Induced NH
2bonding of carbon nanotubes using NH
3plasma-enhanced chemical vapor deposition
Ai-Huei Chiou• Yu-Ming Chang•Wen-Fa Wu•
Chang-Ping Chou•Chun-Yao Hsu
Received: 29 June 2011 / Accepted: 25 August 2011 / Published online: 7 September 2011 Ó Springer Science+Business Media, LLC 2011
Abstract Plasma-enhanced chemical vapor deposition was used to modify the multiwall carbon nanotubes (MWCNTs) using ammonia (NH3) plasma. For various
durations of NH3 plasma treatment, a scanning electron
microscope, X-ray, Raman spectroscopy and contact angle measurement were used to ascertain several characteristics of the MWCNTs. The experimental results show that: (1) the length of the MWCNTs is reduced, if the duration of the plasma treatment is increased; (2) the NH3 plasma
treatment can incorporate amine (NH2-) or amino (NH-)
functional groups onto the MWCNT surface; (3) the plasma treated carbon nanotubes become more hydrophilic.
1 Introduction
Carbon nanotubes (CNTs) are widely used for electronic components, chemical and biological sensors, scanning probe microscopy, field emission tips, high-strength com-posites, hydrogen and ion storage components [1] because of their physical and chemical properties and their
excellent mechanical properties [2]. Although CNTs have many advantages, their surface contains many impurities, including amorphous carbon, carbon oxides and other chemical compounds, which hinder the application of CNTs, in practical systems. For use in biology and medicinal chemistry, CNTs must be purified and dissolved in physiological media. If the CNT surface is to be modi-fied, to alleviate these problems, more precise control of chemical functionalization and an understanding of the related structures are needed.
In recent years, several post treatment techniques, such as chemical processes [3], electron irradiation, plasma treatment and laser exposure [4], have been developed. Each of these techniques has proven to be an effective method of cleaning CNTs, but some techniques can dam-age the structure and properties of the CNT. For example, acid treatment is often used to introduce -COOH and -OH groups, to modify CNTs and increase their aqueous solu-bility [5]. This can damage the structure of the CNT [6], decrease its stability [7] and even sever the CNT into short pieces [8]. In addition, most chemicals have disadvantages such as toxicity, instability and the possibility of environ-mental pollution.
Of these approaches, plasma treatment is the most widely used purification method, because it reduces amorphous carbon and induces bonding onto the CNT surface. The interface between the materials and their environment usually plays a decisive role in many appli-cations, such as biology, or drug delivery systems [9]. Therefore, the surface modification of materials has become an important topic for study, because the properties of the material can be maintained, even if the surface structure is changed [10].
Plasma-enhanced chemical vapor deposition (PECVD) is a special type of CVD process that utilizes plasma to
A.-H. Chiou C.-P. Chou
Department of Mechanical Engineering, National Chiao Tung University, Hsinchu, Taiwan, ROC
Y.-M. Chang
Institute of Nuclear Energy Research, Atomic Energy Council, Executive Yuan, Taoyuan, Taiwan, ROC
W.-F. Wu
National Nano Device Laboratories, Hsinchu, Taiwan, ROC C.-Y. Hsu (&)
Department of Mechanical Engineering, Lunghwa University of Science and Technology, Kueishan, Taiwan, ROC e-mail: [email protected]
enhance the chemical reaction rates of precursors. The process uses radio frequency-generated plasma to assist the deposition reaction at lower temperatures [11]. PECVD has many benefits, in that it requires a lower temperature than conventional CVD, it reduces the thermal stress induced by small fabrication defects and it can also be used to improve printability, wettability, bondability, surface hardness and surface heat resistance. It also provides a method of cleaning surfaces, without the need for solvents and introduces cross-linking, at the surface.
Few studies have considered the effect of amino plasma treatment on CNTs. This investigates using PECVD system with amino-related functional groups for MWCNTs. There are three major sections: (1) the induction of amine (NH2)
bonding in MWCNTs, using PECVD; (2) characterization of the difference between the morphologies and structure of MWCNTs, before and after amino (NH3) plasma
treat-ment, using PECVD and (3) an investigation of the effect of changing the duration of NH3plasma treatment on the
deposition of nitrogen-containing groups onto the surfaces of MWCNTs.
