J. Phys. Chem. B 2005, 109, 20403-20406
20403
Electronic Structure and Field Emission of Multiwalled Carbon Nanotubes Depending on Growth Temperature Sang Won Yoon, Shin Young Kim, and Jeunghee Park* Department of Chemistry, Korea UniVersity, Jochiwon 339-700, Korea
Chan Jun Park and Cheol Jin Lee* Department of Electronics Engineering, Korea UniVersity, Seoul 136-701, Korea ReceiVed: August 17, 2005; In Final Form: September 4, 2005
The electronic structure of multiwalled carbon nanotubes (CNTs) has been investigated, depending on the growth temperature, using synchrotron X-ray photoelectron spectroscopy (XPS) and field emission measurements. The vertically aligned CNTs are grown via pyrolysis of ferrocene and acetylene in a broad temperature range 600-1000 °C. The CNTs have a cylindrical structure with a uniform diameter of 20 nm. As growth temperature increases, due to an improved crystallinity of the graphitic sheets, the width of the XPS C 1s peak becomes narrower and the intensity of the valence band increases. Field emission from the as-grown CNTs exhibits a large enhancement of current density with growth temperature, strongly correlated with the electronic structure revealed by XPS.
Introduction Carbon nanotubes (CNTs) are currently attractive materials for a diverse range of applications because of their extraordinary mechanical and electrical properties. Their application has already been demonstrated in field emission displays,1 nanoscale electronic devices,2-5 biosensors,6 and hydrogen storage mediums.7 Many of these utilizations are based on the electrical properties of CNTs, and therefore, further advance requires more precise control not only of the diameter/length but also the electronic structure. However, although various growth methods have developed considerably,8-13 it is still difficult to manipulate the electronic structure using the growth conditions. In the case of pyrolysis and chemical vapor deposition syntheses, research has already demonstrated that the growth temperature would render the control of the length, structure, and degree of crystalline perfection of multiwalled CNTs (MWCNTs).14-16 In this paper, the electronic structure of MWCNTs reliance on growth temperature is investigated by employing synchrotron X-ray photoelectron spectroscopy (XPS) and field emission measurements. High-purity vertically aligned CNTs were synthesized via pyrolysis of ferrocene (FeC10H10) and acetylene (C2H2) in the wide temperature range of 600-1000 °C. The photon energy of XPS is variable in the range 100-1000 eV using synchrotron radiation. The fine-scanned C 1s peak and the valence band (VB) spectrum have been measured by analyzing the nature of the electronic structure of CNTs. The field-emission properties are examined from the as-grown CNTs. A strong correlation between the current/turn-on voltage of field emission and the electronic structure has been found. Experimental Section A quartz tube reactor was heated in a dual furnace fitted with independent temperature controllers. Ferrocene (Aldrich, 98+%) * Corresponding authors. E-mail:
[email protected] (J.P.); cjlee@ hanyang.ac.kr (C.J.L.).
was used after it was vacuum-dried at 100 °C for 1 h and then placed in the first initial heating stage. Silicon (Si) substrate was positioned in the second heating stage. Ferrocene was vaporized at 200 °C and carried by the Ar flow (200 sccm) into the second heating stage of the quartz tube where pyrolysis was performed at 600-1000 °C. C2H2 (Kyungin Gas, 99.99%) flowed with a rate of 5-30 sccm for 2-10 min. The CNTs were grown on the substrates at the second stage. The morphology and structure were analyzed by scanning electron microscopy (SEM, Hitachi S-4300), field-emission transmission electron microscopy (TEM, FEI TECNAI G2, 200 kV), and Raman spectroscopy (Renishaw micro-Raman 2000) using the 514.5 nm line of an argon ion laser. XPS measurements were performed at the U7 and U10 beam lines of the Pohang Light Source (PLS).17 The data were collected using the photon flux of (4-7) × 1011 ((photons/s)/ 200 mA). The binding energies were corrected for specimen charging by referencing the C 1s peak to 284.6 eV, and the background was subtracted using Shirley’s method.18 The experiment was performed in an ultrahigh vacuum chamber with base pressure e5 × 10-10. The photoelectrons emitted from the surface of the sample were collected, and their energy was analyzed with an electron energy analyzer (Physical Electronics: model PHI 3057 with a 16-channel detector). The analyzer was located at 55° from the surface normal. The conductivity of as-grown CNTs on the substrates was measured using a fourprobe method with a HP4156 parameter analyzer. Field electron emission measurements were performed in a vacuum chamber at a pressure of less than 5 × 10-6 Torr. The cathode consisted of as-grown CNTs on the n-type Si substrate (1 × 2 cm2), and the anode was a copper plate (2 × 2 cm2). The distance between the cathode and the anode was approximately 370 µm, and the measured emission area was 7 mm2. Emission current was monitored with a Keithley 6517 A and recorded at intervals of 0.5 s.
