In Situ Raman Spectroelectrochemistry as a Tool for the Differentiation

Anal. Chem. 2007, 79, 9074-9081

In Situ Raman Spectroelectrochemistry as a Tool for the Differentiation of Inner Tubes of Double-Wall Carbon Nanotubes and Thin Single-Wall Carbon Nanotubes Martin Kalba´cˇ,*,†,‡ Ladislav Kavan,† and Lothar Dunsch‡

J. Heyrovsky´ Institute of Physical Chemistry, v.v.i. Academy of Sciences of the Czech Republic, Dolejsˇ kova 3, CZ-182 23 Prague 8, Czech Republic, and Leibniz Institute of Solid State and Materials Research, Helmholtzstrasse 20, D-01069 Dresden, Germany

In situ Raman spectroelectrochemistry has been used to distinguish between thin single-wall carbon nanotubes (SWCNT) and the inner tubes of double-wall carbon nanotubes (DWCNT). The spectroelectrochemical method is based on the different change of the electronic structure of the inner tube in DWCNT and that of SWCNT during electrochemical charging, which is reflected in the Raman spectra. During electrochemical charging the inner tubes of DWCNT exhibit a delayed attenuation of the intensities of their Raman modes as referred to the behavior of SWCNT of similar diameter. The changes are pronounced for the radial breathing mode (RBM), and thus, these modes are diagnostic for the distinction of inner tubes of DWCNT from the thin SWCNT. The different sensitivities of inner and outer tubes to the applied electrochemical charging is a simple analytical tool for differentiation of SWCNT and DWCNT in a mixture. The significance of the proposed method is demonstrated on a commercial DWCNT sample. Double-wall carbon nanotubes (DWCNT) are considered as a special case of multiwall carbon nanotubes (MWCNT) with one inner and one outer tube. Both the inner and outer tubes are experimentally addressable by Raman spectroscopy. Hence, it is possible to study how any action applied on the outer tube is projected on the inner tube.1-3 High-quality DWCNT can be prepared by a heat treatment of fullerene peapods.4 The frequency of the radial breathing mode (RBM) is generally dependent on the tube diameter.5,6 Since the diameter difference of the inner * Corresponding author. E-mail: [email protected]. † Academy of Sciences of the Czech Republic. ‡ Leibniz Institute of Solid State and Materials Research. (1) Kavan, L.; Kalbac, M.; Zukalova, M.; Krause, M.; Dunsch, L. ChemPhysChem 2004, 5, 274-277. (2) Kalbac, M.; Kavan, L.; Zukalova, M.; Dunsch, L. Adv. Funct. Mater. 2005, 15, 418-426. (3) Kavan, L.; Dunsch, L. ChemPhysChem 2007, 8, 975-998. (4) Bandow, S.; Takizawa, M.; Hirahara, K.; Yudasaka, M.; Iijima, S. Chem. Phys. Lett. 2001, 337, 48-54. (5) Jorio, A.; Saito, R.; Hafner, J. H.; Lieber, C. M.; Hunter, M.; McClure, T.; Dresselhaus, G.; Dresselhaus, M. S. Phys. Rev. Lett. 2001, 86, 1118-1121. (6) Bandow, S.; Takizawa, M.; Kato, H.; Okazaki, T.; Shinohara, H.; Iijima, S. Chem. Phys. Lett. 2001, 347, 23-28.