2 Experimental procedures
2.1 Synthesis and functionalization of MWCNTs
The substrate used in this study was a 6-inch, p-type (100) silicon wafer, with resistivity of 15–25 X-cm. Firstly, in order to remove chemical impurities and particles, the wafer was cleaned, using a standard Radio Corporation of American cleaning process. Mattel PVD was used to deposit a 20 nm thick titanium nitride buffer layer and then a 7 nm thick transition metals (nickel) layer on the wafer substrates. Secondly, MWCNTs were synthesized, using thermal chemical vapor deposition (AT-CVD) [12]. The base pressure of the system was 760 Torr. The AT-CVD was performed at 550°C, with H2:CH4= 5:2 sccm for
1 min.
It has been reported that the structure of CNTs depended on the synthesis parameter, such as reaction temperature, catalyst composition and reaction gas [13]. In order to investigate the influence of the Ni catalyst layers (7 nm) transformed into nanoparticles, after different H2gas flows
rate (100 and 500 sccm) during pre-treatment are presented in Fig.1. It is shown that, with higher H2 gas flow rate
during pre-treatment, lead to denser Ni catalyst nanopar-ticles and smaller RMS surface roughness (Fig.1b). This is also similar to the results of Jian et al. [14]. Figure2shows the influences of H2 flow rate pre-treatment on the
mor-phology of MWCNTs synthesized at 550°C, with H2:CH4= 5:2 sccm for 1 min. By applying higher H2gas
flow rate during pre-treatment (Fig.2b), the amorphous
carbon and carbonaceous particles were decreased, and the MWCNTs yield increases apparently.
Finally, the MWCNTs were modified, using PECVD. For the post treatment, the microwave power was 300 W and the NH3gas flow rate was 700 sccm. In order to study
the effect of varying the duration of the NH3 plasma
treatment, the ammonia plasma post-treatments were functionalized at 300°C, for 1, 3, 5 and 10 min, using PECVD (listed in Table 1).
2.2 Characterization of the morphologies and structures
A scanning electron microscope (SEM, Hitachi S-4000) was used to observe the lengths, density, morphologies and
Fig. 1 AFM images and RMS surface roughness of Ni catalyst nanoparticles by different H2gas flow rate during pretreatment at
functionalized of the MWCNTs, and the functionalized MWCNTs. Fourier transform infrared spectroscopy (FTIR, ASTeX PDS-17) was used to analyse the chemical com-positions. X-ray Photoelectron Spectroscopy (XPS, VG Microlab 310F) was used to investigate the functional
groups and the chemical elements. Raman spectroscopy (Raman, Renishaw 2000) was used to examine the region between the crystallization and atom bands of the MWCNTs. A Contact Angle system (CA, KR }USS GmbH GH-100) was also used to analyze the hydrophilic and the hydrophobic nature of the MWCNTs, and the functional-ized MWCNTs.
3 Results and discussion
3.1 Morphological analysis
The effects of varying the duration of NH3plasma
treat-ment time, on the surface morphologies of untreated and treated samples of MWCNTs are presented in Figs.3, 4,
Fig. 2 SEM images of CNTs by different H2gas flow rate during
pretreatment at a 100 and b 500 sccm
Table 1 Symbols representing MWCNTs samples with different plasma treatment times
System PECVD Exp. parameters C-NH3-1 C-NH3-3 C-NH3-5 C-NH3-10 Plasma type NH3 NH3 NH3 NH3 Process temp. (°C) 300 300 300 300 Aux temp. (°C) 250 250 250 250 Flow rate (sccm) 700 700 700 700 Power (Watt) 100 100 100 100
Treatment time (min) 1 3 5 10
Fig. 3 SEM images of MWCNTs, using AT-CVD, at 550°C, for 1 min, with an H2/CH4 mixture, a gas flow rate of 5:2 sccm and a base pressure of 760 Torr, a top view and b side view
and5. Figure3 shows the morphology of the as synthe-sized MWCNTs, on a silicon wafer. The length of the MWCNTs is observed to be approximately 3.188 lm, as
shown in Fig.3 (b). Figures4 and 5 show the treated MWCNTs, for different durations of NH3 plasma
treatment.
