Tungsten Bronzes Framework as a Glasslike Host for Transition Cations: The Case of Ba 6 Zr 2 Ta 8 O 30 V. Massarotti,* ,² D. Capsoni, ² M. Bini, ² C. B. Azzoni, ‡ M. C. Mozzati, ‡ and P. Galinetto ‡ Dipartimento di Chimica Fisica “M. Rolla” dell’UniVersita ` , UniVersita ` di PaVia, Viale Taramelli 16, and CNISM, Dipartimento di Fisica “A. Volta” dell’UniVersita ` , UniVersita ` di PaVia, Via Bassi 6, 27100 PaVia, Italy ReceiVed: January 29, 2007; In Final Form: March 8, 2007 Structural, microstructural, and spectroscopic properties of Ba 6 Zr 2 Ta 8 O 30 (BZT) and Ba 6 Zr 1.94 M 0.06 Ta 8 O 30 (M ) Ni, Co, Fe, Mn, Cr) related doped compounds are reported. The distribution of Zr, Ta, and M dopant ions on the cationic sites and the M valence states are discussed on the basis of the results obtained by the application of the Rietveld method on X-ray powder diffraction (XRPD) patterns and by electron paramagnetic resonance. The micro-Raman spectra of pure and doped BZT, reported and discussed for the first time, confirm the capability of the lattice to host transition cations in a glasslike matrix. The particle size effects of the samples are investigated by analyzing XRPD line broadening and scanning electron microscopy observations. Comparisons with the isostructural pure and doped Ba 6 Ti 2 Nb 8 O 30 compounds are also reported. Introduction Members of the family of the ferroelectric tungsten bronze 1 were studied mainly for their capability of cation substitution, which allows remarkable variations of their physical properties, 2-6 possibly related to future applications in the fields of electro- optics and nonlinear optics. 1,2 In particular, the framework of Ba 6 Ti 2 Nb 8 O 30 (BTN) compound displays a “spongelike” be- havior due to the possibility of some cations to occupy a multiplicity of sites without modifying the crystalline structure. 7 Really, the BTN structure offers two octahedral sites, occupied by both Ti 4+ and Nb 5+ , and two high-coordination sites suitable for large bivalent cations, like Ca, Sr, and Pb. 8,9 On these sites, a possible substitution with rare earths was also observed. 10-14 In the BTN framework, an interstitial site is usually empty, but it was indicated suitable to host small cations, like Li or Lu. 15 For the Ba 6 Zr 2 Ta 8 O 30 (BZT) compound, isostructural with BTN, 8 substitution capability and cation distribution on the octahedral sites, as well as the pertinent property changes occurring in the substituted materials, are still unknown. The aims of the present paper are to synthesize BZT, the related transition cation-substituted compounds Ba 6 Zr 1.94 M 0.06 - Ta 8 O 30 (M ) Ni, Co, Fe, Mn, Cr) and to investigate the related structural, microstructural, and spectroscopic properties, to be compared and discussed with those of the analogous pure and doped BTN compounds. 7 The first step of the work is to obtain the cation distribution between the two octahedral sites on the basis of refinement of structural parameters (the occupancy factors of the sites) from X-ray powder diffraction (XRPD) data, as well as the valence state of the doping cation by analyzing the electron paramagnetic resonance (EPR) data. Micro-Raman (μ-Raman) spectra of the pure and doped BZT bronzes will also be performed to assess their features and possible changes after doping and to obtain additional evidence concerning the structural properties. Experimental Methods Pure and substituted BZT samples were prepared via solid- state synthesis from a mixture of BaCO 3 (Carlo Erba), ZrO 2 (Merck, >99%), Ta 2 O 5 (Aldrich, 99.99%) and Cr 2 O 3 (Aldrich, 99.995%), MnO (Alfa, 99.5%), Fe 2 O 3 (Alfa, 99.9%), CoO (Aldrich, 99.99+%) or NiO (Aldrich, 99.99%) oxides in the proper amounts to obtain Ba 6 Zr 2 Ta 8 O 30 and Ba 6 Zr 1.94 M 0.06 - Ta 8 O 30 (M ) Ni, Co, Fe, Mn, Cr). Each mixture was treated 30 h at 1273 K, 60 h at 1623 K, and 60 h at 1673 K, with intermediate grinding. This thermal treatment allows us to obtain the impurity free samples, at least at the XRPD detection limit. However, Ni- and Co-doped samples required a further treatment for 30 h at 1673 K to reach the same purity of the other samples. Room temperature (RT) XRPD measurements were per- formed in air on a Bruker D5005 diffractometer with the Cu KR radiation, Ni filter, and a position sensitive detector. Rietveld structural and profile refinement was carried out by means of TOPAS program. 16 The cation distribution was determined by means of a constrained model: the total quantity of Zr, Ta, and dopant ions can freely distribute on the two octahedral cationic sites (1 and 2), maintaining the full occupancy of both sites. The peak broadening was obtained by peak profile fitting with a pseudo-Voigt function, and the crystallite size was determined with the single line method 17 by means of the Win-Crysize 3.0 software. 18 BaF 2 crystalline powder was used after a suitable annealing at 823 K, as the standard for the evaluation of the instrumental broadening contribution. 19 EPR measurements were carried out at about 9.4 GHz at RT with a Bruker spectrometer. Particular care was paid in determining the sample mass and position in the resonant cavity to compare signal intensities (areas) with those of suitable standards (e.g., Li 2 MnO 3 ) and of the previously studied BTN samples. 7 μ-Raman measurements were carried out at RT by using a Labram Dilor H10 spectrometer equipped with an Olympus microscope HS BX40. He-Ne laser beam (632.8 nm) was employed as exciting light with a power less than 10 mW at the sample. The microscope, coupled confocally to the spec- * To whom correspondence should be addressed. E-mail: vincenzo.massarotti@unipv.it. Tel: 39-382-987203. Fax: 39-382-987575. ² Dipartimento di Chimica Fisica “M. Rolla” dell’Universita `. ‡ Dipartimento di Fisica “A. Volta” dell’Universita `. 6857 J. Phys. Chem. C 2007, 111, 6857-6861 10.1021/jp0707586 CCC: $37.00 © 2007 American Chemical Society Published on Web 04/17/2007