- Email: [email protected]
N-doped double-walled carbon nanotubes synthesized by chemical vapor deposition
Chemical Physics Letters 413 (2005) 300–305 www.elsevier.com/locate/cplett
N-doped double-walled carbon nanotubes synthesized by chemical vapor deposition Shin Young Kim a, Jinyoung Lee a, Chan Woong Na a, Jeunghee Park Kwanyong Seo b, Bongsoo Kim b a
a,*
,
Department of Chemistry, Korea University, Jochiwon 339-700, Republic of Korea b Department of Chemistry, KAIST, Daejeon 305-700, Republic of Korea Received 4 July 2005; in final form 19 July 2005 Available online 19 August 2005
Abstract High-purity N-doped double-walled carbon nanotubes (DWNTs) were synthesized by chemical vapor deposition, and investigated by synchrotron X-ray photoelectron spectroscopy (XPS). As the photon energy increases, the N content increases up to 3 at.%, indicating that the inner wall contains the higher N content than the outer wall. The graphite-like and pyridine-like N structures exist as about 1:1 ratio. The self-consistent charge-density-functional-based tight-binding calculation shows that the N-doping of the inner wall yields the more stable DWNT than that of the outer wall, which supports the XPS result. 2005 Elsevier B.V. All rights reserved.
1. Introduction Double-walled carbon nanotubes (DWNTs), as frontier between single-walled carbon nanotubes (SWNTs) and multiwalled carbon nanotubes (MWNTs), have recently attracted much attention because of their unique structural and electrical properties [1–7]. Various methods were developed for the high-yield synthesis of DWNTs [8–12], and the transport properties were also measured for the use of nanoscale electronic devices [6,7]. The control of electrical properties is very important in many applications of carbon nanotubes (CNTs). The doping of CNTs with other chemical elements (e.g., B or N) is a practical and feasible way to tailor their electrical properties. In fact, the N-doped CNTs show n-type behavior regardless of a tube chirality [13]. However, there are few works on the N-doped DWNTs, although arc-discharge synthesis of N-doped SWNTs has been reported [14,15]. *
Corresponding author. Fax: +2 3290 3973. E-mail address: [email protected] (J. Park).
0009-2614/$ - see front matter 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.cplett.2005.07.093
In the present work, we first report the synthesis of N-doped DWNTs using a catalytic chemical vapor deposition method. Their electronic structures were thoroughly investigated by employing synchrotron Xray photoelectron spectroscopy (XPS). The photon energy of XPS is variable in the range of 100–1300 eV. Since the escape depth of a photoelectron increases with its kinetic energy [16], the higher photon energy would provide more information for the inner parts of DWNTs. In order to explore the nature of band structure, we also obtained the valence band (VB) spectrum using low photon energy. Furthermore, we performed ab initio calculations on the N-doped DWNTs, (5,5)@(10,10), to explain the experimental results.
2. Experimental Ethanol solution of Fe(NO3)3 Æ 9H2O and (NH4)6Mo7O24 Æ 4H2O was blended with MgO powder and sonicated for 30 min. The molar ratio was MgO: Fe:Mo = 20:1:0.1. The mixture was baked at 120 C
S.Y. Kim et al. / Chemical Physics Letters 413 (2005) 300–305
duced. The outer diameter of DWNTs is in the range of 1.5–2 nm (Fig. 1c). Fig. 2 displays first-order Raman spectrum. It reveals the characteristic narrow G band at 1591 cm 1, which is originated from the Raman active A1g, E1g, and E2g axial vibrational modes of the graphite sheets [19,20]. The D band at 1340 cm 1 is originated from the level of disordered carbons and its intensity is about 1/10 of that of G band. It also shows a series of peaks in the range of 150–350 cm 1, which is unique to SWNTs and DWNTs arising from the radial breathing mode (RBM). Due to a cut-off filter, the peaks below 100 cm 1 were not detected. Typical axial vibrational modes and RBM confirm the production of DWNTs. The diameter is estimated by the formula that the RBM frequency (mRBM) is proportional to the inverse diameter (d): mRBM (cm 1) = 248/d (nm) [21]. The peaks at 311, 274, 259, 183, 165, and 150 cm 1 correspond to 0.80, 0.91, 0.96, 1.36, 1.50, and 1.65 nm. The diameter of 1.50 and 1.65 nm may correspond to the outer wall of DWNT
1591
Intensity (Arb.)
for 2 h and then ground to fine powders, forming the bimetallic catalyst embedded in MgO powders. The CH4/NH3/Ar mixture was flowed with the rate of 50/ 10/500 sccm, on these MgO-supported catalyst powders at 850 C for 10–30 min. The morphology and structure of the products were examined by scanning electron microscopy (SEM, Hitachi S-4300), field-emission transmission electron microscopy (TEM, FEI TECNAI G2), and Raman spectroscopy (Renishaw RM 1000) using the 514 nm line from an Ar ion laser. The XPS measurements were performed at the U7 (8A1) and U10 (3A1) beam lines of the pohang light source (PLS). These beam lines are designed to provide soft X-rays in the energy range of 50–1500 eV [17] and 10–1000 eV, respectively. The respective spectral resolving power (E/DE) of incident photons is 4000–5000 and 10 000–30 000. The experiment was performed in an ultra-high vacuum chamber with a base pressure 65 · 10 10. The analyzer was located at 55 from the surface normal. Theoretical calculations were performed for (5,5)@(10,10) N-doped DWNTs, using a self-consistent charge density-functional-based tight-binding (SCCDFTB) method [18]. The total energy and the optimized geometry were obtained for the conformers containing the different number of N atoms in two walls. Eight carbon layers along the tube axis were used for DWNTs. ˚ and the average bond length The inner diameter is 7.1 A ˚. is 1.42 A
301
150 160
274 259
183 311
3. Results and discussion Fig. 1a shows the SEM images for as-grown products. High-density bundles were grown completely covering the support. The diameter of the bundles is 10–20 nm. Fig. 1b corresponds to high-resolution TEM image showing that the bundles are mostly consisted of the DWNTs. A negligible amount of MWNTs was pro-
1340
200
300
1200
1500
1800
Raman Shift (cm-1) Fig. 2. Raman spectrum of DWNTs. The excitation wavelength is 514 nm line from Ar ion laser.
