1326
CO Adsorption on Platinum Nanoparticles - the Importance of Size Distribution Studied with in-Situ DRIFTS and DFT Calculations

Tuesday, October 13, 2015: 08:00
211-A (Phoenix Convention Center)
C. Brieger, J. Melke, P. Kaghazchi (Freie Universitaet Berlin), and C. Roth (Freie Universitaet Berlin)
Catalyst degradation due to carbon support corrosion is a key issue in fuel cell technology. One concept towards enhanced durability is to replace carbon by more stable support materials such as oxides.1 However, this can alter the catalytic acitvity dramatically due to particle-support interaction. The particle-support interaction can be tested with the adsorption of small molecules such as CO. Differences in IR spectra of CO adsorption on platinum nanoparticles can reveal information on the influence of the support as well as on the binding sites.

As a model system, we study the temperature dependence of CO adsorption on Pt/SiO2 with a combination of in-situ DRIFTS (Diffuse Reflectance Infrared Fourier Transform Spectroscopy) and DFT calculations to clarify the binding mode and to distinguish between interparticle and intraparticle heterogeneity. The concept of intraparticle heterogeneity implies that CO can be absorbed at different sites (facets, edges and corners) on a nanoparticle of a specific size2-4 whereas interparticle heterogeneity focuses on the adsorption on different nanoparticle sizes.5

CO adsorption was studied in the gas phase between 25°C - 175°C. The CO concentration as well as the pretreatment conditions were varied. Results for the reduced sample under 1% CO are shown in Fig. 1. For the DRIFTS data the corresponding temperature dependent background as well as the spectra without CO were carefully subtracted. Fig. 1 a-b) displays the measurements and their fits under constant CO flow, data in Fig 1 c-d) is taken after purging with nitrogen to remove  excess CO and to test the thermal stability of the adsorbates. In all experiments, almost no bridge bonded CO is detected. The data can be fitted with three bands for linearly bonded CO at 25°C: band 1 at 2080 cm-1, band 2 at 2070 cm-1, and band 3 at 2050 cm-1. The DFT calculations show that the experimentally-observed differences in CO band positions are due to the interparticle heterogeneity, but not due to the intraparticle heterogeneity. We find that band 1 corresponds to CO adsorption on 1.8 nm Pt nanoparticles with a binding energy of 2.00 eV, while band 2 and 3 correspond to CO adsorption on ~1.5 nm (2.1 eV) and ~1.1 nm (2.3 eV), respectively. Under constant CO flow (Fig. 1 a-b)) the intensities of the fitted bands are highest for the most abundant nanoparticles of 1.8 nm.6 Desorption of CO molecules from larger nanoparticles (weaker binding energies) already starts at low temperatures (Fig. 1 c-d)). For temperatures higher than 100°C, CO molecules are only absorbed on the smaller nanoparticles (stronger binding energies).

Our results show that nanoparticle size distribution (interparticle heterogeneity) has to be considered for adsorption studies on catalysts. The temperature dependence of CO adsorption on different cluster sizes could be used to identify reactive sites and thus are important for electrochemical activity and mechanistic studies. In future, these results will be coupled with in-situ XAS experiments and an electrochemical environment.

References:

1. E. Antolini, and E. R. Gonzalez, Solid State Ionics, 180, 746-63, (2009).

2. A. Garnier, S. Sall, F. Garin, M. J. Chetcuti, and C. Petit, Journal of Molecular Catalysis A-Chemical, 373, 127-34, (2013).

3. S. K. Cheah, V. P. Bernardet, A. A. Franco, O. Lemaire, and P. Gelin, The Journal of Physical  Chemistry C, 117, 22756-67, (2013).

4. B. Hammer, O. H. Nielsen, and J. K. Norskov, Catalysis Letters, 46, 31-5, (1997).

5. F. Maillard, E. R. Savinova, P. A. Simonov, V. I.  Zaikovskii, and U. Stimming, Journal of Physical Chemistry B, 108, 17893-904, (2004).

6. J. W. Geus, and P.B. Wells, Applied Catalysis, 18, 231-42, (1985).