10.1021/jp0546305 CCC: $30.25 © 2005 American Chemical Society Published on Web 10/07/2005
20404 J. Phys. Chem. B, Vol. 109, No. 43, 2005
Yoon et al.
Figure 1. (a) SEM micrograph of the vertically aligned CNTs grown at 800 °C. (b) TEM image showing a typical morphology of CNTs and (c) its atomic resolved image.
Figure 2. Raman spectrum of CNTs grown at various temperatures in the range 600-1000 °C. The excitation wavelength is the 514 nm line from an Ar ion laser.
Results and Discussion The SEM micrograph presents the vertically aligned CNTs grown on a large area of the substrates at 800 °C (Figure 1a). The length is approximately 15 µm. The corresponding TEM image is displayed in Figure 1b. The CNTs were dispersed in ethanol, to load on a TEM grid. The diameter of CNTs is 1030 nm, with an average value of 20 nm. The CNTs exclusively exhibit a cylindrical structure with a few encapsulated catalytic particles. The atomic resolved image of a typical CNT is presented in Figure 1c. The wall thickness is 5 nm. The crystalline graphitic sheets are aligned parallel to the tube axis. For all growth temperatures (600-1000 °C), the average diameter is maintained at 20 nm.15,16 The length of CNTs is controlled by the growth time and kept 15-20 µm (Supporting Information, Figure S1). The atomic resolved TEM image reveals that the CNT grown at the higher temperature exhibits a higher degree of crystalline perfection in the graphitic sheets. To obtain information regarding the crystallinity of the entire CNTs, first-order Raman spectroscopy (Figure 2) was employed. Two Raman bands at ∼1580 cm-1 (G band) and ∼1350 cm-1 (D band) originate from the Raman-active in-plane atomic displacement E2g mode and disorder-induced features, due to the finite particle size effect or lattice distortion, respectively.19,20 The intensity ratio of D band to G band (ID/IG) has a linear relation with the inverse of the in-plane crystallite dimension. As temperature increases from 600 to 1000 °C, the value of ID/IG decreases from 0.87 to 0.25. Table 1 lists the ID/IG value for all temperatures. It indicates that the degree of long-range
ordered crystalline perfection increases with the growth temperature. The growth of the present CNTs follows the basegrowth model as described elsewhere.14-16 As the C atoms melt and saturate into the catalytic Fe nanoparticles, they precipitate to start the build up of graphitic sheets. The growth temperature would play a critical role in the growth rate as well as the degree of crystalline perfection, by controlling the concentration of saturated C atoms and the precipitation rate.14-16 Figure 3 shows the fine-scanned XPS C 1s spectrum for the CNTs grown at 600-1000 °C, using the photon energy of 630 eV. The width and shape are nearly independent of the photon energy 360-1000 eV. The respective probing depth of C 1s electrons is estimated to be approximately 2 nm.21 As the growth temperature increases from 600 to 1000 °C, the full-width at half-maximum (fwhm) of asymmetric band centered at 284.6 eV decreases significantly from 0.9 to 0.5 eV. This asymmetric band could be deconvoluted into two bands at 284.5 (PC1) and 285.5 (PC2) by fitting into the Voigt function. It is expected that the binding energy of C atoms bonded to the defects as dangling bonds will appear at a higher energy relative to those of the graphite C atoms. Therefore, the dominant PC1 band can be assigned to the C atoms binding to the graphite network and the weaker PC2 band correspond to the C atoms at the defect sites, consistent with the work of graphite film.22 The fwhm and the fraction of PC1 and PC2 bands are listed in Table 1. As temperature increases, the fwhm of both PC1 and PC2 bands decreases, which is robustly related with the increased crystalline perfection of the graphitic sheets. Figure 4 presents the VB spectrum of XPS using the photon energy 360 eV, for the vertically aligned CNTs grown