Descriptors for noble metal redispersion on CeO2 based catalysts
Résumé
Introduction and Motivations Predicting the state of catalysts during operation is key for the implementation of reactivation strategies and for the synthesis of long-time stable materials. This is particularly important for catalysts consisting of noble metal particles supported on reducible metal oxides. Such systems based on limited and costly resources are applied in various processes, from fine chemical synthesis to fuel cells and emission control. Recent research has focused on the strong interaction between platinum group metals and CeO2, which offers unique structural and catalytic properties [1]. While enhancing the low temperature oxidation activity, the strong interaction leads to Pt redispersion under oxidizing conditions at elevated temperatures, which is often accompanied by catalyst deactivation [2]. Conversely, reducing treatments promote the formation of nanoparticles, significantly enhancing the catalytic activity. However, these reactivation strategies are energy-intensive and should be minimized over the catalyst’s lifecycle. Therefore, designing redispersion-resistant catalysts at the atomic level while maintaining the active noble metal-CeO2 interface is crucial [3]. This study investigates the kinetics of Pt redispersion on CeO2 nanocubes, aiming to quantify it as a function of (1) the local environment, (2) the initial Pt nanoparticle size, and (3) correlating these findings to the integral catalytic activity. Materials and Methods With the ultimate aim of following the redispersion with electron microscopy, well-defined CeO2 nanocubes with a low surface area (23 m²/g) were chosen as support material as the cubic morphology allows to obtain a parallel alignment of the sample to the electron beam. The nanocubes were loaded with 1.0 wt.% Pt via incipient wetness impregnation, maintaining the loading well below an atomic monolayer. XAS revealed that sample treatment in H2 leads to Pt reduction and formation of nanoparticles with a narrow size distribution of 1-2 nm in diameter. This pre-treatment protocol was replicated in an ETEM using 19 mbar of H2 at 500 °C. For the redispersion experiment, oxidizing conditions (500 °C, 1.5 mbar O2) were applied and the time-dependent decay of individual Pt nanoparticles was observed for few seconds in irregular time intervals over 30 minutes. To minimize possible effects of the electron beam, the beam was blanked in between the measurements. CO oxidation measurements were performed in a lean (10% O2, 1000 ppm CO) reaction mixture using a plug-flow reactor. The catalyst activity was monitored during three consecutive light-offs (heating of the catalyst in reaction mixture) before and after a reductive treatment. Results and Discussion By following the size and morphology of five Pt nanoparticles during redispersion using ETEM, we were able to correlate the redispersion rate to the local environment of the individual nanoparticles. Hereby it could be shown that redispersion of Pt in the proximity of other nanoparticles is hindered compared to noble metal entities located on the plane CeO2 surface. Morphological changes during the redispersion process indicate high dynamics, which are also reflected in the interplanar Pt-Pt spacing. Overall, a slight extension of the spacing was observed due to a partial oxidation of the nanoparticles under the conditions used in the microscope. Based on the redispersion rate, the number of available Pt surface sites was calculated over time assuming a hemispherical shape of the nanoparticle. In closest approximation, the number of exposed surface sites influences directly the reaction rate. By this, a link to the catalytic oxidation activity could be established. The catalytic activity under oxygen rich conditions was monitored using CO as reactant, since the reaction is well-known to be catalyzed only by Pt agglomerates while Pt single perform poorly. As expected, the catalytic activity diminished during the three consecutive light-offs, in line with the noble metal redispersion. Strikingly, the time constant of the loss in activity was in the same order of magnitude as the lifetime of the nanoparticles in the lean mixture. Based on this knowledge, a full prediction of the structural evolution and corresponding catalytic properties of Pt/CeO2 nanomaterials seems to be possible. This outcome will also help with the rational design of redispersion resistant materials. References 1.F. Maurer, J. Jelic, J. Wang, A. M. Gänzler, P. Dolcet, C. Wöll, Y. Wang, F. Studt, M. Casapu, J.-D. Grunwaldt, Nat. Catal. 2020, 3 (10), 824–833. 2.A. M. Gänzler, M. Casapu, F. Maurer, H. Störmer, D. Gerthsen, G. Ferré, P. Vernoux, B. Bornmann, R. Frahm, V. Murzin, M. Nachtegaal, M. Votsmeier, J.-D. Grunwaldt, ACS Catal. 2018, 8 (6), 4800–4811. 3.F. Maurer, A. Beck, J. Jelic, W. Wang, S. Mangold, M. Stehle, D. Wang, P. Dolcet, A. M. Gänzler, C. Kübel, F. Studt, M. Casapu, J.-D. Grunwaldt, ACS Catal. 2022, 12 (4), 2473-2486. Significance and Relevance We tracked the formation of Pt agglomerates and quantified the redispersion of individual nano particles using Environmental Electron Transmission Microscopy (ETEM), X-ray Absorption Spectros- copy (XAS), and Density Functional Theoretical (DFT) calculations. By correlating these structural rates to catalytic ones, we were able to identify the redispersion as main descriptor for catalyst deactivation under the given conditions. The observed processes can now be used to develop redispersion-resistant catalysts enhancing the catalytic efficiency of costly and scarce noble metals throughout the lifetime. Acknowledgements The authors thank the Agence Nationale de la Recherche (ANR) and the Deutsche Forschungsgemeinschaft (DFG) for financial support via the DYCAT project (grant No. ANR-19-CE05- 0038 and grant No. 431423888, respectively). FM furthermore acknowledges “Fonds der Chemischen Industrie” (FCI) as well as DFG for funding via the SFB 1441 TrackAct − Project-ID 426888090.