Fig. 1. (a) SEM image shows the nanotube bundles with a diameter of 10–20 nm. (b) HRTEM image reveals that the bundles are consisted mainly of DWNTs. (c) A magnified image showing the outer diameter in the range of 1.5–2 nm.
302
S.Y. Kim et al. / Chemical Physics Letters 413 (2005) 300–305
having the inner diameter of 0.80 and 0.91 nm, respectively. The diameter is in the same range as that observed from the HRTEM images. Fig. 3a shows the XPS survey scan spectrum using the photon energy 360, 630, 1000, and 1260 eV. It shows distinct C, N, and O 1s peaks. The respective probing depth of C 1s electrons is predicted to be about 1, 2, 4, and 5 nm [16]. The N concentration, defined as N/ (C + N) at.%, is estimated by the area ratio of the N 1s and C 1s peaks, taking into consideration of their relative sensitivities. As the photon energy increases from 630 to 1260 eV, the N% increases to 2.9%. The increase C1s
a
360 eV O1s
O1s
O1s
600
500
630 eV N1s 1.2%
1000 eV N1s 2.9%
1260 eV
400
300
200
100
0
Normalized Intensity
b
360 eV 630 eV 1000 eV
PC3 288
PC2 PC1
286
1260 eV
284
282
c
1000 eV
PN2
PN1
1260 eV 406
404
402
400
398
396
394
Binding Energy (eV) Fig. 3. (a) XPS survey scan spectrum. (b) The fine-scanned C 1s peak. The data points (open circles) are fitted by two Voigt functions PC1 (red dots), PC2 (blue dots), and PC3 (green dots). The sum of deconvoluted bands is marked by the lines. (c) The fine-scanned N 1s peak. The data points (open circles) are fitted by two Voigt functions PN1 (blue dots) and PN2 (red dots). The sum of deconvoluted bands is marked by the lines. (For interpretation of the references to color in this Figure, the reader is referred to the web version of this article.)
of the N% with the probing depth indicates the higher N content in the more inside parts of the tubes. The O peak would be originated mainly from the MgO support, with some contribution from the dangling bonds of graphite layers such as C@O. The fine-scanned C 1s spectrum is displayed in Fig. 3b. For all four photon energies, an asymmetric band is centered at 284.6 eV. The full width at half maximum (FWHM) increases from 0.8 to 1.5 eV, as the photon energy increases from 360 to 1260 eV. From the curve fitting by Voigt function, the band can be deconvoluted into three bands at 284.6 (PC1), 285.2 (PC2), and 286.1 eV (PC3). The strongest PC1 is assigned to the C atoms binding to C atoms (C–C), and the weaker PC2 corresponds to the C atoms binding to N atoms (C– N). We ascribe the weakest PC3 to the other binding configurations such as C@O or C„N that can be formed at the edge of graphite layers. This assignment is consistent with our recent work of the N-doped MWNTs [22]. The area % of PC1, PC2, and PC3 is listed in Table 1 for 4 photon energies. As the photon energy increases, the fraction of the N-bonded structure PC2 increases while that of the graphite structure PC1 decreases, which is consistent with the higher N content at the more inside parts of tubes. The bandwidth of the deconvoluted bands increases with the photon energy, implying the wider energy distribution of electronic states. Fig. 3c displays the N 1s XPS spectrum using 1000 and 1260 eV, showing a band centered at 400.5 eV. The asymmetric shape indicates an existence of at least two components, so the band has been deconvoluted into two bands PN1 at 398.3 eV and PN2 at 400.2 eV. The diverse electronic structures of N atoms, including the graphite-like, pyridine-like, pyrrolic, cross-linked sp3 structures, and intercalated N2, were identified by XPS [22–28]. We assigned the PN1 and PN2 to the pyridine-like and graphite-like structures, respectively, which is consistent with the previous work of the Ndoped MWNTs [22,24,25]. The area % of PN1 (pyridine-like) and PN2 (graphite-like) is about 1:1, as listed in Table 1. The XPS of undoped DWNTs was reported by our research group [29]. The FWHM of C 1s peak increases from 0.7 to 1 eV. The peak broadening at the higher photon energy would be due to the van der Waals interaction between the DWNTs in bundles. The present Ndoped DWNTs show a more significant dependence on the photon energy than the undoped DWNTs. The N doping results in the lower degree of crystalline perfection of graphite sheets, which contributes in dispersing the energy distribution of electronic states. Compared to the MWNTs containing the same N content [22], the DWNTs have the higher fraction of pyridine-like structures. Due to the smaller diameter of DWNTs, the formation of pyridine-like structure would be more
S.Y. Kim et al. / Chemical Physics Letters 413 (2005) 300–305
303
Table 1 Area % of deconvoluted bands from the XPS C 1s and N 1s peaks of the N-doped DWNT bundles Photon energy (eV)
PC1 (C–C)
PC2 (C–N)
Position (eV)
FWHM (eV)
Area %
Position (eV)
FWHM (eV)
Area %
Position (eV)
FWHM (eV)
Area %
360 625 1000 1260
284.7 284.6 284.6 284.6
0.49 0.54 0.76 0.92
58 58 48 47
285.2 285.1 285.2 285.3
0.76 0.79 0.79 0.92
20 22 27 32
286.1 286.1 286.1 286.2
1.7 1.6 1.6 1.7
22 20 25 21
1000 1260