9074 Analytical Chemistry, Vol. 79, No. 23, December 1, 2007

and outer tubes is around 0.7 nm, the RBM of those inner tubes prepared by the coalescence of fullerenes inside single-wall carbon nanotubes (SWCNT) are very well distinguished in the Raman spectra.4 High-quality DWCNT can be also prepared by other methods like chemical vapor deposition (CVD).7 The CVD production of DWCNT is obviously cheaper than the preparation based on transformation of fullerene peapods and can be scaled up for industrial purposes. However, depending on experimental conditions used in the CVD process, a mixture of nanostructures is obtained containing SWCNT, DWCNT, and MWCNT. Furthermore, the CVD synthesis of nanotubes provides samples of relatively wide diameter distribution, and thus, the diameters of SWCNT and DWCNT are usually similar. The knowledge of the composition of the samples is crucial for the further studies of electronic, transport, and mechanical properties of these nanotube samples. The best technique to characterize the samples is highresolution transmission electron microscopy (HRTEM). However, the intrinsic drawback of this technique is that the microscopic images are recorded only for a limited amount of a sample, which may or may not be representative for the bulk material. Thus, it is necessary to find a method for routine characterization of larger amounts of samples. Raman spectroscopy is usually the method of choice to characterize nanotube samples due to resonance enhancement. However, a simple Raman spectroscopic measurement is not applicable to the evaluation of samples with respect to the presence of DWCNT and SWCNT, because it is impossible to distinguish between SWCNT and the inner tubes of DWCNT of similar diameters. Here we show that in situ Raman spectroelectrochemistry overcomes this problem and indeed can distinguish and differentiate between SWCNT and the inner tubes of DWCNT. This is because the electrochemical behavior of an inner tube during electrochemical charging is different from that of outer tubes.2 Recently Kim et al.8 proposed to use a combination of Raman spectroscopy and chemical doping of DWCNT with concentrated H2SO4 as a “possible method” to evaluate the purity of the sample (7) Kim, Y. A.; Muramatsu, H.; Hayashi, T.; Endo, M.; Terrones, M.; Dresselhaus, M. S. Chem. Vap. Deposition 2006, 12, 327. (8) Kim, Y. A.; Muramatsu, H.; Kojima, M.; Hayashi, T.; Endo, M.; Terrones, M.; Dresselhaus, M. S. Chem. Phys. Lett. 2006, 420, 377-381. 10.1021/ac071205u CCC: $37.00

© 2007 American Chemical Society Published on Web 11/01/2007

Figure 1. Potential-dependent Raman spectra (excited at 1.92 eV) of SWCNT (HiPco) on a Pt electrode in 0.2 M LiClO4 + acetonitrile as measured by in situ Raman spectroelectrochemistry. The bold line indicates the Raman spectra measured at the starting potential. Spectra are offset for clarity, but the intensity scale is identical for all spectra in the respective window.

with respect to the content of DWCNT. Chemical doping thus can also be used to differentiate the Raman spectra of inner and outer tubes, respectively. However, the level of chemical doping is hard to control, and consequently, it is very difficult to reproduce the doping conditions precisely for different samples. On the other hand, electrochemical doping is known to be very well and easy controllable. As we will show in this work the precise control of the doping level is crucial for interpretation of the data. Thus, we suggest the combination of electrochemical doping and Raman spectroscopy as a technique to analyze the DWCNT/SWCNT mixtures, which can be done in a single experimental step by in situ Raman spectroelectrochemistry. In the present work we first demonstrate the difference in electrochemical behavior of expeapod DWCNT and SWCNT, and then we apply the spectroelectrochemical method to analyze a commercial sample of DWCNT. EXPERIMENTAL SECTION The samples of C60@SWCNT (peapods) were prepared by the reaction of SWCNT with gaseous C60.9 The filling factor has been estimated to be higher than 80%.9 The ex-peapod DWCNT nanotubes were obtained by a heat treatment of peapods at 1200 °C for 8 h in vacuum (10-6 Pa). The commercial sample of DWCNT has been purchased from Nanocyl. The SWCNT fabricated by the high-pressure catalytic decomposition of carbon monoxide (HiPco) were available from our previous study.10 In situ spectroelectrochemical experiments were done at a Ptsupported thin film of nanotubes. These films were fabricated by the evaporation of a sonicated ethanolic slurry of nanotubes and (9) Kavan, L.; Dunsch, L.; Kataura, H.; Oshiyama, A.; Otani, M.; Okada, S. J. Phys. Chem. B 2003, 107, 7666-7675. (10) Kavan, L.; Rapta, P.; Dunsch, L. Chem. Phys. Lett. 2000, 328, 363-368.

outgassed overnight at 70 °C in vacuum. We used thin films to minimize the sample consumption. The film thickness was typically about 1 µm. We did not observe any limitations caused by the film thickness. Only very thick layers of nanotubes can delaminate from the substrate due to problems with adhesion. The electrode was mounted in a spectroelectrochemical cell in a glove box under nitrogen. The cell was equipped with a Pt counter electrode and an Ag wire pseudoreference electrode. The electrolyte solution was 0.2 M LiClO4 in dry acetonitrile (Aldrich). Electrochemical doping was carried out potentiostatically (PG 300HEKA potentiostat). The Raman spectra were measured on a T-64000 spectrometer (Instruments, SA) interfaced to an Olympus BH2 microscope; the laser power impinging on the cell window was between 1 and 5 mW. Spectra were excited by Ar+ or Kr+ laser (Innova 300 series, Coherent). The Raman spectrometer was calibrated before each set of measurements by using the F1g line of Si at 520.2 cm-1 for reference. RESULTS AND DISCUSSION Figures 1-3 show the in situ spectroelectrochemical data obtained on HiPco nanotubes using 1.92, 1.83, and 2.18 eV laser excitation energies, respectively. The resulting data are in a good agreement with those reported previously.11 Electrochemical doping of the samples leads to an overall bleaching of the Raman signal of these materials. The RBMs are more sensitive to electrochemical potential than the tangential displacement (TG) mode. The former is almost completely vanished at the minimum and maximum applied potentials (-1.5 and 1.5 V, respectively). The bleaching is roughly symmetrical for cathodic and anodic doping which reflects the mirror pairs of van Hove singularities. (11) Kavan, L.; Kalbac, M.; Zukalova, M.; Dunsch, L. J. Phys. Chem. B 2005, 109, 19613-19619.

