Reactivity of Cr Species Grafted on SiO 2 /Si(100) Surface: A Reflection Extended X-ray Absorption Fine Structure Study down to the Submonolayer Regime G. Agostini, ² E. Groppo, ² S. Bordiga, ² A. Zecchina, ² C. Prestipino, ‡ F. D’Acapito, § E. van Kimmenade, | P. C. Thu 1 ne, | J. W. Niemantsverdriet, | and C. Lamberti* ,² Department of Inorganic, Physical and Materials Chemistry, and NIS “Center of Excellence”, UniVersity of Torino, Via P. Giuria 7, I-10125 Torino Italy, ESRF, BP220 F-38043 Grenoble, France, CNR-INFM-OGG c/o ESRF, Gilda CRG, 6 Rue Jules Horowits F-38043 Grenoble, France, and Schuit Institute of Catalysis, EindhoVen UniVersity of Technology, 5600 MB EindhoVen, The Netherlands ReceiVed: May 25, 2007; In Final Form: July 24, 2007 In situ X-ray absorption near-edge spectroscopy/extended X-ray absorption fine structure (XANES/EXAFS) experiments are conducted for the first time on a highly diluted Cr/SiO 2 /Si(100) system (2 Cr/nm 2 , representing a model of the Phillips catalyst for the ethylene polymerization) by exploiting the reflection EXAFS (ReflEXAFS) geometry. This experiment, aimed to give a contribution in bridging the gap between surface science and catalysis, demonstrates that it is possible to follow the reversible red-ox reactivity of surface species grafted on a single well-defined surface, at a concentration limit that is far below the monolayer coverage level and for a highly sensitive sample. A further improvement on the impurity level of the ReflEXAFS chamber is however required in order to be able to follow in situ the polymerization reaction. Our results demonstrate that the red-ox ability of the isolated surface Cr species is not enough to make a polymerization active species. 1. Introduction X-ray absorption spectroscopy, in both the EXAFS and XANES regions, being able to discriminate oxidation states, coordination states, and the local environment of a selected atomic species, is a useful technique to probe the reactivity of surface species. 1-11 Most of the quoted examples refer to transmission or fluorescence experiments performed on samples where the reactive species were grafted (hosted) on a high surface area material (oxides, zeolites, active carbons, etc...). However, the development of beamlines ad hoc conceived to measure grazing angle geometry has allowed the surface science community to investigate the surface of single crystals and thin films. The high photon flux available on third generation synchrotron radiation sources has been able to push the surface detection limit down to the monolayer and submonolayer regime. 12-18 In this regard, only a few examples are present where the same experiment combines the monolayer sensitivity of grazing angle geometry acquisitions with the control of the sample environment (temperature and reaction atmosphere) needed to perform chemical reactions. 19-30 The Cr/SiO 2 Phillips catalyst, 31 although used in industrial plants since the 1960s for C 2 H 4 polymerization, is still one of the most debated systems, concerning both the molecular structure of the active sites and the related initiation mechanism, for which a unifying picture is still missing. 32-35 The main reasons why these two strictly connected questions are not properly addressed are the high intrinsic heterogeneity of the Cr sites formed at the surface of amorphous silica and the high Cr dilution (typically less than 1 wt % Cr). The second point is related to the formation of catalytically inactive Cr 2 O 3 clusters at higher Cr loadings. 11 Attempts to reduce the complexity of the catalyst surface have been continuously made over the last decades. In this respect, the annealing method of McDaniel and co-workers, 32 further developed by Groppo et al., 35 represents a way to fine-tune the relative population of Cr(II) sites. The use of organometallic precursors has been proposed by the group of Scott. 36,37 Recently, a method to completely remove the heterogeneity of the Cr II /SiO 2 system has been reported. 38-40 With the use of the complex TAC (1,3,5-tribenzylhexahydro 1,3,5-triazine) ligand, as a surface-modifying agent, a single site Cr species was made, resulting in the formation of polyethylene with a very low polydispersity index. It has recently been found that the much simpler CH 2 Cl 2 molecule acts as a surface- modifying agent, having the dual function to selectively enhance the catalytic activity of a small fraction of Cr sites, while poisoning the remaining ones. This approach represented a much simpler method to reduce the Cr heterogeneity, increasing at the same time the catalytic performances of the system. 41 An alternative approach to simplify the CrO x /SiO 2 system is to replace the porous silica substrate by a flat silicon wafer with thermal oxide layers, 42-49 hereafter SiO 2 /Si(100). While the preparation chemistry remains unaltered with respect to con- ventional Phillips catalysts, these model systems feature an almost atomically flat silicon oxide surface, which renders the active surface equally accessible for surface characterization and for the reactants during polymerization. This in turn allows one to correlate surface chemistry 47 to intrinsic catalytic reactivity 48 and polymerization kinetics. 49 In this work we want to extend the XANES and EXAFS investigation to the SiO 2 /Si(100) model catalysts. The extremely low number of Cr atoms grafted on the support for a load of 2 Cr atom/nm 2 implies that we are below the threshold of 1000- 500 ppm concentration commonly accepted to obtain reasonable quality EXAFS spectra. As a consequence, the choice of ReflEXAFS geometry has to be considered mandatory. In fact, * Corresponding author. Phone: +39011-6707841. Fax: +39011- 6707855. E-mail: carlo.lamberti@unito.it. ² University of Torino. ‡ ESRF. § CNR-INFM-OGG. | Eindhoven University of Technology. 16437 J. Phys. Chem. C 2007, 111, 16437-16444 10.1021/jp074066t CCC: $37.00 © 2007 American Chemical Society Published on Web 10/06/2007