398.3 398.3
3.1 3.4
42 46
2.8 3.0
58 54
PN1 (pyridine-like)
PN2 (graphite-like) 400.1 400.2
favorable by releasing a strain of graphite sheets. This will be explained further using the following SCCDFTB calculations. The VB spectrum measured using the photon energy 133 eV is displayed in Fig. 4. The spectrum of the Ndoped DWNTs has been compared with that of the undoped DWNTs. The zero energy is chosen at the Fermi level Ef, which is the threshold of the emission spectrum. The spectrum is normalized using the intensity at 13 eV. The intensity of the region between 2 and 12 eV is noticeably higher for the N-doped DWNTs than the undoped DWNTs. This enhanced region is separated into two parts; region A at 2–6.5 eV and region B at 6.5– 12 eV, which attributes to C 2p and N 2p electrons associated with p bonds in graphite structure, and to C 2p and N 2p electrons associated with bonds, respectively [30,31]. Thus the presence of the N-binding structure increases the intensity of the A and B regions. Our research group had already reported the relative energy of three (5,5)@(10,10) DWNT conformers in which the N structure is graphite-like and the total number of N atoms is 6 [32]. The energy of the conformers G1, G2, and G3 is listed in Table 2. Herein, we calculate
N-doped DWNTs
Normalized Intensity
Undoped DWNTs
A
B
12
9
PC3 (defects)
6
3
0
Binding Energy (eV) Fig. 4. VB spectrum of XPS using photon energy 133 eV for N-doped and undoped DWNTs. The spectrum consists of two regions; region A at 2–6.5 eV and region B at 6.5–12 eV.
the relative energy of the DWNT conformers containing the pyridine-like N atoms. Two conformers P1 and P2, in which three N atoms replace the C atoms of the outer and inner wall, respectively, are displayed in Fig. 5. The 4% N-doping of the inner wall results in the more stable conformer ( 0.76 eV) than the 2% N-doping of the outer wall, as listed in Table 2. In order to examine the effect of defects on the stability of DWNTs, we calculate the relative energy of the conformers D1 and D2 with one C atom lacking in the outer and inner wall, respectively. The defect in the inner wall induces more stable conformer than that in the outer wall by 1.12 eV. The calculation shows that the N-doping of the inner wall is more favorable than that of the outer wall for both graphite-like and pyridine-like N structures. The formation of N-binding structure would release the strains of curved graphite layers, more efficiently for the inner wall than the outer wall. The defects in the inner wall are also favorable by the same reason. The incorporation of N atoms preferentially in the inner wall may be thus thermodynamically driven during the growth. For the pyridine-like N structures, the energy decrease corresponding to 2% N increase of the inner wall is more significant than the case of graphite-like N structure. Due to the deficiency of C atoms, the pyridine-like N structure would release the strain better than the graphite-like N structure. Such stain-releasing effect becomes larger for the smaller diameter, which explains the higher fraction of the pyridine-like N structure compared to the MWNTs. In summary, we synthesized high-purity N-doped DWNTs on the MgO-supported Fe–Mo catalysts using the CVD of CH4/NH3 mixture at 850 C. The DWNTs form the bundles with a diameter of 10–20 nm. The outer diameter is in the range of 1.5–2 nm. As the photon energy increases, the N content increases up to 3%, indicating a higher N concentration at the inner wall. The deconvolution of fine-scanned N 1s peaks shows that the pyridine-like and graphite-like N structures exist as 1:1 ratio. The XPS valence band analysis reveals that the N-doped DWNTs have an enhanced 2p-p and 2pr electron density due to the doped N atoms. We calculate the stability of N-doped DWNT (5,5)@(10,10)
304
S.Y. Kim et al. / Chemical Physics Letters 413 (2005) 300–305
Table 2 Relative energy of the N-doped DWNT conformers calculated using SCC-DFTB Conformer
N content (%)
Relative energy (eV)
Inner
Outer
G1 G2 G3
1 4 5
3 2 1
0
P1 P2
0 4
2 0
0
0.31 0.37
The total number of N atoms is 6. We only vary the number between outer and inner wall ([31]) The number of N atoms replacing C atoms is 3
0.76
C deficiency (%)
D1 D2
Inner
Outer
0 1
0.6 0
0
The number of C deficiency is 1 1.12
Fig. 5. A (5,5)@(10,10) DWNT doped with pyridine-like N atoms: (a) 2% in the outer wall and (b) 4% inner wall. The energy of (b) is 0.76 eV lower than that of (a).
conformers using SCC-DFTB method. The N doping of the inner wall than the outer wall is thermodynamically favorable, which is more significant for the pyridine-like N structure compared to the graphite-like N structure.
Acknowledgements This work was supported by KOSEF (Project No. R14-2003-033-01003-0; R02-2004-000-10025-0) and KRF (2003-015-C00265). BK thanks NT-IT Fusion Strategy of Advanced Technology for financial support. SEM and TEM analysis was performed at the Korea Basic Science Institute (KBSI). Experiments at PLS were supported in part by MOST and POSTECH.