Analytical Chemistry, Vol. 79, No. 23, December 1, 2007

9075



The development of the spectra of SWCNT is in general independent of the direction of the potential scan. To ensure the equilibrated condition we measured at a constant electrode potential when the current dropped bellow detection limit. The small deviation from the ideal behavior can occur due to decomposition of the sample at high potentials. The TG mode is furthermore up-shifted at high anodic potentials (1.5 V), whereas 9076 Analytical Chemistry, Vol. 79, No. 23, December 1, 2007

there is no significant change of the position at high cathodic potentials (-1.5 V). We have shown previously that the extensive doping of SWCNT leads to a strong decrease of the intensity of the Raman signal.11 The electrochemical charging of SWCNT causes a shift of the Fermi level. As far as the Fermi level achieves the level of the van Hove singularity it erases the electronic transitions

from/to this particular singularity. If the Raman signal is in resonance with such a singularity strong bleaching of the intensity of the Raman spectra is expected. However, a more detailed analysis of the spectra shows that the bleaching is not identical for all bands in the RBM region. For example (Figure 2), the fast bleaching of the RBM at 300 cm-1 and the relatively slow bleaching of the mode at 285 cm-1 can be followed. The obtained data are in agreement with previous results on electrochemical11,12 and chemical doping13-15 of SWCNT. The effect has been explained by assuming the diameter selectivity of chemical doping. At these conditions, the doping sensitivity was negligible for tubes of diameters from 0.9 to 1.2 nm, but it increased for both the narrower and thicker tubes. This trend was reproduced by in situ Raman spectroelectrochemistry.12 Moreover, it was found that the relative RBM attenuation depends on the electrolyte counterion, which compensates the electronic charge at SWCNT.12 Although the counterion sensitivity of n-doping correlated expectedly with the size of the chargecompensating ion,12 the nonmonotonous diameter selectivity in doping was interpreted by more complex arguments.12-15 Apparently, the nonmonotonous doping/diameter correlation12-15 would contradict the conclusion that the redox potential of nanotubes is a linear function of the band gap (or inverse diameter).16 Furthermore, the study of chemical p-doping with liquid HNO3 and H2SO4 did not confirm the nonmonotonous diameter selectivity of doping, either.17 Raman spectra were interpreted in terms of variations in the resonance condition caused by a Fermi level shift into the valence band.17 Therefore, the previously reported nonmonotonous diameter sensitivity of doping has been revised, and a new mechanism has been proposed.12 In general, the intensity of Stokes resonant Raman scattering (I) scales with the laser photon energy, EL, and the optical transition energy, Eii:

I)

c |(EL - Eii - iγ)(EL + Eph - Eii - iγ)|2

(1)

where c is a Raman matrix element (considered here as constant), Eph is the phonon energy of the vibration, and γ is the damping constant; typical values for the RBM are Eph ≈ 0.02 eV and γ ≈ 0.05 eV. Hence, the intensity of the Raman signal is at maximum if EL approaches Eii, in other words, the particular tube is in resonance. It was found that tubes that are in good resonance are also doping-sensitive. On the other hand, tubes whose optical transition energy is larger than that needed for a perfect fulfilling of the resonance condition respond less strongly to the doping. From eq 1 it is obvious that if |(EL - Eii - iγ)(EL + Eph - Eii iγ)|2 is approaching zero, then a small change of Eii would have (12) Kavan, L.; Dunsch, L. Nano Lett. 2003, 3, 969-972. (13) Kukovecz, A.; Pichler, T.; Pfeiffer, R.; Kuzmany, H. Chem. Commun. 2002, 1730-1731. (14) Kukovecz, A.; Pichler, T.; Pfeiffer, R.; Kramberger, C.; Kuzmany, H. Phys. Chem. Chem. Phys. 2003, 5, 582-587. (15) Kuzmany, H.; Kukovecz, A.; Simon, F.; Holzweber, A.; Kramberger, C.; Pichler, T. Synth. Met. 2004, 141, 113-122. (16) O’Connell, M. J.; Eibergen, E. E.; Doorn, S. K. Nat. Mater. 2005, 4, 412418. (17) Zhou, W.; Vavro, J.; Nemes, N. M.; Fischer, J. E.; Borondics, F.; Kamaras, K.; Tanner, D. B. Phys. Rev. B 2005, 71, 205423.