at different temperatures. Zero energy is chosen as the Fermi level (Ef), which, as shown in the figure, is calibrated using the threshold energy of Au foil. The spectrum is normalized using the intensity at 13 eV. Following the works of graphite films and MWCNTs, the spectrum in the region between 2.0 and 7.6 eV (marked by “A”) is assigned to graphite 2p-π, which overlaps on the top of graphite 2p-σ.23,24 As growth temperature increases, the intensity of this “A” region increases. It is reported that the MWCNTs exhibit a lower intensity in this region than the graphite, resulting from the curvature of graphite sheets.24 The π electrons tend to behave less delocalized in small tubes due to the curvature-induced strain.25 Therefore, the enhanced intensity with the growth temperature might come from the more delocalized 2p-π electrons due to the better crystallinity of the graphitic sheets. The vicinity of Ef in the range of 0-1.0 eV (marked by “B”) exhibits a significant feature, in which the intensity rises increasingly at the Ef, as the growth temperature increases. This steeper rise is directly related with the increased density of states (DOS) at the Ef. It is well-known that the metal exhibits a steep rise at the Ef, as can be seen from the Au foil. Thus, the steepness of the DOS at the Ef also represents the
Multiwalled Carbon Nanotubes
J. Phys. Chem. B, Vol. 109, No. 43, 2005 20405
TABLE 1: Fwhm and Area % of Deconvoluted Bands from the XPS C 1s Peak for the CNTs Grown at Various Temperatures fitting params PC1 (graphite)
PC2 (defects)
growth temp (°C)
Raman ID/IGa
XPS C 1s fwhm (eV)b
fwhm (eV)
area (%)
fwhm (eV)
area (%)
conductivity (S/cm)
600 700 800 900 1000
0.87 0.53 0.44 0.33 0.25
0.9 0.8 0.65 0.6 0.5
0.8 0.8 0.6 0.6 0.5
84 86 86 88 88
0.8 0.8 0.6 0.6 0.5
16 14 14 11 11
0.57 3.2 5.7 7.1 8.2
a
The intensity ratio of D band relative to G band of Raman spectrum with an error of 5%. b The XPS C 1s spectrum measure using 630 eV.
Figure 3. XPS C 1s spectrum of CNTs grown at various temperatures, using the photon energy 630 eV. The data points (open circles) of the C 1s band are fitted by two Voigt functions PC1 and PC2 (dotted lines).
Figure 4. XPS valence band spectrum using 360 eV for the vertically aligned CNTs on the substrates.
metallic properties. The higher growth temperature would result in the greater metallic behavior of the CNTs. There is the question of why the DOS at the Ef shows significant enhancement with the growth temperature. Suzuki et al. proposed that the dangling bond density at the spherically curved tips contributes to increasing the DOS at the Ef.26,27 However, the enhancement of the DOS with the growth temperature cannot be explained properly by dangling bond defects, since the analysis of the Raman and XPS C 1s peak proves a higher degree of crystalline perfection formed at the higher growth temperature. Recently, the field emission measurement of individual MWCNT shows that the electron is emitted predominantly from a tip of MWCNT.28 It was suggested that the pentagonal carbon rings at the tip would have a higher DOS than the hexagonal carbon rings, so the electrons
would be emitted dominantly. The DOS along a SWNT capped by a hemispherical C240 tip apex containing six pentagons are calculated to show a prominent peak just above the Fermi level, which is associated with the presence of the pentagons at the tips.29 Therefore, it is expected that as the growth temperature increases, the formation of pentagon rings at the curved tips may be facilitated to reduce the defect sites and release the strains of the more crystalline graphitic sheets. The increased number of pentagon rings at the curved graphitic sheets of the tips may increase the DOS at the Ef. There is a possibility that the presence of encapsulated Fe catalytic particles contributes in the steep rise of DOS at the Ef. However, a negligible amount (∼1%) of Fe is detected by