Abstract
Infrared absorption enhancement is reported for a polyacrylic acid film deposited on evaporated gold film in normal transmission geometry. Comparison between absorption intensities measured with and without the self-assembled monolayer of p-nitrothiophenol demonstrates that the absorption intensities of polyacrylic acid are independent of the presence of a self-assembled monolayer. Results suggest that the first layer effect in the physical enhancement mechanism does not exist in this thermally evaporated island film.
Export citation and abstract BibTeX RIS
1. Introduction
Hartstein et al [1] and later Hatta et al [2] reported that the Kretschmann ATR configuration enhances infrared absorption of molecules adsorbed onto evaporated silver or gold films [3]. For surface-enhanced infrared absorption, the extended range of that enhanced electric field has remained a problem of interest. In the case of a chemical effect typified by charge transfer (CT) between a metal and a molecule, the range is restricted to the first layer. Moreover, a chemical effect exists in combination of a specific molecule and a substrate. In addition to charge transfer, a so-called first-layer-effect (FLE), such as involvement of electron-hole pairs [4] and perturbation by molecules adsorbed by the dielectric constant of metal particles [5], has also been proposed. For this FLE, a molecule with absorption enhancement is limited to the molecules adsorbed directly to the metal. In other words, the effects can be regarded as a kind of chemical effect.
Discriminating between short-range fields (FLE) and longer-range fields is apparently a matter of prime importance for elucidating the enhancement mechanism. The centre of our interest is the extent from the surface to which the enhanced field expands in the physical effect.
An earlier report [6] described experiments using a polycyanoacrylate film on Ag evaporated by vacuum heating on the bottom of the Ge prism: the extent of the resultant electric field was ca. 3–5 nm. Moreover, it was theoretically predicted using the Square Columnar Model [7] that the enhanced field is concentrated on the channel part between the metal particles. This concentration phenomenon was verified experimentally using a Ag evaporated film [8] and a Au array prepared using electron beam lithography [9]. Nevertheless, the distribution of the electric field within the channel has not been clarified.
One report of Finite-difference time-domain calculation in a specific structure such as a metal nanoantenna described that an enhanced field exists locally at the edge of metal particles [10]. Surface-enhanced Raman scattering indicated FLE as a physical effect [11].
The present study specifically examined physical effects of FLE in a general form of evaporated island film, not a unique structure such as a nanoantenna or a ring resonator [12]. To verify the FLE, a method of blocking the first layer can be conceived. Specifically, a blocking monolayer is formed only on surfaces of nanoparticles prepared using thermal evaporation or a similar process. The molecule to be measured is deposited thereon. Results are then compared with measured results for a normal sample without a blocking layer. The value of the first layer can be evaluated from this comparison, as shown in figure 1.
Figure 1. Schematic drawing of samples: (a) without blocking layer and (b) with blocking layer (SAM).
Download figure:
Standard image High-resolution imageTherefore, as described herein, after two equivalent substrates are placed as mutually adjacent in the holder, equivalent nanoparticle films are deposited simultaneously. After forming self-assembled monolayer (SAM) on only one side, chemically stable polymer is deposited by spin coating to block any direct polymer–metal contact. The polymer is deposited directly on the other gold film to facilitate verification of the first layer effect by observation and comparison of the absorbance of the polymers of these two samples.
2. Experimental
2.1. Infrared measurement
Infrared absorption spectra were obtained at normal incidence of radiation using a Fourier transform infrared spectrophotometer (FT-IR, MB-100; ABB Bomem) equipped with a standard deuterated triglycine sulfate (DTGS) detector averaging 128 scans at 4 cm−1 resolution. The incident light is irradiated to the surface of PAA film.
2.2. Chemical materials
As blocking material for the first layer, we used reagent grade p-nitrothiophenol (PNTP; Fujifilm-wako Pure Chemical Inds. Ltd) without further purification because of its self-assembling properties. After ultrasonic cleaning of substrate with acetone, a self-assembled PNTP monolayer was prepared on the gold film on a Si substrate by immersion of a gold film on Si into 2 mM ethanolic solutions of PNTP for 0.5 h at room temperature. Then the monolayer was rinsed thoroughly with ethanol. Morphologies of the Au films deposited on a Si wafer were observed using a field-emission scanning electron microscope (FE-SEM, JSM-7000F; JEOL Ltd) at 10 keV.
As the infrared absorbing material, polyacrylic acid (PAA; Fujifilm-wako Pure Chemical Inds. Ltd) was spin-coated onto the substrate with Au evaporated film. A 200 μL droplet of 10 g L−1 PAA ethanol solution was dropped on the spin-coater-mounted substrate. Before infrared measurements, the substrate was rotated (5400 rpm, 5 min) and dried to evaporate the ethanol.
