. Ovsitser, E.V. Kondratenko, M. Schwidder, W. Grünert, A. Brückner, Journal of Catalysis, 2007, 249, 67.