Surface-Enhanced Raman Scattering of 4-Aminobenzenethiol and 4-Nitrobenzenethiol at Silver and Gold Nanogaps and Nanostructures

Abstract

In Chapter 1, The General Introduction, we provide background information on surface-enhanced Raman scattering (SERS), along with the characteristics of finite-difference time-domain (FDTD) method adopted in this thesis. In Chapter 2, SERS spectra of 4-nitrobenzenethiol (4-NBT) and 4-aminobenzenethiol (4-ABT) on Ag obtained under ambient conditions and in icy environments at 77 K are presented. This study was conducted to resolve the debate on the origin of b2-type bands appearing in the SERS of 4-NBT and 4-ABT. The origin of b2-type bands in the SERS of 4-NBT and 4-ABT has recently been debated because these bands are very similar to those attributed to a photoreaction product such as 4,4′-dimercaptoazobenzene (4,4′-DMAB). We confirmed in this work that under ambient conditions, the b2-type bands distinctly appeared in the SERS spectra of both 4-NBT and 4-ABT. In contrast, no b2-type peaks appeared in the SERS of 4-NBT in icy environments, suggesting that 4-NBT did not undergo a photoreaction. However, the SERS spectral pattern of 4-ABT was the same both at room temperature and in icy conditions. Based on our separate observation that hot electrons are plasmonically generated from Ag even in icy environments, the lack of photoreaction of 4-NBT is likely a result of the small spaces between the ice crystals, rendering the N–O bond difficult to break. The situation of 4-ABT on Ag is identical to that of 4-NBT on Ag in the same conditions

therefore, the b2-type bands observed in icy conditions must be because of the 4-ABT, and not because of the production of 4,4′-DMAB or other photoreaction products. Regardless of temperature, hot electrons were more easily generated at lower excitation wavelengths, and the b2-type bands appeared more distinctly with a decrease in the excitation wavelength. From these observations, we conclude that the hot electrons, as well as the b2-type bands of 4-ABT, are associated with the charge-transfer chemical enhancement mechanism in SERS. In Chapter 3, SERS of 4-ABT at the nanogaps between metal nanoparticles and a flat Au substrate is described. This study was conducted to understand the characteristics of one kind of hot site for SERS. In fact, although no Raman signal is observable when 4-ABT, for instance, is self-assembled on a flat Au substrate, a distinct spectrum is obtained when Ag or Au nanoparticles are adsorbed on the pendent amine groups of 4-ABT. This is definitely due to the electromagnetic coupling between the localized surface plasmon of Ag or Au nanoparticle with the surface plasmon polariton of the planar Au substrate, allowing an intense electric field to be induced in the gap even by visible light. On this basis, firstly, we have thoroughly examined the size effect of Ag nanoparticles, along with the excitation wavelength dependence, by assembling 4-ABT between planar Au and a variable-size Ag nanoparticle (from 20- to 80-nm in diameter). Regarding the size dependence, a higher Raman signal was observed when larger Ag nanoparticles were attached onto 4-ABT, irrespective of the excitation wavelength. Regarding the excitation wavelength, the highest Raman signal was measured at 568 nm excitation, slightly larger than that at 632.8 nm excitation. The Raman signal measured at 514.5 and 488 nm excitation was an order of magnitude weaker than that at 568 nm excitation, in agreement with the three-dimensional finite-difference time domain (3-D FDTD) simulation. It is noteworthy that placing an Au nanoparticle on 4-ABT, instead of an Ag nanoparticle, the enhancement at the 568 nm excitation was several tens of times weaker than that at the 632.8 nm excitation, suggesting the importance of the localized surface plasmon resonance of the Ag nanoparticles for an effective coupling with the surface plasmon polariton of the planar Au substrate to induce an very intense electric field at the nanogap. In addition, secondly, the Raman spectral characteristics of 1,4-phenylenediisocyanide (1,4-PDI) and 4-ABT positioned at the nanogap formed by Au/Ag alloy nanoparticles and a flat Au substrate were examined, and 3-D FDTD calculations were carried out. More intense Raman signal was measured, regardless of the excitation wavelength, when Ag-rich Au/Ag alloy nanoparticles were used to form the nanogaps. Regarding the excitation wavelength, 568 nm light was the most effective in inducing a Raman signal, particularly when Ag nanoparticles were adsorbed on 1,4-PDI or 4-ABT, whereas 632.8 nm light was slightly more effective than 568 nm light when Au nanoparticles were adsorbed onto them. The Raman spectra of 1,4-PDI could be attributed to the electromagnetic enhancement mechanism. The dependencies of the Raman spectra of 1,4-PDI on the excitation wavelength and the type of Au/Ag alloy nanoparticle were comparable to those predicted by the 3-D FDTD calculations. From the measured NC stretching frequencies, the surface of 35-nm sized Au/Ag alloy nanoparticles containing more than 5 mole percent of Ag atoms was concluded to be covered fully with Ag atoms. The Raman spectra of 4-ABT were interpreted to be a product of electromagnetic and chemical enhancement mechanisms. Assuming that the Raman intensity ratios of the b2- and a1-type bands were indicative of the extent of chemical enhancement, the Ag-to-4-ABT electron transfer appeared more facile than the Au-to-4-ABT transfer did and more favorable by excitation with a 514.5 nm laser than 568 nm or 632.8 nm laser. In Chapter 4, SERS of 4-NBT at the nanogaps between a planar Au substrate and a Ag nanostructure is described. This study was conducted to appreciate the effectiveness of hot electrons plasmonically generated from Ag nanoparticles. As described in Chapter 2, 4-NBT adsorbed on a nanostructured Ag substrate can be reduced to 4-ABT by the irradiation of a visible laser. In order to evaluate the effectiveness of hot electrons generated from Ag, we have carried out a SERS study by forming a nanogap system composed of a planar Au substrate and an Ag-coated micrometer-sized silica bead, wherein 4-NBT was adsorbed firstly onto the Au substrate and then Ag-coated silica beads, derivatized with 1-alkanethiols, were spread over the 4-NBT layer: the distance between 4-NBT and a nanostructured Ag substrate was varied by the chain length of alkanethiol molecules. Although the planar Au substrate itself was ineffective in the reduction of 4-NBT, hot electrons usable in the reduction of 4-NBT were generated from the Ag-coated silica beads. The hot electrons generated by 514.5-nm radiation were more effective in the reduction of 4-NBT to 4-ABT than those generated by 632.8-nm radiation, although the nanogap was more SERS-active with the excitation at 632.8-nm than at 514.5-nm. The photoreduction efficiency of hot electrons nonetheless decreased linearly with the distance they travelled from the Ag surface: the reduction capability at a distance of 2 nm apart is about one fourth of that in contact situations.