Rev H. Chadli, F. Fergani, et al. Filling rate dependence on theoretical Raman spectra of C60 peapods CONFERENCE ARTICLE (International Symposium on Physics of Data Storage-1) FILLING RATE DEPENDENCE ON THEORETICAL RAMAN SPECTRA OF CARBON C60 PEAPODS H. Chadli, F. Fergani, S. A. Ait Abdelkader, AH Rahmani, B. Fakrach, A. Rahmani Laboratoire d’Etude des Matériaux Avancés et Applications (LEM2A), Université Moulay Ismail, FSM-ESTM-ESTK, BP 11201, Zitoune, 50000 Meknes, Morocco. Corresponding author : h.chadli@fpe.umi.ac.ma RECEIVED: 27 September 2016 / RECEIVED IN FINAL FORM: 21 December 2016 / ACCEPTED: 21 December 2016 Abstract : We use the spectral moments method in the framework of the bond-polarization theory to calculate nonresonant Raman spectra of C60 peapods as a function of the concentration of fullerenes inside the single wall carbon nanotubes. The evolution of the average Raman intensity ratios between Raman mode of C 60 molecules and nanotube as a function of the concentration of fullerenes has been analyzed and a general good agreement is found between calculations and measurements. Keywords : CARBON NANOTUBES, PEAPODS, RAMAN SPECTROSCOPY, SPECTRAL MOMENT METHOD, SIMULATION Introduction Single walled carbon nanotubes (SWNT) can encapsulate small and large molecules, including C60 fullerenes[1,2,3]. Those systems consist of SWNT in which fullerene molecules are inserted and called peapods. This hybrid system between fullerene and SWNT have generated a lot of interest for future electronic applications. Due to their special one- dimensional nanosized structure and their tunable electronic properties, peapods are considered important materials for several potential applications from high temperature superconductor[4] to memory elements[5]. From an experimental point of view, peapods structural analysis can be performed using transmission electron microscopy[1,2] or electron diffraction[3], X-ray scattering[6] or using Raman spectroscopy[7]. From a theoretical point of view, it was shown that C60 molecules inside SWNTs with a wide range of diameters from 1.25 nm to 2.71 nm can form ten different packing arrangements[8]. Some of these predicted phases have been experimentally observed[9]. Kuzmany et al. have performed a detailed Raman analysis of the concentration of C60 fullerenes in peapods[7]. The authors measured the scattering relative intensity from the fullerenes to that from the tube in order to obtain the relative C60 filling concentrations. The relative concentrations derived from the measurement of normalized intensity ratio for each mode of C60 are close (see Table 1 of Ref. [7]). In previous works [10,11], we calculated the nonresonant Raman spectra of C60 peapods for linear and zigzag configurations. In this paper, in order to Materials and Devices, Vol 1(1), 1221 (2016) – DOI: 10.23647/ca.md20161221 mailto:h.chadli@fpe.umi.ac.ma H. Chadli, F. Fergani, et al. Filling rate dependence on theoretical Raman spectra of C60 peapods improve the comparison with the experimental Raman data measured on peapod samples, we calculate the nonresonant Raman spectra of C60 peapods with large diameters (2.15-2.3 nm) in which the C60 molecules phase can take a double helix or a two molecules layers configuration depending on the diameter. The relative intensity ratio between a Raman mode of C60 and a mode of SWCNT, normalized by the same ratio calculated for a reference filling factor, gives a useful method to derive from Raman experiments the relative concentration of C60 in peapods samples prepared with the same batch that the reference filling factor sample. Model and method The structure of C60s inside SWNT depends primarily on tube diameter. We have already studied the different possible configurations of C60 inside SWCNT and found that the C60 molecules adopt a linear configuration for diameters below 1.45 nm and a zigzag configuration for diameters between 1.45 and 2.20 nm[9]. In this study, we consider SWNT in a diameter range from 2.20 to 2.30 nm in which the C60 molecules can take a double helix or a two molecules layers configuration depending on the nanotube diameter. These configurations of the C60 molecules inside the nanotubes are derived from