Fig. 4 Top-view SEM images of MWCNTs, after plasma treatment with an NH3 gas Flow rate of 700 sccm, microwave power of 100 W and process temperature of 300°C, for a1 min, b 3 min, c 5 min and d10 min
Fig. 5 Side-view SEM images of MWCNTs after plasma treatment with an NH3 gas flow rate of 700 sccm, microwave power of 100 W and process temperature of 300°C, for a1 min, b 3 min, c 5 min and d10 min
The surface morphology of MWCNTs with no NH3
plasma treatment is very rough. However, the surface morphology of MWCNTs becomes smooth, after NH3
plasma treatment. In addition, the length of MWCNTs decreases, as the duration of the plasma treatment increa-ses, as shown in Fig.6. It can be seen that NH3 plasma
treatment performs an effective etching and cleaning pro-cess on MWCNTs [15], which can be established by Raman analysis. This method may cause interconnection in MWCNTs, which gives better surface roughness.
3.2 Graphite analysis
Raman spectroscopy is widely used to analyze the structure of MWCNTs [16]. This technique is highly sensitive to the disorder on the surfaces, because of the optical skin depth [17]. Figure7 shows Raman spectra of MWCNTs before and after the NH3 plasma treatment, as a function of
dif-ferent treatment time. These Raman spectra exhibit two distinct peaks, at 1,300–1,380 cm-1 (D band) and 1,550–1,600 cm-1(G band). The D band peak is generally caused by defects in the curved graphite sheet, sp3carbon, or other impurities and the G band shows the presence of crystalline, graphitic carbon in MWCNTs associated with the motion of two neighboring carbon atoms in a graphite sheet, in the reverse direction [18]. The ratio of the inten-sities of the D and G band peaks can be used to evaluate the comparative extent of structural defects. It has been con-firmed [19] that the ID/IG ratio can be effectively
deter-mined by crystal space interaction. When ID/IG ratio
decreases, the quality of MWCNTs improves.
Analysis of the results shows that the value of the ID/IG
ratio, for MWCNTs with no modification, is 1.180. This value decreases, after 1, 3, and 5 min of NH3 plasma
treatment, as shown in Fig.7. The decrease in ID/IGvalues
probably occurs because of increased nitrogen content and
the etching effect, which reduces surface impurities, as presented in Table2. However, the ID/IG ratio increases,
for the sample with 10-min NH3 plasma treatment. The
increase in ID/IGratio for the plasma modified MWCNTs is
probably due to over etching of the MWCNTs by NH3
plasma treatment, which is consistent with SEM observations.
3.3 Hydrophobic properties
Water contact angle (WCA) introduced by Thomas Young [20] as a way of investigating wetting behavior and surface tension, now have many fields of application in the mea-surement of the characteristics of single, or multi-layer surface treatment techniques [21], such as the aging char-acteristics of modified surfaces and the migration of hydrophobic and hydrophilic functional groups, in aqueous and nonaqueous environments [22]. Contact angle analysis is sensitive to the chemical composition of the top molecular layer and is a relatively simple, inexpensive and popular technique for the characterization of material
Fig. 6 Effect of the duration of plasma treatment on the length of MWCNT, for an NH3 gas flow rate of 700 sccm, microwave power of 100 W and process temperature of 300°C
Fig. 7 Raman analysis of MWCNTs after plasma treatment, with an NH3 gas flow rate of 700 sccm, microwave power of 100 W and process temperature of 300°C, for a 0 min, b 1 min, c 3 min, d5 min and e 10 min
Table 2 Values of peaks at the D band and G band in the Raman spectra for the MWCNTs before and after different durations (1, 3, 5 and 10 min) of plasma treatment
Sample D-peak G-peak ID/IG
CNTs 1,328 1,588 1.180
C-NH3-1 1,322 1,595 1.174
C-NH3-3 1,324 1,598 1.144
C-NH3-5 1,330 1,593 1.177
surfaces. After plasma treatment, the bulk surface can induce hydrophobic functional groups.
Figure8shows the WCA testing of MWCNTs, without and with modification by NH3 plasma treatments. The
unmodified MWCNTs have hydrophobic features, with a contact angle of 101.3°. The different contact angles, after various NH3plasma treatments, are shown in Table 3. This
result indicates that the treated MWCNTs become hydro-philic. It also presents evidence of the functionalization of the MWCNT surface. This occurs because of the formation of nitrogen-containing functional groups at the MWCNTs surface, through the reactions between the active species from the NH3 plasma and the surface atoms. The NH3
plasma increases the surface energy of MWCNTs surface. Figure8 shows that the number of hydrophilic ions increases, with the duration of NH3plasma treatment. This
phenomenon has also been observed by several researchers [23, 24]. Ammonia is considered to be an effective pre-cursor to the introduction of amine functionalities to carbon nanotubes [23] and other materials [24], enhancing their hydrophilicity and biocompatibility. The use of NH3, with
PECVD may induce nitrogen-containing functional groups onto the MWCNT surface, which can be confirmed by XPS and FTIR analysis. WCA analysis shows a correlation between ammonia and hydrophilicity.