References [1] R. Saito, G. Dresselhaus, M.S. Dresselhaus, J. Appl. Phys. 73 (1993) 494. [2] K. Tanaka, H. Aoki, H. Ago, T. Yamabe, K. Okahara, Carbon 35 (1997) 121.
[3] M.B. Nardelli, C. Brabec, A. Marti, C. Roland, J. Bernholc, Phys. Rev. Lett. 80 (1998) 313. [4] Y.-K. Kwon, D. Toma´nek, Phys. Rev. B 58 (1998) R16001. [5] R. Saito, R. Matsuo, T. Kimura, G. Dresselhaus, M.S. Dresselhaus, Chem. Phys. Lett. 348 (2001) 187. [6] M. Kociak, K. Suenage, K. Hirahara, Y. Saito, T. Nakahira, S. Iijima, Phys. Rev. Lett. 89 (2002) 155501. [7] T. Shimada, T. Sugai, Y. Ohno, S. Kishimoto, T. Mizutani, H. Yoshida, T. Okazaki, H. Shinohara, Appl. Phys. Lett. 84 (2004) 2412. [8] T. Sugai, H. Yoshida, T. Shimada, T. Okazaki, H. Shinohara, NanoLetters 3 (2003) 769. [9] S. Bandow, M. Takizawa, K. Hirahara, M. Yudasaka, S. Iijima, Chem. Phys. Lett. 337 (2001) 48. [10] S.C. Lyu, T.J. Lee, C.W. Yang, C.J. Lee, Chem. Commun. (2003) 1404. [11] E. Flahaut, R. Bacsa, A. Peigney, C. Laurent, Chem. Commun. (2003) 1442. [12] J. Wei, B. Jiang, D. Wu, B. Wei, J. Phys. Chem. B 108 (2004) 8844. [13] R. Czerw, M. Terrones, J.-C. Charlier, X. Blase, B. Foley, R. Kamalakaran, N. Grobert, H. Terrones, D. Tekleab, P.M. Ajayan, W. Blau, M. Ru¨hle, D.L. Carroll, NanoLetters 1 (2001) 457. [14] R. Droppa Jr., P. Hammer, A.C.M. Carvalho, M.C. dos Santos, F. Alvaretz, J. Non-Cryst. Solid 299 (2002) 874. [15] M. Glerup, J. Steinmetz, D. Samaille, O. Ste´phen, S. Enouz, A. Loiseau, S. Roth, P. Bernier, Chem. Phys. Lett. 387 (2004) 193. [16] B.K. Teo, EXAFS: Basic Principles and Data Analysis, SpringerVerlag, Berlin, 1986, p. 92. [17] M.-K. Lee, H.J. Shin, Nucl. Instrum. Methods Phys. Rev. A 467468 (2001) 508. [18] M. Elstner, D. Porezag, G. Jungnickel, J. Elsner, M. Haugk, Th. Frauenheim, S. Suhai, G. Seifert, Phys. Rev. B 58 (1998) 7260. [19] A. Kasuya, Y. Sasaki, K. Tohji, Y. Nishina, Phys. Rev. Lett. 78 (1997) 4434. [20] A.M. Rao, P.C. Eklund, S. Bandow, A. Thess, R.E. Smalley, Nature 388 (1997) 257. [21] A. Jorio, R. Saito, J.H. Hafner, C.M. Lieber, M. Hunter, T. McClure, G. Dresselhaus, M.S. Dresselhaus, Phys. Rev. Lett. 86 (2001) 1118. [22] H.C. Choi, J. Park, B. Kim, J. Phys. Chem. B 109 (2005) 4333. [23] R. Sen, B.C. Satishkumar, A. Govindaraj, K.R. Harikumar, M.K. Renganathan, C.N.R. Rao, J. Mater. Chem. 7 (1997) 2335. [24] M. Terrones, P. Redlich, N. Grobert, S. Trasobares, W.K. Hsu, H. Terrones, Y.Q. Zhu, J.P. Hare, C.L. Reeves, A.K. Cheetham,
S.Y. Kim et al. / Chemical Physics Letters 413 (2005) 300–305 M. Ru¨hle, H.W. Kroto, D.R.M. Walton, Adv. Mater. 11 (1999) 655. [25] M. Nath, B.C. Satishkumar, A. Govindaraj, C.P. Vinod, C.N.R. Rao, Chem. Phys. Lett. 322 (2000) 333. [26] X. Wang, Y. Liu, D. Zhu, L. Zhang, H. Ma, N. Yao, B. Zhang, J. Phys. Chem. B 106 (2002) 2186. [27] A.G. Kudashov, A.V. Okotrub, L.G. Bulusheva, I.P. Asanov, Yu.V. Shubin, N.F. Yudanov, L.I. Yudanova, V.S. Danilovich, O.G. Abrosimov, J. Phys. Chem. B 108 (2004) 9048.
305
[28] J.W. Jang, C.E. Lee, S.C. Lyu, T.J. Lee, C.J. Lee, Appl. Phys. Lett. 84 (2004) 2877. [29] H.C. Choi, S.Y. Kim, W.S. Jang, S.Y. Bae, J. Park, K.L. Kim, K. Kim, Chem. Phys. Lett. 399 (2004) 255. [30] S. Souto, M. Pickholz, M.C. dos Santos, F. Alvarez, Phys. Rev. B 57 (1998) 2536. [31] Z.Y. Chen, J.P. Zhao, T. Yano, T. Ooie, J. Appl. Phys. 92 (2002) 281. [32] H.C. Choi, S.Y. Bae, J. Park, K. Seo, C. Kim, B. Kim, H.J. Song, H.-J. Shin, Appl. Phys. Lett. 85 (2004) 5742.
