Reply to “Comment on ‘Scanning-probe Raman spectroscopy with single-molecule sensitivity' ”

In the Comment by Domke and Pettinger [Phys. Rev. B 75, 236401 (2007)] on our study on “Scanning-probe Raman spectroscopy with single-molecule sensitivity” [C. C. Neacsu et al. Phys. Rev. B 73, 193406 (2006)] the authors raise concerns whether our tip-enhanced Raman response is due to carbon clusters as molecular decomposition products of the malachite green molecules used. Their arguments are based on the different spectral characteristics we observe between the tip-enhanced and far-field Raman response. Here, we show the results of systematic control experiments to support the conclusions drawn by Neacsu et al. that all spectral features can be related to malachite green molecules. Comparing Raman spectra for different degrees of near-field enhancement, the absence of spectral changes during bleaching, examining a broad spectral range, and the explicit comparison with Raman signals from carbon impurities validates our original assignment, including the single-molecule response.

Figure 1

(Color online) [(a)–(d)] Tip-enhanced Raman spectra for $\sim 1$ ML malachite green for different degrees of enhancement in comparison with the corresponding far-field Raman spectrum (bottom graph) and the result of DFT calculation for the mode assignment. The Raman enhancement factors derived are (a) $3\times {10}^8$, (b) $7\times {10}^7$, (c) $1\times {10}^7$, and (d) $1\times {10}^6$, respectively, with an experimental uncertainty of a factor of 3–5 for each value. Data are acquired for (a) $1\mathrm{s}$, [(b) and (c)] $100\mathrm{s}$, and (d) $30\mathrm{s}$, respectively. Spectral resolution is $25{\mathrm{cm}}^{-1}$ for the near-field and $1{\mathrm{cm}}^{-1}$ for the far-field spectra.

Figure 2

(Color online) Time series of 100 consecutive near-field Raman spectra (acquired for $1\mathrm{s}$ each) for $\sim 1$ ML MG on Au (Raman enhancement of $1.3\times {10}^7$). The Raman spectrum clearly resembles the far-field spectrum. The signal decays to zero due to bleaching. No new spectral features appear from the photoreaction products.

Figure 3

(Color online) Bleaching kinetics derived from Fig. 2 for the spectrally integrated Raman time series and the region from $1550\text{to}1650{\mathrm{cm}}^{-1}$ (inset). An exponential decay constant of $\tau =40\pm 5\mathrm{s}$ results for the enhancement of $1.3\times {10}^7$ and the incident laser fluence of $3\times {10}^4\mathrm{W}∕{\mathrm{cm}}^2$ in this experiment.

Figure 4

(Color online) Left: time series of tip-scattered Raman spectra (Fig. 3 in Ref. 2). Right: comparison between sum spectrum of data shown in left panel (olive) and tip-enhanced Raman spectra for different degrees of enhancement [red, (a) and (b)] and sum spectrum of data shown in Fig. 2 blue).

Figure 5

(Color online) Tip-enhanced (enhancement $\sim {10}^8$) and far-field MG Raman spectrum for $\sim 1$ ML of MG including the higher-energy spectral range. Although spectrally different from the far-field response, the absence of any spectral signature above $1800{\mathrm{cm}}^{-1}$ for the tip-enhanced response contrasts the observation for carbonaceous species.

Figure 6

(Color online) Time series of consecutive near-field Raman spectra acquired for a photodecomposed MG multilayer. The large and uncorrelated spectral fluctuations and the band at $1870–2100{\mathrm{cm}}^{-1}$ are characteristic for carbon clusters.

Figure 7

(Color online) Illustration of calculated Raman active normal modes of the MG molecule at (a) $1293{\mathrm{cm}}^{-1}$ and (b) $1544{\mathrm{cm}}^{-1}$. (a) In-plane $\mathrm{C}\mathrm{H}$ deformation mode together with $\mathrm{C}\mathrm{C}$ stretching of the methane group. This mode is observed in all near-field spectra irrespective of the enhancement level. (b) skeletal stretching motions combined with in-plane $\mathrm{C}\mathrm{H}$ bending motions of the conjugated di-methyl-amino-phenyl rings. This mode is not present in the FF spectrum, but its intensity grows with the near-field enhancement (see Fig. 1).