One-dimensional plasmonic hotspots located between silver nanowire dimers evaluated by surface-enhanced resonance Raman scattering

Hotspots of surface-enhanced resonance Raman scattering (SERRS) are localized within 1 nm at gaps or crevices of plasmonic nanoparticle dimers. We demonstrate SERRS hotspots with volumes that are extended in one dimension tens of thousand times compared to standard zero-dimensional hotspots using crevices of plasmonic nanowire dimers. According to the polarization measurements, a plasmon resonance in the direction along the dimer width generates SERRS hotspots. SERRS images show oscillating patterns between edges of the hotspot. The SERRS intensity becomes the strongest at the edges, indicating that Fabry-Perot-type resonance of surface plasmons is involved in the Raman enhancement. These optical properties of the SERRS hotspots are quantitatively reproduced by numerical calculations based on the electromagnetic (EM) mechanism. EM coupling energy between dye molecule excitons and plasmons is evaluated using spectral changes in plasmon resonance reflected in a loss of SERRS activity at the hotspots. The coupling energies are consistent with the calculated EM enhancement factors.

Figure 1

(a1)–(a4) Elastic light scattering spectra of isolated NWs. Insets: dark-field images (left panels) and enhanced dark-field images (right panels), respectively. Spectra correspond to white circles in dark-field images of the insets. Images of (b) are the elastic light scattering, (c) SERRS, and (d) SEM of the NWs obtained from the same area on the sample glass plate covered with an ITO film. (e) Enlarged SEM images of the NWs corresponding to white boxes (1–3) in (b–d). Scale bars in (a–d) and (e) are 10 and 1.0 μm, respectively.

Figure 2

(a1) SEM image of the NW dimer. Insets: (a2) and (a3) enlarged SEM images of the edges of the dimer crevice. (b),(c) Dark-field and SERRS images of the dimer. (d) Superposition of the SEM and the dark-field image. (e) Superposition of the SEM and SERRS images. Scale bars in (a)–(c) are 5 μm.

Figure 3

(a) Spectrum of the elastic light scattering for the NW dimer and (inset) monomer perpendicular (red line) and parallel (black line) to the dimer length. (b) Spectrum of the SERRS intensity perpendicular (red line) and parallel (black line) to the dimer length. (c) Polarization dependence (circles and line) of the plasmon resonance maxima at 520 nm (red) and 720 nm (black). (d) Polarization dependence of the SERRS intensity at 560 nm (red circles and red line).

Figure 4

(a1),(a2) Spectra of the elastic light scattering efficiency for the direction perpendicular to the NW dimer length and (insets) corresponding dark-field images. (b1),(b2) Spectra of the SERRS intensity for the direction perpendicular to the NW dimer length and (insets) corresponding images. Red and blue spectra correspond to red and blue circles in insets, respectively. (c) First peak wavelengths vs the second peak wavelengths in the elastic light scattering spectra of the NW dimers; (○) and (×) denote the presence and absence of the second peak, respectively. (d1) SERRS image of the NW dimer showing spatial oscillatory patterns along the NW dimer length. (d2),(d3) SERRS images perpendicular and parallel to the NW dimer length, respectively. (d4) Enhanced SERRS image of (d3). Scale bars in (a)–(d) are 5.0 μm.

Figure 5

(a) Image of a NW dimer and direction of the incident light (yellow arrow) with the coordinate system used for the FDTD calculations. Red and black arrows indicate the direction of the light polarization. (b) Calculated spectra of the elastic light scattering for the NW dimer and (inset) monomer perpendicular (red line) and parallel (black line) to the dimer (monomer) length $(X$ axis). (c1)–(e1) Images of the dimer cross section in $X,Y$, and $Z$ planes, respectively. Red and black arrows indicate the direction of the polarization. (c2)–(e2) EM field distributions at 532 nm for the incident light polarization along the dimer width $(Z$ axis). (c3)–(e3) EM field distributions at 532 nm for the incident light polarization along $X$.

Figure 6

(a) NW dimer scattering cross-section spectra for different NW diameters for the elastic light scattering in the direction perpendicular to the dimer length: (○), (◊), (▾), (△), and (□) correspond to the diameters of 110, 100, 90, 70, and 50 nm, respectively. Inset: Peak wavelength of the dipolar plasmon resonance vs the NW diameter. (b) Spectra of the maximum EM field amplitude in the $X$ plane and (c) $Z$ plane for the incident light polarization perpendicular to the dimer length for the NW diameters in (a). Insets: Respective maximum EM field amplitudes at 532 nm vs the NW diameter.

Figure 7

Elastic light scattering spectrum of the NW dimer (a1) showing the SERRS activity (a2) after a loss of the SERRS activity, (a3) after recovery of the SERRS activity. (b) Linewidth vs peak energy for the elastic light scattering spectra of the NW dimers (red) showing the SERRS activity and (black) after a loss of the SERRS activity. (c1) Experimental elastic light scattering spectra of the NW dimer (red line) showing the SERRS activity and (black line) after a loss of the SERRS activity. (c2) Elastic light scattering spectrum calculated using Eq. (3) with the coupling energy of 100 meV (red line) and 0 meV (black line). (d) Linewidth vs peak energy for the elastic light scattering spectra calculated using Eq. (3) with the coupling energy changing from 150 to 0 meV. The values of the plasmon resonance energies for 0 meV coupling energy are (▾) 2.14, (◊) 2.21, and (○) 2.29 eV for the width of 580 meV, and (□) 2.21, (△) 2.29 eV for the width of 490 meV.