N-doped double-walled carbon nanotubes synthesized by chemical vapor deposition Shin Young Kim a, Jinyoung Lee a, Chan Woong Na a, Jeunghee Park Kwanyong Seo b, Bongsoo Kim b a
a,*
,
Department of Chemistry, Korea University, Jochiwon 339-700, Republic of Korea b Department of Chemistry, KAIST, Daejeon 305-700, Republic of Korea Received 4 July 2005; in final form 19 July 2005 Available online 19 August 2005
Abstract High-purity N-doped double-walled carbon nanotubes (DWNTs) were synthesized by chemical vapor deposition, and investigated by synchrotron X-ray photoelectron spectroscopy (XPS). As the photon energy increases, the N content increases up to 3 at.%, indicating that the inner wall contains the higher N content than the outer wall. The graphite-like and pyridine-like N structures exist as about 1:1 ratio. The self-consistent charge-density-functional-based tight-binding calculation shows that the N-doping of the inner wall yields the more stable DWNT than that of the outer wall, which supports the XPS result. 2005 Elsevier B.V. All rights reserved.
1. Introduction Double-walled carbon nanotubes (DWNTs), as frontier between single-walled carbon nanotubes (SWNTs) and multiwalled carbon nanotubes (MWNTs), have recently attracted much attention because of their unique structural and electrical properties [1–7]. Various methods were developed for the high-yield synthesis of DWNTs [8–12], and the transport properties were also measured for the use of nanoscale electronic devices [6,7]. The control of electrical properties is very important in many applications of carbon nanotubes (CNTs). The doping of CNTs with other chemical elements (e.g., B or N) is a practical and feasible way to tailor their electrical properties. In fact, the N-doped CNTs show n-type behavior regardless of a tube chirality [13]. However, there are few works on the N-doped DWNTs, although arc-discharge synthesis of N-doped SWNTs has been reported [14,15]. *
Corresponding author. Fax: +2 3290 3973. E-mail address: [email protected] (J. Park).
0009-2614/$ - see front matter 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.cplett.2005.07.093
In the present work, we first report the synthesis of N-doped DWNTs using a catalytic chemical vapor deposition method. Their electronic structures were thoroughly investigated by employing synchrotron Xray photoelectron spectroscopy (XPS). The photon energy of XPS is variable in the range of 100–1300 eV. Since the escape depth of a photoelectron increases with its kinetic energy [16], the higher photon energy would provide more information for the inner parts of DWNTs. In order to explore the nature of band structure, we also obtained the valence band (VB) spectrum using low photon energy. Furthermore, we performed ab initio calculations on the N-doped DWNTs, (5,5)@(10,10), to explain the experimental results.
2. Experimental Ethanol solution of Fe(NO3)3 Æ 9H2O and (NH4)6Mo7O24 Æ 4H2O was blended with MgO powder and sonicated for 30 min. The molar ratio was MgO: Fe:Mo = 20:1:0.1. The mixture was baked at 120 C
S.Y. Kim et al. / Chemical Physics Letters 413 (2005) 300–305
duced. The outer diameter of DWNTs is in the range of 1.5–2 nm (Fig. 1c). Fig. 2 displays first-order Raman spectrum. It reveals the characteristic narrow G band at 1591 cm 1, which is originated from the Raman active A1g, E1g, and E2g axial vibrational modes of the graphite sheets [19,20]. The D band at 1340 cm 1 is originated from the level of disordered carbons and its intensity is about 1/10 of that of G band. It also shows a series of peaks in the range of 150–350 cm 1, which is unique to SWNTs and DWNTs arising from the radial breathing mode (RBM). Due to a cut-off filter, the peaks below 100 cm 1 were not detected. Typical axial vibrational modes and RBM confirm the production of DWNTs. The diameter is estimated by the formula that the RBM frequency (mRBM) is proportional to the inverse diameter (d): mRBM (cm 1) = 248/d (nm) [21]. The peaks at 311, 274, 259, 183, 165, and 150 cm 1 correspond to 0.80, 0.91, 0.96, 1.36, 1.50, and 1.65 nm. The diameter of 1.50 and 1.65 nm may correspond to the outer wall of DWNT
1591
Intensity (Arb.)
for 2 h and then ground to fine powders, forming the bimetallic catalyst embedded in MgO powders. The CH4/NH3/Ar mixture was flowed with the rate of 50/ 10/500 sccm, on these MgO-supported catalyst powders at 850 C for 10–30 min. The morphology and structure of the products were examined by scanning electron microscopy (SEM, Hitachi S-4300), field-emission transmission electron microscopy (TEM, FEI TECNAI G2), and Raman spectroscopy (Renishaw RM 1000) using the 514 nm line from an Ar ion laser. The XPS measurements were performed at the U7 (8A1) and U10 (3A1) beam lines of the pohang light source (PLS). These beam lines are designed to provide soft X-rays in the energy range of 50–1500 eV [17] and 10–1000 eV, respectively. The respective spectral resolving power (E/DE) of incident photons is 4000–5000 and 10 000–30 000. The experiment was performed in an ultra-high vacuum chamber with a base pressure 65 · 10 10. The analyzer was located at 55 from the surface normal. Theoretical calculations were performed for (5,5)@(10,10) N-doped DWNTs, using a self-consistent charge density-functional-based tight-binding (SCCDFTB) method [18]. The total energy and the optimized geometry were obtained for the conformers containing the different number of N atoms in two walls. Eight carbon layers along the tube axis were used for DWNTs. ˚ and the average bond length The inner diameter is 7.1 A ˚. is 1.42 A
301
150 160
274 259
183 311
3. Results and discussion Fig. 1a shows the SEM images for as-grown products. High-density bundles were grown completely covering the support. The diameter of the bundles is 10–20 nm. Fig. 1b corresponds to high-resolution TEM image showing that the bundles are mostly consisted of the DWNTs. A negligible amount of MWNTs was pro-
1340
200
300
1200
1500
1800
Raman Shift (cm-1) Fig. 2. Raman spectrum of DWNTs. The excitation wavelength is 514 nm line from Ar ion laser.
