Nanoscale analysis of multiwalled carbon nanotube by tip-enhanced Raman spectroscopy
Introduction
Carbon nanotubes (CNTs), particularly single-walled carbon nanotubes (SWNTs), have been extensively studied and utilized in many applications owing to their one-dimensional properties and their wide range of electronic properties depending on their chiral indices. Multiwalled carbon nanotubes (MWNTs), which comprise two or more CNT 'walls' with characteristics strongly depending on the chirality of each wall, are also promising for applications. For example, it has been shown that the outermost wall of an MWNT can be functionalized while preserving the SWNT-like properties of the inner wall [1]. The dependence of transport properties on the chirality of the inner and outer walls has also been demonstrated in a CNT field effect transistor device [2]. Therefore, the study of MWNTs and the interaction between CNT walls is of fundamental importance.
Raman spectroscopy has become a powerful tool for the study of sp2 carbon systems. The three most prominent spectral features, namely the D, G, and 2D bands, serve as the 'fingerprint' for carbon materials and, also, provide information on the sample volume, the stress and strain, and the stacking orientation [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14]. Additionally, low-energy Raman modes can be used to identify the different allotropes, for example, the radial breathing mode (RBM) (100–400 cm−1) for CNTs and the layer breathing (94 cm−1) and torsion modes (52 cm−1) for twisted bilayer graphene [15], [16], [17]. In this study, we focus on the relationship between the G band and 2D band in the Raman spectra of an MWNT.
The G band is usually the most intense peak in the Raman spectrum of bulk graphite, hence its name. The second most intense peak is the 2D band, which is the overtone of the defect-induced D band, but since the momentum conservation of the 2D band is satisfied by two phonons with opposite wave vectors, it can be observed in defect-free graphite and for this reason it is also referred to as the G′ band [5], [11]. Even though the G and 2D band intensities are usually proportional to the volume of the bulk sample, this phenomenon does not occur at the nanoscale, for which it is widely accepted that the ratio between the 2D and G band intensities is proportional to the number of graphene layers rather than the absolute intensity [5], [11], [13].
An interesting property of the 2D band that differs from the G band is its sensitivity to the number of graphene layers and the stacking orientation. For example, the integrated intensity ratio and intensity ratio of the 2D band to the G band increase with the number of CVD-grown turbostratic stacked graphene [10], [13], [14]. However, for turbostratic stacked graphene prepared by mechanically placing one graphene layer on top of another, the above relationship is not observed since there is very weak interaction between the layers, thus the relative intensities of the Raman peaks resemble those for single-layer graphene [4], [5], [10]. For Bernal stacked graphene, the intensity ratios decrease with increasing number of layers owing to the large energy splitting between the 2D band components, which results in broadening of the 2D band [3], [5], [11]. This phenomenon is caused by quantum interference between the 2D band phonon scattering paths, which is induced by the π–π interaction between graphene layers. Therefore, this phenomenon should be observable in other sp2 carbon systems similar to stacked graphene, such as MWNTs. The π–π interaction between the CNT walls should also exhibit the quantum interference phenomenon.
The electronic structure of MWNTs is similar to that of stacked graphene in many ways, the main difference being the quantization of the wave vector. The π–π interaction between CNT walls is expected to exhibit quantum interference similar to that of turbostratic stacked graphene since an MWNT consists of stacked CNT walls with different chiral indices [18]. Thus, the intensity ratios should increase with increasing number of CNT walls.
However, because of the one-dimensional geometry of CNTs, conventional Raman spectroscopy would yield a spectrum averaged over all CNTs within the laser beam spot, resulting in the loss of detailed information at the nanoscale [16]. Thus, we cannot study individual MWNTs using a conventional Raman system. On the other hand, tip-enhanced Raman spectroscopy (TERS) is an emerging powerful Raman spectroscopy technique capable of achieving nanoscale resolution. By irradiating a laser onto the tip of a sharp metallic probe of a scanning microscope such as an atomic force microscope (AFM) or a scanning tunneling microscope (STM), the induced surface plasmons result in an intensified electromagnetic field confined at the tip apex, with which an enormous amount of Raman scattering can occur, and spatial resolution down to 1 nm has been demonstrated [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30]. In this study, we observed the relationship between and and the number of CNT walls in the same MWNT by TERS.
