Distribution pattern of length, length uniformity, and density of TiO3(2-) quantum wires in an ETS-10 crystal revealed by laser-scanning confocal polarized micro-Raman spectroscopy

ETS-10 is a highly intriguing microporous titanosilicate that has shown an excellent propensity for the selective removal of harmful heavy-metal ions, the potential to work as an effective catalyst for various reactions, and that can be used as a material for solar cells. Such important features arise from the TiO3 2 quantum wires with the diameter (d) of approximately 0.67 nm running along the [110] and [110] directions in the crystal (Figure 1). The TiO3 2 quantum wire is a one-dimensional (1D) extreme of three-dimensional (3D) bulk titanates, which are widely used in industry as, for example, capacitors. It also exhibits an interesting 1D quantum confinement effect. The TiO3 2 quantum wires are not expected to be connected all the way from one face to the opposite face of a crystal owing to the large number of randomly distributed defects. Now the questions are what is the average length of the wires, to what degree do the lengths vary (how does the length homogeneity vary), how does the local density of the quantum wire vary from one region to another within a crystal, do they vary randomly or in accordance with a certain pattern? Answers to the above questions will be highly useful for understanding the mechanism of ETS-10 formation and growth, the refinement of its structure, improvements of its catalytic activities, and its future applications. However, there have been no methods to gain such information. The TiO3 2 quantum wire in ETS-10 gives a strong Raman shift band between 724 and 840 cm , arising from a longitudinal vibrational mode of the -Ti-O-Ti-Ochain. Its frequency at the band maximum (nmax), its bandwidth (full width at half maximum, fwhm), and intensity (I) reflect the relative average length, length homogeneity, and density of the quantum wire, respectively. The Raman band frequency decreases as the length increases, owing to the increase in the reduced mass of the quantum wire. The smallest frequency ever observed is 724 cm . Bandwidths between 23 and 120 cm 1 have been observed, and the bandwidth decreases as the length uniformity increases. The intensity increases as the number of the TiO3 2 quantum wire increases. Accordingly, the frequency, bandwidth, and intensity have served as the three important criteria for comparison of the relative average lengths, relative average length uniformities, and relative average densities of the TiO3 2 quantum wires in the ETS-10 crystals. This information indicates that we can also apply the same principle to obtain their distribution pattern within an ETS-10 crystal if we can obtain a matrix of Raman spectra measured from a large number of artificially divided very small sections of a crystal. Furthermore, the obtained data would be more informative if we can obtain a map of these three data sets for the TiO3 2 quantum wires running along the [110] and [110] directions, respectively. We now report that laser scanning confocal polarized micro-Raman (LSC-PMR) spectroscopy is a highly useful tool for the above purpose and the novel fact that the TiO3 2 quantum wires are not evenly distributed within ETS-10 crystals but distributed in a symmetrical manner according to an interesting pattern. Figure 1. a) Illustrations of a typical morphology (truncated bipyramid) of an ETS-10 crystal and three-dimensional networks of SiO2 channels (cyan) and TiO3 2 quantum wires (red) in the case of polymorph B and b) a single TiO3 2 quantum wire.