Fig. 1. (a) SEM image shows the nanotube bundles with a diameter of 10–20 nm. (b) HRTEM image reveals that the bundles are consisted mainly of DWNTs. (c) A magnified image showing the outer diameter in the range of 1.5–2 nm.
302
S.Y. Kim et al. / Chemical Physics Letters 413 (2005) 300–305
having the inner diameter of 0.80 and 0.91 nm, respectively. The diameter is in the same range as that observed from the HRTEM images. Fig. 3a shows the XPS survey scan spectrum using the photon energy 360, 630, 1000, and 1260 eV. It shows distinct C, N, and O 1s peaks. The respective probing depth of C 1s electrons is predicted to be about 1, 2, 4, and 5 nm [16]. The N concentration, defined as N/ (C + N) at.%, is estimated by the area ratio of the N 1s and C 1s peaks, taking into consideration of their relative sensitivities. As the photon energy increases from 630 to 1260 eV, the N% increases to 2.9%. The increase C1s
a
360 eV O1s
O1s
O1s
600
500
630 eV N1s 1.2%
1000 eV N1s 2.9%
1260 eV
400
300
200
100
0
Normalized Intensity
b
360 eV 630 eV 1000 eV
PC3 288
PC2 PC1
286
1260 eV
284
282
c
1000 eV
PN2
PN1
1260 eV 406
404
402
400
398
396
394
Binding Energy (eV) Fig. 3. (a) XPS survey scan spectrum. (b) The fine-scanned C 1s peak. The data points (open circles) are fitted by two Voigt functions PC1 (red dots), PC2 (blue dots), and PC3 (green dots). The sum of deconvoluted bands is marked by the lines. (c) The fine-scanned N 1s peak. The data points (open circles) are fitted by two Voigt functions PN1 (blue dots) and PN2 (red dots). The sum of deconvoluted bands is marked by the lines. (For interpretation of the references to color in this Figure, the reader is referred to the web version of this article.)
of the N% with the probing depth indicates the higher N content in the more inside parts of the tubes. The O peak would be originated mainly from the MgO support, with some contribution from the dangling bonds of graphite layers such as C@O. The fine-scanned C 1s spectrum is displayed in Fig. 3b. For all four photon energies, an asymmetric band is centered at 284.6 eV. The full width at half maximum (FWHM) increases from 0.8 to 1.5 eV, as the photon energy increases from 360 to 1260 eV. From the curve fitting by Voigt function, the band can be deconvoluted into three bands at 284.6 (PC1), 285.2 (PC2), and 286.1 eV (PC3). The strongest PC1 is assigned to the C atoms binding to C atoms (C–C), and the weaker PC2 corresponds to the C atoms binding to N atoms (C– N). We ascribe the weakest PC3 to the other binding configurations such as C@O or C„N that can be formed at the edge of graphite layers. This assignment is consistent with our recent work of the N-doped MWNTs [22]. The area % of PC1, PC2, and PC3 is listed in Table 1 for 4 photon energies. As the photon energy increases, the fraction of the N-bonded structure PC2 increases while that of the graphite structure PC1 decreases, which is consistent with the higher N content at the more inside parts of tubes. The bandwidth of the deconvoluted bands increases with the photon energy, implying the wider energy distribution of electronic states. Fig. 3c displays the N 1s XPS spectrum using 1000 and 1260 eV, showing a band centered at 400.5 eV. The asymmetric shape indicates an existence of at least two components, so the band has been deconvoluted into two bands PN1 at 398.3 eV and PN2 at 400.2 eV. The diverse electronic structures of N atoms, including the graphite-like, pyridine-like, pyrrolic, cross-linked sp3 structures, and intercalated N2, were identified by XPS [22–28]. We assigned the PN1 and PN2 to the pyridine-like and graphite-like structures, respectively, which is consistent with the previous work of the Ndoped MWNTs [22,24,25]. The area % of PN1 (pyridine-like) and PN2 (graphite-like) is about 1:1, as listed in Table 1. The XPS of undoped DWNTs was reported by our research group [29]. The FWHM of C 1s peak increases from 0.7 to 1 eV. The peak broadening at the higher photon energy would be due to the van der Waals interaction between the DWNTs in bundles. The present Ndoped DWNTs show a more significant dependence on the photon energy than the undoped DWNTs. The N doping results in the lower degree of crystalline perfection of graphite sheets, which contributes in dispersing the energy distribution of electronic states. Compared to the MWNTs containing the same N content [22], the DWNTs have the higher fraction of pyridine-like structures. Due to the smaller diameter of DWNTs, the formation of pyridine-like structure would be more
S.Y. Kim et al. / Chemical Physics Letters 413 (2005) 300–305
303
Table 1 Area % of deconvoluted bands from the XPS C 1s and N 1s peaks of the N-doped DWNT bundles Photon energy (eV)
PC1 (C–C)
PC2 (C–N)
Position (eV)
FWHM (eV)
Area %
Position (eV)
FWHM (eV)
Area %
Position (eV)
FWHM (eV)
Area %
360 625 1000 1260
284.7 284.6 284.6 284.6
0.49 0.54 0.76 0.92
58 58 48 47
285.2 285.1 285.2 285.3
0.76 0.79 0.79 0.92
20 22 27 32
286.1 286.1 286.1 286.2
1.7 1.6 1.6 1.7
22 20 25 21
1000 1260