3.4 Elemental analysis
In order to evaluate the functional groups formed by the NH3 plasma treatment and to discover the reason for the
reduction in contact angle, the chemical bonds on MWCNT surface treated by NH3 plasma were determined, using
FTIR and XPS analysis. Figure9shows the results of FTIR analysis of the MWCNT chemical bonds, for various durations of modification. The peak, at 2,800–2,950 cm-1, is attributed to C–H stretching vibration. N–H stretching vibration bands, at 3,200–3,400 cm-1, can be observed. These provide further information concerning the effects of NH3 plasma on the surface of MWCNTs, as shown in
Fig.9. The results show that the peak at 3,100–3,500 cm-1 is caused by NH2 groups [25]. These results agree with
those reported by Yook et al. [26].
XPS analysis was used to investigate the effect of NH3
plasma treatment of MWCNTs. XPS analysis determines the difference between the untreated MWCNT and the modified MWCNT, through the elemental chemical shifts. Figure10shows the results of XPS analysis for MWCNT chemical bonds, for various durations of modification. Before modification, the full spectrum appears to be composed of graphitic carbon (284.8 eV), as shown in Fig.10. The modification of MWCNTs by NH3 plasma
treatment does not cause a large change in the C1 s and
Fig. 8 WCA observation of aoriginal MWCNTs samples and after plasma modification b1 min, c 3 min, d 5 min and e10 min, with an NH3gas flow
rate of 700 sccm, microwave power of 100 W and process temperature of 300°C, using a CCD
Table 3 Measured water contact angle (WCA) for MWCNTs, with different durations of plasma treatment (1, 3, 5 and 10 min) Parameter CNTs C-NH3-1 C-NH3-3 C-NH3-5 C-NH3-10
WCA (°) 101.3 97.92 82.75 60.71 46.20
Fig. 9 Comparison of FTIR spectra for a original MWCNTs samples and after NH3plasma treatment b 1, c 3, d 5 and e 10 min, with an
NH3 gas flow rate of 700 sccm, microwave power of 100 W and
O1 s spectra, but there is an observable change in the N1 s spectra. The atomic compositions of untreated MWCNTs are 97.13% C and 2.87% O. After modification, N content increases linearly, with duration of NH3plasma treatment,
as shown in Table4. Detailed examination of the C1 s, O1 s and N1 s spectra allowed the acquisition of more information about the bonding structure of the MWCNTs. Figure11 shows the intensity of the N1 s spectrum. There are no observable peaks for the untreated MWCNTs, but peaks appear for the NH3plasma modified MWCNTs.
The N1 s spectrum may consist of more than one element, for example, -C : N (from 398.3 to 399.2 eV), C–NH2
(from 398.3 to 399.2 eV), N = C (from 399.0 to 400.1 eV), -NH-(from 399.7 to 400.0 eV), -C–NH3 (from
399.7 to 400.0 eV), C–N (from 399.7 to 400.0 eV) and O = C–NH2 (at 400.4 eV) [27]. This is caused by the
decomposition of the ammonia that bonded the surface of MWCNTs, to free radicals (NH2or NH) [28]. Figure12
shows the typical high-resolution C1 s spectrum, for MWCNTs, with and without NH3 plasma treatment. The
findings suggest that: (1) sp2-hybridized graphite-like car-bon atom (C = C) is at 284.1 ± 0.2 eV; (2) the sp3
hybridized carbon atom (C–C) is at 285.1 ± 0.2 eV; (3) peaks at 286.0 ± 0.2 eV are considered to be C–O and C– N groups; (4) peaks at 287.2 ± 0.2 eV and at 288.7 ± 0.2 eV are considered to be C = O group and O– C = O group, respectively [29,30].
