The surface-enhanced Raman scattering (SERS) of 4-aminobenzenehtiol (4-ABT) has seen a surge of interest recently, since its SERS spectral features are dependent not only on the kinds of SERS substrates but also on the measurement conditions. The most unusual SERS feature is the appearance of b 2-type bands in the region 1100–1500 cm–1, in contrast to their absence in the normal Raman spectrum, but their origin is not yet clarified. However, propositions have been made suggesting that their appearance is associated with either a charge transfer phenomenon or a surface-induced photoreaction product such as 4,4′-dimercaptoazobenzene (4,4′-DMAB). In this work, we found that the b 2-type bands of 4-ABT are strongly affected also by the solution pH. Regardless of the excitation wavelength and kind of SERS substrates, the b 2-type bands appeared very weakly or negligibly at acidic pHs, while they were observed very distinctly at basic pHs. For the case of 4,4′-DMAB, any such pH dependence was not observed at all in its SERS spectra. Since the pH dependence in the SERS of 4-ABT was observed reversibly, the appearance and disappearance of the b 2-type bands must have nothing to do with formation of any surface-induced photoreaction product like 4,4′-DMAB. Consulting the pH-dependent UV–vis absorption spectra and ab initio quantum mechanical calculation, the disappearance of the b 2-type bands at acidic pHs is presumed to be associated with the upshift of the lowest unoccupied molecular orbital level of 4-ABT caused by protonation of the amine group: the charge transfer resonance chemical enhancement will then be less likely to occur.

No abstract

In order to obtain information on the nature of the "hot" sites in surface-enhanced Raman scattering, we have fabricated a model nanostructure composed of 4-aminobenzenethiol (4-ABT) that has further been sandwiched between Au nanoparticles and a macroscopically smooth Au substrate. In the absence of Au nanoparticles, that is, when 4-ABT is adsorbed solely onto a planar Au substrate, no Raman peak is identified at all using 632.8 nm radiation as the excitation source. When Au nanoparticles are attached onto the amine groups of 4-ABT on a planar Au substrate, Raman peaks are readily observed, however. What is intriguing is that a more intense Raman signal is obtained with increasing size of the Au nanoparticles at least with dimensions of under ∼100 nm. Much the same trend is confirmed from the finite-difference time-domain calculation. This is thought to suggest that the electromagnetic coupling of the localized surface plasmon of Au nanoparticles with the surface plasmon polariton of the underneath smooth Au substrate occurs more favorably with largersized nanoparticles leading to the molecules trapped in the gap exhibiting strong Raman peaks.

Raman scattering measurements were conducted for 4-aminobenzenethiol (4-ABT) monolayers assembled on a macroscopically smooth Pt substrate. At the beginning, no Raman peak was detected for 4-ABT on Pt, but upon attaching Ag nanoparticles to the amine groups of 4-ABT on Pt (Ag@4-ABT/Pt), distinct Raman spectra were observed. Considering the fact that almost no Raman peaks are observed when Ag nanoparticles are attached to 4-aminophenylsilane monolayers assembled on a silicon wafer, the Raman spectra observed for Ag@4-ABT/Pt must be surface-enhanced Raman scattering (SERS) spectra, occurring through an electromagnetic (EM) coupling of the localized surface plasmon of Ag nanoparticles with the surface plasmon polariton of the Pt substrate. From the excitation wavelength dependence, we also confirmed the contribution of the charge-transfer enhancement in the SERS spectra of Ag@4-ABT/Pt: it became more important at short-wavelength excitation. Overall, the SERS intensity of Ag@4-ABT/Pt gradually decreased as the excitation wavelength was increased from 488 to 514.5, 568, and 632.8 nm. A similar trend was observed with a finite-difference time-domain calculation, suggesting that the EM coupling should also be strong at short-wavelength excitation. Accordingly, the experimental enhancement factor per Ag nanoparticle was estimated to be as large as 7.9 × 102 under the illumination of 514.5 nm radiation. The present observation clearly demonstrates that the inherent obstacles to the more widespread use of SERS can be overcome by the judicious use of SERS-active nanoparticles directly or indirectly.