energy minimization calculations previously described in previous works[10]. The geometrical optimizations of C60 inside SWNT are derived. We found that the C60 molecules adopt a double helix arrangement for SWNT's diameter in the range 2.16-2.23nm and layers of two molecules for tube diameter in the range 2.23-2.28nm. For the SWCNT, the C-C interactions are described using the force constant model introduced by Saito[12]. The interactions between carbon atoms for the C60 molecules are calculated using the force constants model described by Jishi and Dresselhaus[13]. The interactions between carbon atoms belonging to two different molecules (tube-tube, C60-C60 or tube-C60) are described by the familiar (12-6) Lennard-Jones potential, U LJ=4ε[(σ /r) 12 −(σ /r) 6 ] (1) Where r is the carbon atom-atom distance, ε=2.964 meV and σ=0.3407 nm are van der Waals parameters. Results Raman spectra calculations are based on the spectral moments method which has proved to be a powerful approach for determining the infrared, Raman, and inelastic neutron scattering spectra of harmonic systems (for mor details see Refs.[14]). The calculated ZZ-polarized Raman spectra of individual peapods for each configuration of the C60 molecules inside the tube are shown in figure 1 along with their corresponding unfilled nanotubes and the unoriented C60 molecule. In the following, we mainly discuss the polarized ZZ Raman spectra of unfilled zigzag SWNTs ((n, 0) with n = 17, 28 and 29), and for their corresponding peapods with linear [C60@(17,0)], double-helix [C60@(28,0)], and two-molecule layers phases [C60@(29,0)]. Concerning the high-wavenumber modes range, above 1200 cm-1, the main modes of C60 are located around 1219 (Hg(6)), 1403 Hg(7)), 1468 (Ag(2)), and 1582 cm-1 Hg(8)), in C60@(17,0), C60@(28,0), and C60@(29,0), independently of the configuration of C60 molecules inside the tube, these modes show a small upshift (1-2 cm-1) with respect to their position in the free C60 spectrum. As already found in linear and zigzag peapods (see figure 2 in Ref. [10]), We calculate that the Tangential-Like Modes (TLM) of SCNTs are almost not affected by the insertion of C60 molecules inside the nanotube. Concerning the main modes of (28,0) and (29,0) SWNT located at 168, 102 and 98 cm-1 are upshift to 174, 102 and 100 cm-1, respectively. These shift and splitting of Radial Breathing-Like Modes (RBLM) depend on one side, on the configuration of the C60 molecules inside SWNT, and the other side on the coupling effects related to the interactions between the C60 molecules and the nanotube host, and on inter- C60 interactions. Using the spectral moments method, we investigate the evolution of the average non resonant intensity ratios between Raman mode of C60 molecules and nanotube as a function of the concentration of C60s inside the tubes. Obviously, this non-resonant approach is not able to calculate the dependence of the absolute intensity of each mode with the excitation energies. Nevertheless, concerning the dependence of the Raman spectrum with the filling factor, the relative intensity ratio between a Raman mode of C60 and a mode of SWCNT, normalized by the same ratio calculated for a reference filling factor, gives a useful method to derive from Raman experiments the relative concentration of C60 in peapods samples prepared with the same batch than the reference filling factor sample. We calculate the dependence of the Raman spectrum at different filling factor. To facilitate the comparison with the experimental results, we performed an average of Materials and Devices, Vol 1(1), 1221 (2016) – DOI: 10.23647/ca.md20161221 H. Chadli, F. Fergani, et al. Filling rate dependence on theoretical Raman spectra of C60 peapods the Raman spectra over the peapod orientations with regards to the laboratory frame. Raman spectra are calculated in the VV configuration for unoriented peapod samples. For each filling factor, the integrated intensity ratios between a Raman mode of C60 and the RBLM (for C60 phonon modes: Hg(2), Ag(1), Hg(3) and Hg(4)) or G-mode (for C60 phonon modes: Hg(7) and Ag(2)) are calculated. These calculated intensity ratios are normalized with respect to the same intensity ratios calculated for