a dramatic effect on the spectral intensity. On the other hand, when |(EL - Eii - iγ)(EL + Eph - Eii - iγ)|2 is larger (but still small enough that the tube is found in Raman spectra), then the intensity I is less sensitive to the change of optical properties. Therefore, the dependence of the bleaching of the spectra in dependence on potential can be rationalized. Nevertheless, independent of the origin of the different bleaching behaviors of tubes with different RBM positions, it is clear that the bleaching of the nanotube bands during doping is not a simple process, and thus the interpretation must be done with caution. Figures 4-6 present the in situ spectroelectrochemical data obtained on ex-peapod DWCNT using 1.92, 1.83, and 2.18 eV laser excitation energy, respectively. The ex-peapods DWCNT has been prepared from peapods with a known diameter distribution. The simple comparison of the original Raman spectra of the peapods (not shown) and the spectra of ex-peapods DWCNT at the particular laser excitation energy provides the information which RBMs are attributed to outer and inner tubes, respectively. Furthermore, the diameter distribution of the inner tubes is limited by the diameter of the parent peapod. The optimum diameter of the tube for the formation of C60@SWCNT peapod is around 1.4 nm.18 The optimum diameter of the inner tubes is then 0.7 nm. Hence, the RBMs are well distinguished for inner and outer tubes. For example, at a laser excitation energy of 1.92 eV the outer tube RBMs are found in the region between 150 and 200 cm-1, whereas the RBMs of inner tubes are in the region between 280 and 370 cm-1. Hence, the RBM behavior during electrochemical charging can be easily followed separately for inner and outer tubes. In agreement with previous reports,1,2 the inner tube modes of expeapod DWCNT are extinguished less than the bands of the outer tube when applying an electrochemical potential (Figures 4-6), because the inner tubes are shielded by the outer tubes. This effect is most obvious on the inner tube’s RBM, but it can be also followed for the defect-induced mode at around 1320 cm-1 (D mode), the TG mode at around 1590 cm-1, the high-frequency mode at around 2500 cm-1 (G′ mode),19 and even for the intermediate frequency modes.20 Furthermore the D, TG, and G′ lines upshift in frequency at high positive potentials. This is a consequence of stiffening of the CC bonds due to electrochemical doping. However, the latter effect applies only for outer tubes and not for inner tubes. Therefore the D, TG, and G′ bands of inner and outer tubes of ex-peapod DWCNT can be distinguished at high anodic potentials.19 The doping-induced splitting of the TG mode is unique for DWCNT, and thus, it has been suggested for evaluation of the purity of DWCNT.8 However, as shown in Figure 6 the splitting of the TG mode is sometimes difficult to be distinguished. Hence, the application of this mode for the analysis of DWCNT samples is limited. On the other hand, the Raman spectra in the RBM region clearly demonstrate the presence of DWCNT. In other words the TG mode splitting at positive potentials can be used as an indication of DWCNT in the sample, but a nonsplit TG mode (18) Otani, M.; Okada, S.; Oshiyama, A. Phys. Rev. B 2003, 68, 125424. (19) Kalbac, M.; Kavan, L.; Zukalova, M.; Dunsch, L. Carbon 2004, 42, 29152920. (20) Kalbac, M.; Kavan, L.; Zukalova, M.; Dunsch, L. Chem. Eur. J. 2006, 16, 4451-4457.


9077

Figure 4. Potential-dependent Raman spectra (excited at 1.92 eV) of ex-peapod DWCNT on a Pt electrode in 0.2 M LiClO4 + acetonitrile as measured by in situ Raman spectroelectrochemistry. The bold line indicates the Raman spectra measured at the starting potential. Spectra are offset for clarity, but the intensity scale is identical for all spectra in the respective window.