EDX and XPS (Supporting Information, Figure S2), irrespective of growth temperature. Thus, it is apparently difficult to correlate the temperature dependence of DOS at the Ef with encapsulated Fe catalytic particles. Nevertheless further investigation is required for the origin of the DOS at the Ef. Conductivity measurement has been carried for the as-grown CNTs on the substrates. Table 1 lists the conductivity (S/cm) of the samples. The standard four-probe technique was used, and the spacing of tips was 35 µm. As growth temperature increases, the conductivity enhances greatly by a factor of 15. The conductivity of 0.6-8 S/cm is nearly metallic. As the crystallinity of the graphitic sheets increases, the contact resistance of inter-CNTs is much reduced. The field emission characteristics of as-grown CNTs are presented in Figure 5a. As mentioned above, the length of CNTs is 15-20/µm, for all growth temperatures. A typical turn-on field, which produces a current density of 0.1 µA/cm2, is about 2.9, 1.4, and 1.2 V/µm, respectively, for the CNTs grown at 600, 800, and 1000 °C. The emission curve was measured from more than 10 different spots of each sample, and its average is plotted in this figure. The data in log scale are plotted in Supporting Information, Figure S3. The respective emissioncurrent density reaches 1 mA/cm2 at an applied field of about 6.0, 3.0, and 2.5 V/µm. As growth temperature increases from 600 to 1000 °C, the turn-on voltage is reduced and the emission current density is enhanced by a factor of more than 2. A bump is observed from the emission curve of the CNTs grown at 600 and 1000 °C, which may originate from the inhomogeneous morphology of the tips over the measurement area. The corresponding Fowler-Nordheim (FN) plot is shown in Figure 5b. To calculate the field enhancement factor (β), the FowlerNordheim (FN) equation is used:
J)A
( ) (
)
Bφ3/2d β2V2 exp βV φd2
Here J is the current density, A ) 4.43 × 10-22, B ) 6.83 × 109, β is a field enhancement factor, φ is the work function, d is the distance between the anode and the cathode, and V is the applied voltage. When the work function of CNTs is assumed
20406 J. Phys. Chem. B, Vol. 109, No. 43, 2005
Yoon et al. indicating a strong correlation with the crystallinity of the graphitic sheets. Acknowledgment. This work was supported by the KOSEF (Project No. R14-2003-033-01003-0 and R02-2004-000-100250) and KRF (Project No. 2003-015-C00265). SEM and TEM analyses were performed at Korea Basic Science Institute (KBSI). Experiments at PLS were supported in part by MOST and POSTECH. Supporting Information Available: SEM images, XPS, and field emission data in log scale. This material is available free of charge via Internet at http://pubs.acs.org. References and Notes
Figure 5. (a) Emission current densities from the vertically aligned CNTs grown on the Si substrates. (b) Corresponding Fowler-Nordheim plot.
to be 5 eV, the field enhancement factor is estimated to be 1500, 3300, and 3800, the value of which is sufficient for applications of field emission displays. It is believed that improved emission properties at higher growth temperatures are caused by more delocalized 2p-π electrons, as well as the higher DOS at the Ef. The field emission properties are thus directly correlated with the degree of crystalline perfection in the graphitic sheets. This correlation further suggests that the electrons are emitted predominantly from the tips. Therefore, the crystallinity of the graphitic sheets plays an important role in determining the field emission properties for the application of field emission displays. Conclusion In summary, the vertically aligned CNTs are synthesized on the Si substrates via pyrolysis of ferrocene and acetylene in the temperature range 600-1000 °C. The CNTs usually have a cylindrical structure and a uniform diameter of 20 nm. Raman spectroscopy reveals that the degree of crystalline