Single- and double-resonance Raman $G$-band processes in carbon nanotubes

This work reports Raman spectroscopy measurements from single wall carbon nanotubes (SWNTs) addressing the nature of the $G$-band resonance Raman spectra. Experimental results on different samples are presented, i.e., aligned and misaligned SWNT samples, as well as isolated and bundled SWNTs. It is shown that the Raman spectra from nondefective SWNTs, both isolated and bundled, are composed of strong first-order single resonance Raman features. Defective materials, however, are found to exhibit lower intensity spectra with contributions from both single resonance and defect-induced double resonance features.

Figure 1

(Color online) $G$-band resonance Raman spectrum from a fiber of aligned SWNT bundles $(E_{\text{laser}}=2.71\mathrm{eV})$. The inset to the upper panel shows an optical image of the sample. The spectrum in the upper panel was taken from location $\mathbf{1}$ (see upper inset) with the (ZZ) polarization scattering geometry, and the spectrum in the lower panel was acquired from location $\mathbf{2}$. The inset to the lower panel shows the $E_{\text{laser}}$ dependence of the two peaks indicated by arrows. The solid curves in this inset are predictions for the $E_{\text{laser}}$ dependence of the $G$-band double resonance features (Refs. 6, 7).

Figure 2

The left panels show the $G$-band Raman spectra in the (ZZ) polarization geometry at location $\mathbf{1}$ [see inset in Fig. 1(a)], using four different laser energies, as displayed in the figure. The right panels show the spectra measured with the same laser energies in location $\mathbf{2}$ of the sample. The peak frequencies are displayed in ${\mathrm{cm}}^{-1}$.

Figure 3

The $D$ (a), $G^{-}$ and $G^{+}$ (b) band frequencies observed for different isolated nanotubes on a sapphire substrate (open symbols) and on a $\mathrm{Si}∕\mathrm{Si}{\mathrm{O}}_2$ substrate (filled symbols) using several different laser lines. The stars symbol data points come from Ref. 8. The dark solid line in (a) shows the prediction from the double resonance model for the $D$-band ${\omega }_D$. The solid line was downshifted to fit the experimental results according to Ref. 18.