Figure 5. Potential-dependent Raman spectra (excited at 1.83 eV) of ex-peapod DWCNT on a Pt electrode in 0.2 M LiClO4 + acetonitrile as measured by in situ Raman spectroelectrochemistry. The bold line indicates the Raman spectra measured at the starting potential. Spectra are offset for clarity, but the intensity scale is identical for all spectra.

does not necessarily evidence the absence of DWCNT in the tested sample. The different behaviors of the RBM of inner and outer tubes in ex-peapod DWCNT can be observed already at low electrode potentials. However, both inner and outer tube modes in expeapods DWCNT spectra are strongly bleached at a high level of doping. The potential of 1.5 V leads to a strong bleaching of all 9078 Analytical Chemistry, Vol. 79, No. 23, December 1, 2007

bands of inner tubes similarly as in the case of SWCNT. Chemical doping analogously leads to a faster bleaching of the outer tube bands.21 In the latter case the different effect of the dopant on inner and outer tubes has been explained by a three-layer capacitor (21) Chen, G. G.; Bandow, S.; Margine, E. R.; Nisoli, C.; Kolmogorov, A. N.; Crespi, V. H.; Gupta, R.; Sumanasekera, G. U.; Iijima, S.; Eklund, P. C. Phys. Rev. Lett. 2003, 90, 257403.

Figure 6. Potential-dependent Raman spectra (excited at 2.18 eV) of ex-peapods DWCNT on a Pt electrode in 0.2 M LiClO4 + acetonitrile as measured by in situ Raman spectroelectrochemistry. The bold line indicates the Raman spectra measured at the starting potential. Spectra are offset for clarity, but the intensity scale is identical for all spectra in the respective window.

model. It was concluded that most of the holes are located on the outer tube.21 The hole concentration on outer tubes was found to be approximately 10 times higher than that on inner tubes. This explains the sluggish intensity attenuation of the inner tube bands. The same conclusion was drawn also for electrochemical doping.2 Hence, during electrochemical doping the charge is located on outer tube, the van Hove singularities of inner tubes are filled at higher potential, and the inner tube bands start to bleach later. The postponed bleaching of the bands is therefore an indication of the features of the inner tubes. The bleaching behavior of RBMs seems to be reliable for the evaluation of SWCNT/DWCNT mixtures. Nevertheless, similarly as in the case of SWCNT the more detailed analysis also shows that the attenuation of inner tubes’ dependence on the applied potential is different for different RBMs. For example (Figure 5), the band at 303 cm-1 attenuates much faster than the band at 290 cm-1. Hence, it is obvious that the “resonance rule”11 is applied even for inner tubes of ex-peapods DWCNT. This is also demonstrated in Figure 7, comparing the attenuation of the Raman intensities of selected outer and inner tubes’ dependence on excitation wavelength. The situation is furthermore complicated by the presence of an outer tube which causes the postponed bleaching. Nevertheless, the direct comparison of the bleaching of the RBM of inner tubes in DWCNT and the RBM of SWCNT shows that the effect of shielding is much stronger than the effect of “resonance”. Thus, the problem can be elegantly solved by a comparison of the behavior of ex-peapod DWCNT and SWCNT spectra during electrochemical doping. This is because electrochemistry allows the precise reproduction of the doping condition for these two materials. In other words the comparison of the relative change of the RBM intensity at two different potentials

Figure 7. Potential-dependent intensity of the RBM Raman peaks of the outer tube at 180 cm-1 (full squares) and inner tube of DWCNT at 340 cm-1 (empty squares). Raman spectra were excited at 2.18 (A), 1.92 (B), and 1.83 eV (C). The intensities of the particular modes at different potentials have been normalized to the intensity at 0 V.

for SWCNT and DWCNT will provide information on the composition of an unknown sample. This is hardly possible by chemical doping as the method does not allow a controllable tuning of the doping level. Figure 8 demonstrates the applicability of our spectroelectrochemical method to distinguish between inner and outer tubes of a commercial sample, whose diameter distribution is unknown. The spectroelectrochemical data were acquired for a DWCNT material purchased from Nanocyl. There are two main groups of bands centered at around 200 and 285 cm-1, respectively. On the basis of the comparison of these data with the spectroelectrochemical behavior of the HiPco nanotubes and the ex-peapod DWCNT the bands at around 200 cm-1 correspond obviously to the outer tubes, whereas the bands at 285 cm-1 are attributed to Analytical Chemistry, Vol. 79, No. 23, December 1, 2007

9079

Figure 8. Potential-dependent Raman spectra (excited at 1.83 eV) of DWCNT (Nanocyl) on a Pt electrode in 0.2 M LiClO4 + acetonitrile as measured by in situ Raman spectroelectrochemistry. The bold line indicates the Raman spectra measured at the starting potential. Spectra are offset for clarity, but the intensity scale is identical for all spectra.

