EXAFS and XANES Investigation of the ETS-10 Microporous Titanosilicate C. Prestipino,* ,² P. L. Solari, ‡ and C. Lamberti ² Department of Inorganic, Physical and Materials Chemistry and NIS Centre of Excellence, Via P. Giuria 7, I-10125 Torino, Italy, and ESRF, rue J. Horowitz 6, BP 220 F-38043 Grenoble France ReceiVed: January 11, 2005; In Final Form: March 16, 2005 In this work, we report state-of-the-art analysis of both Ti K-edge high-resolution XANES and EXAFS data collected on the ETS-10 molecular sieve at the GILDA BM8 beamline of the ESRF facility. The interatomic distances and the angles obtained in our EXAFS study are in fair agreement with the single-crystal XRD data of Wang and Jacobson (Chem. Commun. 1999, 973) and with the recent ab initio periodic study of Damin et al. (J. Phys. Chem. B 2004, 108, 1328) Differently from previous EXAFS work (J. Phys. Chem. 1996, 100, 449), our study supports a model of ETS-10 where the Ti atoms are bonded with two equivalent axial oxygen atoms. This model is also able to reproduce the edge and the post-edge region of the XANES spectrum. Conversely, the weak but well-defined pre-edge peak at 4971.3 eV can be explained only by assuming that a fraction of Ti atoms are in a local geometry similar to that of the pentacoordinated Ti sites in the ETS-4 structure. These Ti atoms in ETS-10 should be the terminal of the -Ti-O-Ti-O-Ti- chains, of which the actual number is strongly increased by the high crystal defectivity (Ti vacancies). 1. Introduction Microporous titanosilicates have been of interest since the discovery of important catalytic properties in titanium silicalite-1 (TS-1) 1 in oxidation and epoxidation reactions. 2-5 Among the huge number of porous titanosilicates, ETS-10, first reported by Kuznicki et al., 6 occupies an important role owing to its interesting chemical and physical properties. Because of the high number of counterions present in the material, this molecular sieve has been appreciated for its ionic exchange properties 6,7 and as a bare catalyst. 8 The much higher density of framework Ti(IV) centers with respect to other titanosilicates, such as TS-1 2-5 or Ti-, 9 makes ETS-10 a promising photocatalyst. Its advantage, with respect to the classical high-surface-area TiO 2 oxide, consists of a three- dimensional microporous structure that provides to ETS-10 a shape selectivity toward the photodegradation of large organic molecules. 10,11 Very recently, Llabre ´s i Xamena et al. 12 have enhanced the activity in the shape-selectivity photocatalytic degradation of large aromatic molecules of ETS-10 by controlled defects production with fluoridic acid. 1.1. Brief Overview of the Structure of ETS-10. ETS-10 is a microporous titanosilicate with as-synthesized stoichiometry of (Na, K) 2 TiSi 5 O 13 . For the first time, a convincing model of the structure of the ETS-10 framework was proposed by Anderson et al. 13 on the basis of a detailed high-resolution electron microscopy (HREM), NMR, powder x-ray diffraction (XRD), and structural modeling study. The same authors successively reported a deeper discussion on the model else- where. 14,15 However, up to 1999, the small-crystal size of samples of ETS-10 has prevented accurate structure refinement from single-crystal data, despite efforts to improve the syntheses. After that time, Wang and Jacobson 16 have been able to grow crystals with the necessary size to perform an accurate single- crystal analysis. A much more accurate knowledge of the structure of this material has consequently been achieved. Its structure is formed by chains of corner-sharing TiO 6 octahedra linked to each other by tetrahedral SiO 4 units, generating a large-pore (twelve-membered ring) and forming a three-dimensional channel structure. Although the material is crystalline, it is characterized by a high level of disorder, owing to the possibility to build the crystal with different stacking arrangements. In fact, although two “ideal” polymorphs of ETS- 10 have been identified, each formed by a particular stacking arrangement of the (001) layer, in reality, the stacking sequence is virtually random, resulting in an inherently disordered structure. The two polymorphs are shown in Figure 1. Poly- morph A (tetragonal, P4 1 or P4 3 ) is formed by a zigzag stacking of layers, while polymorph B (monoclinic, C2/c) results from a diagonal stacking arrangement. The occurrence of stacking defects has even been revealed by the electron microscopy studies by Anderson et al. 13-15,17 This structural complexity has prevented the use of standard approaches for the refinement of the diffraction data, complicating the structure solving process. One of the most interesting aspects of the ETS-10 structure is its possibility to be thought of as a network of one-dimensional TiO 2 semiconductor “quantum wires” embedded in a siliceous insulating matrix. This idea was first introduced by Torino’s group, 18,19 who by following this concept have been able to explain the main features of the UV-vis spectrum of this material. A serious understanding of the electronic properties of ETS-10 came only later in the ab initio works of Bordiga et al. 20 and of Damin et al. 21 1.2. Previous XAFS Study of ETS-10. Extended x-ray absorption fine structure (EXAFS) spectroscopy is an important tool for the characterization of new materials, particularly when a degree of disorder is present in the structure. The first application of this techniques on ETS-10 is the work of Davis et al., 22 which appeared in 1995, just one year after the structure proposed by Anderson et al. 13 Successively, a more accurate * Corresponding author: C. Prestipino, phone +39011-6707841; fax +39011-6707855; e-mail carmelo.prestipino@unito.it. ² Department of Inorganic, Physical and Materials Chemistry and NIS Centre of Excellence. ‡ ESRF. 13132 J. Phys. Chem. B 2005, 109, 13132-13137 10.1021/jp050183h CCC: $30.25 © 2005 American Chemical Society Published on Web 06/17/2005