Most of the Raman work (except NSOM) performed near metal surfaces has been concerned with SERS. We do not comment on SERS models here, but note which peaks not normally seen in Raman are expressed. We discuss elsewhere [4] the SERS work on benzene, several cyanides, and C 60 . Here we concentrate on pyrazine, which is a benzene ring with two opposing carbons replaced with N. It has been studied extensively since it is complicated enough to have several vibration modes (with D 2h symmetry), but simple enough that the bulk vibrations are either Raman or IR active, but not both. It has been studied on Ni, Cu, Au, and Ag. [5-7] When shifts of the usual Raman lines between bulk or solution and surface Raman are used as a gauge of substrate interaction, the pyrazine reacts most with Ni, then Au and Cu, and least with Ag. [5] We therefore concentrate on the Ag results. The strong IR line at1484 cm -1 is the strongest new line in the various papers, but with intensity comparable to the normally allowed modes in some studies [6, 7], and much reduced from them in others. [5] Another strong IR line at 1418 cm -1 is often nearly as strong as the 1484 cm -1 line, [5, 7] but absent in other studies [6] and for studies involving the strongly interacting metals. [5] The strong IR – strong new Raman lines analogy for GFR holds, but the irreproducibility is troublesome. Perhaps the orientation of the molecules, which matters in GFR, depends strongly on sample preparation. The intensity of these lines is sensitive to electrochemical potential, [7] and the orientation of pyrazine on Ni does change with concentration, lying flat at low concentrations and on-end at higher concentrations. [5] This could explain why these in-plane (so not GFR affected for flat orientation) vibrations are not completely reproducible. In NSOM, a sharpened optical fiber is coated with aluminum to form an aperture. The probe is positioned near the surface under lateral force feedback. The NSOM is used in illumination mode, with 514 nm Ar ion laser light coupled into the fiber probe. Reflected light is collimated with a 0.50 NA lens, passed through a holographic filter, focused into a Czerny-Turner spectrometer, and finally collected onto a cooled (-45 C) CCD camera. A near field spectra of KTP is shown in Fig. 1. The general shape is typical of the spectra at various distances from the sample. This region of the Raman spectrum contains vibrations primarily from TiO 6 stretching modes in the KTP. The near-field peak shown in green in figure 1 is the strong totally symmetric, A1, vibration mode, which has been observed before in both near- and far-field measurements. [3] The changes observed as the probe approaches the surface are rather small, so we resort to comparative spectra. Several spectra taken far from the surface were averaged and subtracted from single spectra acquired closer to the surface. Two peaks at different energies than the original peaks are observed: the B1 peak (not allowed in the
Read full abstract