2.3. Thermal evaporation
In a vacuum chamber of 2 × 10−7 Pa at room temperature, gold films of 2 or 3 nm mass thickness were deposited simultaneously on two adjacent silicon wafers using electron-beam evaporation. The mass thickness of the Au films was found using a quartz crystal oscillator (XTM/2; Leybold Inficon) placed adjacent to the Si surface. After evaporation, some Au films were annealed in situ at 300 °C in an ultra-high-vacuum chamber (2.2 × 10−8 Pa base pressure). Each sample was kept for 2 h, then cooled.
3. Results and Discussion
Figure 2 depicts a field emission scanning electron microscopy image of the as-deposited 2 nm Au film. Results show formation of a typical island film. The infrared absorption spectrum of the self-assembled PNTP monolayer formed on the as-deposited 2 nm Au film is presented in figure 3. Bands peculiar to PNTP and appearing clearly at 1340 and 852 cm-1 were assigned respectively to νs(NO2) and δ(NO2), which indicates that SAM is formed on Au. The clearly observed SAM spectrum confirms that SAM is formed at the site where absorption enhancement occurs.
Figure 2. SEM image of as depo 2 nm Au film.
Download figure:
Standard image High-resolution image
Figure 3. Infrared absorption spectrum of self-assembled monolayer of PNTP formed on as depo 2 nm Au film.
Download figure:
Standard image High-resolution imageFigure 4 portrays the IR absorption spectrum of PAA deposited on the as-deposited 2 nm Au film. The solid line shows results without SAM. Open circles show those with SAM. Results obtained with and without SAM match within experimental error, indicating that FLE does not exist. The absorption band by PNTP is readily apparent (figure 3), indicating that SAM is formed even if the absorption enhanced site is not perfectly covered by the SAM. In other words, if FLE exists, then some difference in the absorption intensity from the site covered with SAM would be apparent if the enhanced site with no SAM even partially remains. Therefore, perfect coverage of the enhanced site by SAM is unnecessary. Furthermore, the formed PNTP layer is not very thick because the IR absorption intensity of PAA is consistent with or without SAM. The two spectra mutually coincide within experimental error, demonstrating that the deposited PNTP volume is only negligibly smaller than the PAA volume.
Figure 4. The IR absorption spectrum of PAA deposited on as-deposited 2 nm Au film: the solid line represents data without SAM, open circles represent data with SAM.
Download figure:
Standard image High-resolution imageSimilar measurements of films having other morphologies were conducted to ascertain whether the absence of FLE presents difficulties peculiar to the as-deposited 2 nm film. Figures 5 (3 nm, as-deposited) and 6 (3 nm, post-annealed) respectively portray SEM images of gold films (a) and the infrared absorption spectra of PAA deposited on the film (b). The nanoparticle films were created with various particle sizes and inter-particle distances. The solid line shows the PAA spectrum without SAM. Open circles show the spectrum with SAM. For all films, results without SAM and with SAM are consistent within experimental error. All results, obtained with or without SAM, are consistent for a deposited island film having various forms, which clarifies that FLE does not exist as a physical effect of the thermal evaporated island film.
Figure 5. SEM image and the IR absorption spectrum of PAA deposited on as-deposited 3 nm Au film: the solid line represents data without SAM, open circles represent data with SAM.
Download figure:
Standard image High-resolution image
Figure 6. SEM image and IR absorption spectrum of PAA deposited on post-annealed 3 nm Au film: the solid line represents data without SAM, open circles represent data with SAM.
Download figure:
Standard image High-resolution image
In the surface-enhanced infrared spectrum of an electrode system, FLE has been reported [13, 14]. However, these FLE are attributable to chemical effects in chemisorbed molecules. That condition differs from the cases examined in the present study, which uses physisorbed polymer. Moreover, FLE examined using Raman spectroscopy, as reported by Otto [11] and FLE examined using FT-IR, as reported by Otto et al [15] are only useful if the wave vector is preserved. Atomic-scale surface roughness is necessary to preserve wave vectors in these dynamic charge transfer processes, but atomic-scale roughness does not exist in island films deposited under the conditions used for the present study. Therefore, FLE is considered not to exist in this study.
In conclusion, the first layer effect of physical enhancement mechanism in surface-enhanced infrared absorption was experimentally demonstrated by this study. Comparison between the polyacrylic acid film absorbance measured with and without a self-assembled monolayer of PNTP reveals that the intensities are independent of the monolayer presence. Results suggest that the FLE in the physical enhancement mechanism does not exist in this thermally evaporated island film.