the 60% filling factor sample. The relative concentrations are derived by this way for the typical C60@(28,0) (double helix chain of C60), C60@(29,0) (two molecule layers), C60@(13,13) (zigzag chain of C60) and C60@(10,10) (linear chain of C60) (Fig. 2a, Fig.2b, Fig 2c and Fig.2d, respectively). . Figure 1: The ZZ calculated Raman spectra of several phases of C60 inside SWNTs. Spectra are displayed in the Breathing mode (A), intermediate (B) and TLM (C) ranges. Figure 2: Calculated average intensity ratios, normalized on the 60% filling factor intensities, between modes of the C60 in peapods crystal and the RBLM (Hg(2),Ag(1),Hg(3), Hg(4)) or G-mode (Ag(2)) of nanotube. The considered configurations of C60 are linear (a), zigzag (b), double helix (c) and two-molecules layers (d) inside (10,10), (16,16), (28,0) and (29,0), respectively. Materials and Devices, Vol 1(1), 1221 (2016) – DOI: 10.23647/ca.md20161221 H. Chadli, F. Fergani, et al. Filling rate dependence on theoretical Raman spectra of C60 peapods As expected, for all the investigated peapods diameter, the relative concentrations calculated for each C60 mode are close. The comparison of the data of figure 2 shows that the relative concentrations calculated for infinite peapods increase when the peapod diameter increases. Indeed, for a filling factor around 20%, the relative concentration is close to 0.2, 0.4, 0.6 and 0.7 in C60@(10,10) (diameter of (10,10) is 1.35 nm), C60@(13,13) (diameter of (13,13) is 1.76 nm), C60@(28,0) (diameter of (28,0) is 2.19 nm) and C60@(29,0) (diameter of (29,0) is 2.27 nm), respectively. In the C60@(10,10), the average relative concentrations are found between 0.98 and 1.08 following the different C60 phonons modes, in good agreement with the experimental relative concentration evaluated around 1.13[7]. For concentrations around 20% (corresponding to the L43 sample (EELS concentration 25±10) in Ref. [7]), the average relative concentration is calculated around 0.22± 0.02 except for Ag(2) mode showing an overestimated concentration, in good agreement with the experimental relative concentration evaluated around 0.19. Conclusion In this work, we calculate the nonresonant Raman spectrum of isolated peapods with large diameter and using the spectral moments method. The dependence of the Raman spectrum with the C60 filling level have been discussed. The evolution of the average intensity ratios between Raman mode of C60 molecules and nanotube as a function of the concentration of C60s have been analyzed and a general good agreement is found between calculations and measurements. Acknowledgements: The work was supported by the CNRS-France/CNRST-Morocco agreement. REFERENCES 1. B. W. Smith, M. Monthioux, D. E. Luzzi, Nature 396, p 323 (1998). 2. H. Kataura, Y. Kumazawa, Y. Maniwa, I. Umezu I, S. Suzuki, Y. Ohtsuka, Y. Achiba Synth. Metals 121 p.1195 (2000) 3. H. Hirahara, K. Suenaga, S. Bandow, H. Kato, T. Okazaki, H. Shinohara, S. Iijima Phys. Rev. Lett. 85 p.5384 (2000) 4. R. F. Service, Science 292, p 45 (2001). 5. Y. K. Kwon, D. Tomanek, and S. Iijima, Phys. Rev. Lett. 82, p 1470 (1999). 6. H. Kataura, Y. Maniwa, M. Abe, A. Fujiwara, T. Kodama, K. Kikuchi, H. Imahori, Y. Misaki, S. Suzuki, Y. Achiba, Applied Physics A 74, p 349 (2002) 7. H. Kuzmany, R. Pfeiffer, C. Kramberger, T. Pichler, X. liu, M. Knupfer, J. Fink, H. Kataura, Y. Achiba, B.W. Smith, D.E. Luzzi Appl. Phys. A 76, p.449 (2003) 8. M. Hodak, L. A. Girifalco, Phys. ReV. B 68, p 085405 (2003). 9. A. Khlobystov, D. A. Britz, A. Ardavan, G. A. D. Briggs, Phys. Rev. Lett. 92, p 245507 (2004). 10. H. Chadli, A. Rahmani, K. Sbai, P. Hermet, S. Rols, and J.-L. Sauvajol, Phys. Rev. B 74, p 205412 (2006). 11. H. Chadli, F.Fergani, M.Bentaleb, B.Fakrach, K.Sbai, A.Rahmani, J.-L.Bantignies, J.-L. Sauvajol, PhysicaE 71, p 31 (2015). 12. R. Saito, T. Takeya, T. Kimura, G. Dresselhaus, and M. S. Dresselhaus Phys. Rev. B 57 p 4145 (1998). 13. R. A. Jishi, R. M. Mirie, and M. S. Dresselhaus, Phys. Rev. B 45, p 13685 (1992). 14. A. Rahmani, J.-L. Sauvajol, S. Rols, and C. Benoit, Phys. Rev. B 66, p 125404 (2002). Materials and Devices, Vol 1(1), 1221 (2016) – DOI: 10.23647/ca.md20161221 Introduction Model and method Results Conclusion