Figure 9. Potential-dependent intensity of the RBM Raman peaks at around (A) 295, (B) 285, and (C) 200 cm-1 of different nanotube samples. Raman spectra were excited at 1.83 eV. The full squares, empty squares, and triangles correspond to HiPco, DWCNT (Nanocyl), and ex-peapod DWCNT, respectively. The intensities of the particular mode at different potentials have been normalized to the intensity at 0 V. The experimental error of intensity evaluation was estimated to be about 10%.

we used the nearest frequencies at 197, 284, and 298 (HiPco) and 284 and 297 cm-1 (ex-peapod DWCNT). Depending on the purity and the overall quality of the sample, the intensities of Raman spectra can vary in a broad range. Hence,the intensities at different electrode potentials for each mode are normalized to the intensity at a potential of 0 V. Figure 9 clearly confirms that the mode at 200 cm-1 corresponds to the outer tube, whereas the modes at 285 and 295 cm-1 correspond to inner tubes. It is apparent that the Nanocyl sample contains hardly narrow SWCNT as an impurity (with RBMs at 285 and 295 cm-1), but it may or may not contain wide SWCNT (with RBMs around 200 cm-1). We must point out that a distinction between DWCNT’s outer tubes and the SWCNT is difficult based on their Raman intensity/charging profiles. Nevertheless, the thin SWCNT and the inner tubes of DWCNT can be unambiguously assigned.

the inner tubes of DWCNT. This can be refined by a detailed analysis of the charging profiles of three particular modes at around 200, 285, and 295 cm-1, which is presented in Figure 9. It is obvious that the position of the RBM of the inner tube of the DWCNT is slightly shifted with respect to the position of the RBM of SWCNT for the same particular tube. This is caused by the different environments of the SWCNT and that of the inner tubes in DWCNT. Nevertheless, the differences in bleaching of SWCNT within a narrow interval of RBM frequencies seems to be less significant than the difference between the bleaching of the RBM of inner tube and SWCNT. Therefore, the shift caused by a different environment of the tubes can be neglected for the purpose of the evaluation of SWCNT and DWCNT samples. For the comparison of the spectra of the Nanocyl DWCNT sample

CONCLUSION In situ Raman spectroelectrochemical feedback of inner tubes in DWCNT is different from that of SWCNT even if their diameters are comparable. While the SWCNT bleaches rapidly upon applying an electrochemical potential, the intensity of the signal of inner tubes remains almost unchanged. This effect is also mirrored in the TG mode spectra where the TG mode splits upon application of high positive potential for DWCNT but shifts only for SWCNT. However, the actual splitting of the TG mode depends on the laser excitation energy, and thus, the absence of a splitting is not a direct proof for the absence of DWCNT in the sample. Contrary to chemical doping, electrochemistry provides much more detailed information on the sample. The attenuation of the RBM of inner tubes in DWCNT is weaker, because the inner tubes are shielded by the outer tubes. Thus, it is possible to identify the RBM of inner tubes and that of thin SWCNT even in their

9080 Analytical Chemistry, Vol. 79, No. 23, December 1, 2007

mixture. However, a special attention must be paid to distinguish between the two different effects causing the delayed bleaching of the signal in Raman spectra: (1) the relatively large distance of the optical transition of particular nanotube from the energy of laser excitation (“resonance rule”) and (2) the shielding of inner tubes due to presence of an outer tube in DWCNT. We show that a precise electrochemical doping allows a proper evaluation of the data. For the best accuracy we propose to use standard samples of DWCNT and SWCNT for the quantification of the bleaching behavior.

ACKNOWLEDGMENT This work was supported by the Academy of Sciences of the Czech Republic (Contract No. A400400601), by the Czech Ministry of Education, Youth and Sports (Contract No. LC-510), and by IFW Dresden. M. Kalbac acknowledges funding by the Alexander von Humboldt society.

Received for review June 7, 2007. Accepted September 13, 2007. AC071205U


9081

In Situ Raman Spectroelectrochemistry as a Tool for the Differentiation

Recommend Documents