perfection increases significantly as growth temperature increases. Synchrotron XPS is employed to investigate the dependence of the electronic states on the growth temperature. The narrower bandwidth of the C 1s peak at the higher growth temperature is strongly correlated with the higher degree of crystalline perfection of graphitic sheets. The valence band analysis reveals significantly more delocalized 2p-π electrons and the increased DOS at the Ef with the growth temperature. It is suggested that as crystalline perfection of graphitic sheets improves, the formation of pentagon rings at the curved tips is favorable to release the strains, enhancing the DOS at the Ef. Field emission from the as-grown CNTs presents a lower turn-on voltage and a largely enhanced current density with the growth temperature,
(1) Saito, Y.; Hamaguchi, K.; Hata, K.; Uchida, K.; Tasaka, Y.; Ikazaki, F.; Yumura, M.; Kasuya, A.; Nishina, Y. Nature 1997, 389, 554. (2) Hu, J. T.; O. Y. Min, Yang, P. D.; Lieber, C. M. Nature 1999, 399, 48. (3) Antonov, R. D.; Johnson, A. T. Phys. ReV. Lett. 1999, 83, 3274. (4) Fuhrer, M. S.; Nygard, J.; Shih, L.; Forero, M.; Yoon, Y.-G.; Mazzoni, M. S. C.; Choi, H. J.; Ihm, J.; Louı`e, S. G.; Zettl, A.; Mceuen, P. L. Science 2000, 288, 494. (5) Tans, S. J.; Dekker: C. Nature 2000, 404, 834. (6) Kong, J.; Franklin, N. R.; Zhou, C.; Chapline, M. G.; Peng, S.; Cho, K.; Dai, H. Science 2000, 287, 622. (7) Liu, C.; Fan, Y. Y.; Liu, M.; Cong, H. T.; Cheng, H. M.; Dresselhaus, M. S. Science 1999, 286, 1127. (8) Bethune, D. S.; Kiang, C. H.; deVries, M. S.; Gorman, G.; Savoy, R.; Vazquez, J.; Beyers, R. Nature 1993, 363, 605. (9) Thess, A.; Lee, R.; Nikolaev, P.; Dai, H.; Petit, P.; Robert, J.; Xu, C.; Lee, Y. H.; Kim, S. G.; Rinzler, A. G.; Colbert, D. T.; Scuseria, G. E.; Toma´nek, D.; Fisher, J. E.; Smalley, R. E. Science 1996, 273, 483. (10) Journet, C.; Maser, W. K.; Bernier, P.; Loiseau, A.; Lamy de la Chapelle, M.; Lefrant, S.; Deniard, P.; Lee, R.; Fischer, J. E. Nature 1997, 388, 756. (11) Terrones, M.; Grobert, N.; Olivares, J.; Zhang, J. P.; Terrones, H.; Kordatos, K.; Hsu, W. K.; Hare, J. P.; Townsend, P. D.; Prassides, K.; Cheetham, A. K.; Kroto, H. W.; Walton, D. R. M. Nature 1997, 388, 52. (12) Ren, Z. F.; Huang, Z. P.; Xu, J. W.; Wang, J. H.; Bush, P.; Siegal, M. P.; Provencio, P. N. Science 1998, 282, 1105. (13) Rao, C. N. R.; Govindaraj. Acc. Chem. Res. 2002, 35, 996. (14) Lee, Y. T.; Park, J.; Choi, Y. S.; Ryu, H.; Lee, H. J. J. Phys. Chem. B 2002, 106, 7614. (15) Lee, Y. T.; Kim, N. S.; Park, J.; Han, J. B.; Choi, Y. S.; Ryu, H.; Lee, H. J. Chem. Phys. Lett. 2003, 372, 853. (16) Kim, K.-E.; Kim, K.-J.; Jang, W. J.; Bae, S. Y.; Park, J.; Choi, J.; Choo, J. Chem. Phys. Lett. 2005, 401, 459. (17) Lee, M. K.; Shin, H.-J. Nucl. Instrum. Methods Phys. ReV., Sect. A 2001, 467-468, 508. (18) Shirley, D. A. Phys. ReV. B 1972, 5, 4709. (19) Tuinstra, F.; Koenig, J. L. J. Chem. Phys. 1970, 53, 1126. (20) Wilhelm, H.; Lelaurain, M.; McRae, E.; Humbert, B. J. Appl. Phys. 1998, 84, 6552. (21) Teo, B. K.; EXAFS: Basic Principles and Data Analysis; SpringerVerlag: Berlin, 1986; p 92. (22) Estrade-Szwarckopf, H. Carbon 2004, 42, 1713. (23) McFeely, F. R.; Kowalczyk, S. P.; Ley, L.; Cavell, R. G.; Pollak, P. A.; Shirley, D. A. Phys. ReV. B 1974, 9, 5268. (24) Chen, P.; Wu, X.; Sun, X.; Lin, J.; Ji, W. Tan, K. L. Phys. ReV. Lett. 2002, 82, 2548. (25) Ajayan, P. M.; Iijima, S.; Ichihashi, T. Phys. ReV. B 1993, 47, 6859. (26) Suzuki, S.; Watanabe, Y.; Kiyokura, T.; Nath, K. G.; Ogino, T.; Heun, S.; Zhu, W.; Bower, C.; Zhou, O. Phys. ReV. B 2001, 63, 245418. (27) Suzuki, S.; Watanabe, Y.; Ogino, T.; Heun, S.; Gregoratti, L.; Kaulich, B.; Kiskinova, M.; Zhu, W.; Bower, C.; Zhou, O. Phys. ReV. B 2002, 66, 035414. (28) Konishi, Y.; Hokushin, S.; Tanaka, H.; Pan, L.; Akita, S.; Nakayama, Y. Jpn. J. Appl. Phys. 2005, 44, 1648. (29) Charlier, J.-C. J. Mater. Res. 1998, 13, 2368.