398.3 398.3
3.1 3.4
42 46
2.8 3.0
58 54
PN1 (pyridine-like)
PN2 (graphite-like) 400.1 400.2
favorable by releasing a strain of graphite sheets. This will be explained further using the following SCCDFTB calculations. The VB spectrum measured using the photon energy 133 eV is displayed in Fig. 4. The spectrum of the Ndoped DWNTs has been compared with that of the undoped DWNTs. The zero energy is chosen at the Fermi level Ef, which is the threshold of the emission spectrum. The spectrum is normalized using the intensity at 13 eV. The intensity of the region between 2 and 12 eV is noticeably higher for the N-doped DWNTs than the undoped DWNTs. This enhanced region is separated into two parts; region A at 2–6.5 eV and region B at 6.5– 12 eV, which attributes to C 2p and N 2p electrons associated with p bonds in graphite structure, and to C 2p and N 2p electrons associated with bonds, respectively [30,31]. Thus the presence of the N-binding structure increases the intensity of the A and B regions. Our research group had already reported the relative energy of three (5,5)@(10,10) DWNT conformers in which the N structure is graphite-like and the total number of N atoms is 6 [32]. The energy of the conformers G1, G2, and G3 is listed in Table 2. Herein, we calculate
N-doped DWNTs
Normalized Intensity
Undoped DWNTs
A
B
12
9
PC3 (defects)
6
3
0
Binding Energy (eV) Fig. 4. VB spectrum of XPS using photon energy 133 eV for N-doped and undoped DWNTs. The spectrum consists of two regions; region A at 2–6.5 eV and region B at 6.5–12 eV.
the relative energy of the DWNT conformers containing the pyridine-like N atoms. Two conformers P1 and P2, in which three N atoms replace the C atoms of the outer and inner wall, respectively, are displayed in Fig. 5. The 4% N-doping of the inner wall results in the more stable conformer ( 0.76 eV) than the 2% N-doping of the outer wall, as listed in Table 2. In order to examine the effect of defects on the stability of DWNTs, we calculate the relative energy of the conformers D1 and D2 with one C atom lacking in the outer and inner wall, respectively. The defect in the inner wall induces more stable conformer than that in the outer wall by 1.12 eV. The calculation shows that the N-doping of the inner wall is more favorable than that of the outer wall for both graphite-like and pyridine-like N structures. The formation of N-binding structure would release the strains of curved graphite layers, more efficiently for the inner wall than the outer wall. The defects in the inner wall are also favorable by the same reason. The incorporation of N atoms preferentially in the inner wall may be thus thermodynamically driven during the growth. For the pyridine-like N structures, the energy decrease corresponding to 2% N increase of the inner wall is more significant than the case of graphite-like N structure. Due to the deficiency of C atoms, the pyridine-like N structure would release the strain better than the graphite-like N structure. Such stain-releasing effect becomes larger for the smaller diameter, which explains the higher fraction of the pyridine-like N structure compared to the MWNTs. In summary, we synthesized high-purity N-doped DWNTs on the MgO-supported Fe–Mo catalysts using the CVD of CH4/NH3 mixture at 850 C. The DWNTs form the bundles with a diameter of 10–20 nm. The outer diameter is in the range of 1.5–2 nm. As the photon energy increases, the N content increases up to 3%, indicating a higher N concentration at the inner wall. The deconvolution of fine-scanned N 1s peaks shows that the pyridine-like and graphite-like N structures exist as 1:1 ratio. The XPS valence band analysis reveals that the N-doped DWNTs have an enhanced 2p-p and 2pr electron density due to the doped N atoms. We calculate the stability of N-doped DWNT (5,5)@(10,10)
304
S.Y. Kim et al. / Chemical Physics Letters 413 (2005) 300–305
Table 2 Relative energy of the N-doped DWNT conformers calculated using SCC-DFTB Conformer
N content (%)
Relative energy (eV)
Inner
Outer
G1 G2 G3
1 4 5
3 2 1
0
P1 P2
0 4
2 0
0
0.31 0.37
The total number of N atoms is 6. We only vary the number between outer and inner wall ([31]) The number of N atoms replacing C atoms is 3
0.76
C deficiency (%)
D1 D2
Inner
Outer
0 1
0.6 0
0
The number of C deficiency is 1 1.12
Fig. 5. A (5,5)@(10,10) DWNT doped with pyridine-like N atoms: (a) 2% in the outer wall and (b) 4% inner wall. The energy of (b) is 0.76 eV lower than that of (a).
conformers using SCC-DFTB method. The N doping of the inner wall than the outer wall is thermodynamically favorable, which is more significant for the pyridine-like N structure compared to the graphite-like N structure.
Acknowledgements This work was supported by KOSEF (Project No. R14-2003-033-01003-0; R02-2004-000-10025-0) and KRF (2003-015-C00265). BK thanks NT-IT Fusion Strategy of Advanced Technology for financial support. SEM and TEM analysis was performed at the Korea Basic Science Institute (KBSI). Experiments at PLS were supported in part by MOST and POSTECH.