Using STM-TERS (referred to simply as TERS, hereafter), we first discuss the “difference spectra”, in which the far-field background has been eliminated from the TERS spectrum. As a result, we were able to obtain reliable Raman spectra at different points along an MWNT. The and ratios vary in a steplike manner, which corresponds well with the number of CNT walls. Additionally, at the crossing point between two CNT bundles, an increase in the 2D band intensity was observed. These results indicate that the π–π interaction between CNT walls is similar to the interlayer interaction in turbostratic stacked graphene. Since the peak splitting of the 2D band did not increase with the number of layers, we attribute this phenomenon to the quantum interference between Raman scattering paths of the 2D band phonons, in which phonons from different paths exhibit similar phonon energies, thus appearing as a single peak. To the best of our knowledge, such a phenomenon has never been directly investigated for CNTs. These results provide a better understanding of the interaction between CNTs and a means of determining the number of CNT walls by utilizing the characteristics of the 2D band.
Section snippets
Experimental methods
The TERS experiments were carried out with an ultra-high vacuum (UHV) STM system (Unisoku Co., USM1400) controlled by an STM100 controller (RHK Technology Inc.), equipped with a Raman spectrometer (Nanofinder FLEX, Tokyo Instruments Inc.). All STM images were acquired at room temperature under a pressure below 1.0 × 10−8 Pa in constant-current mode using electrochemically etched Ag tips (Unisoku Co.). Commercially available 300-nm-thick Ag(111) films evaporated on mica (Georg Albert PVD) was
Results and discussion
Fig. 1(a) shows an STM topographic image of an isolated SWNT adsorbed on a Ag(111) substrate. The SWNT has an apparent height of 1 nm, estimated from the cross-sectional profile in Fig. 1(b). Fig. 1(c) shows the difference Raman spectrum of the SWNT, which is the difference between the near-field (tip approached) and far-field spectra (tip retracted) observed at the position indicated with an × in Fig. 1(a). The inset of Fig. 1(c) shows the raw near-field and far-field spectra, which were
Conclusion
The relationship between the 2D band and G band intensities and the number of CNT walls in an MWNT was investigated. The difference spectrum was used for the analysis of TERS data. By using the TERS spectrum on the substrate, we eliminated the far-field spectral background from the near-field spectrum.
We discovered a steplike feature of the value, which was related to the number of CNT walls in the MWNT. This phenomenon can be explained by the quantum interference between Raman scattering
Acknowledgements
This work was supported by a Grant-in-Aid for Scientific Research (S) (#24221009) from the Ministry of Education, Culture, Sports, Science and Technology, Japan. The corresponding author would like to express his gratitude towards the Quantum Engineering Design Course program for financial support.
References (38)
- et al.
Raman spectroscopy on isolated single wall carbon nanotubes
Carbon
(2002) - et al.
Raman spectroscopy of carbon nanotubes
Phys. Rep.
(2005) - et al.
Optical and electrical properties of inner tubes in outer wall-selectively functionalized double-wall carbon nanotubes
J. Phys. Chem. C
(2011) - et al.
Chirality-dependent transport properties of double-walled nanotubes measured in situ on their field-effect transistors
J. Am. Chem. Soc.
(2009) - et al.
Raman studies of monolayer graphene: the substrate effect
J. Phys. Chem. C
(2008) - et al.
Raman spectroscopy of graphene on different substrates and influence of defects
Bull. Mater. Sci.
(2008) - et al.
Spatially resolved raman spectroscopy of single- and few-layer graphene
Nano Lett.
(2007) - et al.
Raman spectra of misoriented bilayer graphene
Phys. Rev. B
(2008) - et al.
Angle-resolved raman imaging of interlayer rotations and interactions in twisted bilayer graphene
Nano Lett.
(2012) - et al.
Raman spectroscopy study of rotated double-layer graphene: misorientation-angle dependence of electronic structure
Phys. Rev. B
(2012)