References [1] R. Saito, G. Dresselhaus, M.S. Dresselhaus, J. Appl. Phys. 73 (1993) 494. [2] K. Tanaka, H. Aoki, H. Ago, T. Yamabe, K. Okahara, Carbon 35 (1997) 121.
[3] M.B. Nardelli, C. Brabec, A. Marti, C. Roland, J. Bernholc, Phys. Rev. Lett. 80 (1998) 313. [4] Y.-K. Kwon, D. Toma´nek, Phys. Rev. B 58 (1998) R16001. [5] R. Saito, R. Matsuo, T. Kimura, G. Dresselhaus, M.S. Dresselhaus, Chem. Phys. Lett. 348 (2001) 187. [6] M. Kociak, K. Suenage, K. Hirahara, Y. Saito, T. Nakahira, S. Iijima, Phys. Rev. Lett. 89 (2002) 155501. [7] T. Shimada, T. Sugai, Y. Ohno, S. Kishimoto, T. Mizutani, H. Yoshida, T. Okazaki, H. Shinohara, Appl. Phys. Lett. 84 (2004) 2412. [8] T. Sugai, H. Yoshida, T. Shimada, T. Okazaki, H. Shinohara, NanoLetters 3 (2003) 769. [9] S. Bandow, M. Takizawa, K. Hirahara, M. Yudasaka, S. Iijima, Chem. Phys. Lett. 337 (2001) 48. [10] S.C. Lyu, T.J. Lee, C.W. Yang, C.J. Lee, Chem. Commun. (2003) 1404. [11] E. Flahaut, R. Bacsa, A. Peigney, C. Laurent, Chem. Commun. (2003) 1442. [12] J. Wei, B. Jiang, D. Wu, B. Wei, J. Phys. Chem. B 108 (2004) 8844. [13] R. Czerw, M. Terrones, J.-C. Charlier, X. Blase, B. Foley, R. Kamalakaran, N. Grobert, H. Terrones, D. Tekleab, P.M. Ajayan, W. Blau, M. Ru¨hle, D.L. Carroll, NanoLetters 1 (2001) 457. [14] R. Droppa Jr., P. Hammer, A.C.M. Carvalho, M.C. dos Santos, F. Alvaretz, J. Non-Cryst. Solid 299 (2002) 874. [15] M. Glerup, J. Steinmetz, D. Samaille, O. Ste´phen, S. Enouz, A. Loiseau, S. Roth, P. Bernier, Chem. Phys. Lett. 387 (2004) 193. [16] B.K. Teo, EXAFS: Basic Principles and Data Analysis, SpringerVerlag, Berlin, 1986, p. 92. [17] M.-K. Lee, H.J. Shin, Nucl. Instrum. Methods Phys. Rev. A 467468 (2001) 508. [18] M. Elstner, D. Porezag, G. Jungnickel, J. Elsner, M. Haugk, Th. Frauenheim, S. Suhai, G. Seifert, Phys. Rev. B 58 (1998) 7260. [19] A. Kasuya, Y. Sasaki, K. Tohji, Y. Nishina, Phys. Rev. Lett. 78 (1997) 4434. [20] A.M. Rao, P.C. Eklund, S. Bandow, A. Thess, R.E. Smalley, Nature 388 (1997) 257. [21] A. Jorio, R. Saito, J.H. Hafner, C.M. Lieber, M. Hunter, T. McClure, G. Dresselhaus, M.S. Dresselhaus, Phys. Rev. Lett. 86 (2001) 1118. [22] H.C. Choi, J. Park, B. Kim, J. Phys. Chem. B 109 (2005) 4333. [23] R. Sen, B.C. Satishkumar, A. Govindaraj, K.R. Harikumar, M.K. Renganathan, C.N.R. Rao, J. Mater. Chem. 7 (1997) 2335. [24] M. Terrones, P. Redlich, N. Grobert, S. Trasobares, W.K. Hsu, H. Terrones, Y.Q. Zhu, J.P. Hare, C.L. Reeves, A.K. Cheetham,
S.Y. Kim et al. / Chemical Physics Letters 413 (2005) 300–305 M. Ru¨hle, H.W. Kroto, D.R.M. Walton, Adv. Mater. 11 (1999) 655. [25] M. Nath, B.C. Satishkumar, A. Govindaraj, C.P. Vinod, C.N.R. Rao, Chem. Phys. Lett. 322 (2000) 333. [26] X. Wang, Y. Liu, D. Zhu, L. Zhang, H. Ma, N. Yao, B. Zhang, J. Phys. Chem. B 106 (2002) 2186. [27] A.G. Kudashov, A.V. Okotrub, L.G. Bulusheva, I.P. Asanov, Yu.V. Shubin, N.F. Yudanov, L.I. Yudanova, V.S. Danilovich, O.G. Abrosimov, J. Phys. Chem. B 108 (2004) 9048.
305
[28] J.W. Jang, C.E. Lee, S.C. Lyu, T.J. Lee, C.J. Lee, Appl. Phys. Lett. 84 (2004) 2877. [29] H.C. Choi, S.Y. Kim, W.S. Jang, S.Y. Bae, J. Park, K.L. Kim, K. Kim, Chem. Phys. Lett. 399 (2004) 255. [30] S. Souto, M. Pickholz, M.C. dos Santos, F. Alvarez, Phys. Rev. B 57 (1998) 2536. [31] Z.Y. Chen, J.P. Zhao, T. Yano, T. Ooie, J. Appl. Phys. 92 (2002) 281. [32] H.C. Choi, S.Y. Bae, J. Park, K. Seo, C. Kim, B. Kim, H.J. Song, H.-J. Shin, Appl. Phys. Lett. 85 (2004) 5742.