Table5 indicates that N/C ratios increase, from 0 to 0.048, and O/C ratios increase, from 0.030 to 0.265, if the duration of NH3plasma treatment is increased. This occurs
because of the formation of some nitrogen groups on the MWCNT surfaces, due to NH3 plasma treatment.
There-fore, the ammonia plasma post-treatments can introduce functional groups, such as C–NH2and C–N. These results
demonstrate that PEVCD treatment is a reliable and effective method, which can expediently endow MWCNTs with various new functional groups, while maintaining their original bulk structures.
Fig. 10 The XPS wide scan spectra for MWCNTs samples, with an NH3gas flow rate of 700 sccm, microwave power of 100 W and
process temperature of 300°C, with plasma modification for a 0 min, b1 min, c 3 min, d 5 min and e 10 min
Table 4 Atomic concentration of C, N, and O, in MWCNTs with different surface modification times
Element contain (%)
Samples (treatment time) C1 s (%) N1 s (%) O1 s (%)
Untreated MWCNTs 97.13 – 2.87
C-NH3-1 (1 min) 86.69 2.5 7.81
C-NH3-3 (3 min) 82.43 2.99 14.58
C-NH3-5 (5 min) 76.17 3.66 20.17
Fig. 11 N1 s spectra for MWCNTs samples, with an NH3gas flow
rate of 700 sccm, microwave power of 100 W and process temper-ature of 300°C, with plasma modification for a 0 min, b 1 min, c3 min, d 5 min and e 10 min
Fig. 12 C1 s spectra for MWCNTs samples, with an NH3gas flow
rate of 700 sccm, microwave power of 100 W and process temper-ature of 300°C, with plasma modification for a 0 min, b 1 min, c3 min, d 5 min and e 10 min
4 Conclusion
In this study, the induction of amino groups onto an MWCNT surface, using a PECVD system, was investi-gated. After NH3plasma treatment, the surface roughness
of the MWCNT improved but the quality (ID/IG ratio) of
the MWCNT decrease, after plasma treatment for a longer period (5 and 10 min). NH3plasma treatment shortened the
length of the MWCNTs grown on the blanket.
The results of XPS and Raman spectra analysis indicate that treatment time affects the atomic composition and structural properties of MWCNTs. The approach taken is an efficient method for the introduction of amino groups onto MWCNTs surfaces. NH3plasma treatment was
con-firmed to be an effective precursor to the introduction of amine functionalities onto the MWCNT surface, to enhance their hydrophilicity and biocompatibility. These results demonstrate that the induction of amino groups onto an MWCNT surface, using a PECVD system, is an effec-tive and reliable method.
References
1. M. Paradise, T. Goswami, Mater. Des. 28, 1477 (2007) 2. T. Hertel, R. Walkup, P. Avouris, Phys. Rev. B 58, 13870 (1998) 3. D. Tasis, N. Tagmatarchis, A. Bianco, M. Prato, Chem. Rev. 106,
1105 (2006)
4. P.X. Hou, S. Bai, Q.H. Yang, C. Liu, H.M. Cheng, Carbon 40, 81 (2002)
5. T. Thundat, Nat. Nanotechnol. 3, 133 (2008)
6. S. Polarz, A. Roy, M. Lehmann, M. Driess, F.E. Kruis, A. Hoffmann, P. Zimmer, Adv. Funct. Mater. 17, 1385 (2007) 7. Q. Ma, Z. Ye, H. He, L. Zhu, J. Huang, Y. Zhang, Scr. Mater. 58,
21 (2008)
8. D.H.W. Li, J.C. Lam, C.C.S. Lau, T.W. Huan, Renew. Energ. 29, 921 (2004)
9. J.H. Loh, Med. Device Technol. 10, 24 (1999)
10. M. Ozdemir, H. Sadikoglu, Trends Food Sci. Technol. 9, 159 (1998) 11. S. Iijima, T. Ichihashi, Nature 363, 603 (1993)
12. A. Thess, R. Lee, P. Nikolaev, H. Dai, P. Petit, J. Robert, Science 273, 483 (1996)
13. H. Li, C. Shi, X. Du, C. He, J. Li, N. Zhao, Mater. Lett. 62, 1472 (2008)
14. S.R. Jian, Y.T. Chen, C.F. Wang, H.C. Wen, W.M. Chiu, C.S. Yang, Nanoscale Res. Lett. 3, 230 (2008)
15. C.Y. Zai, X.D. Bai, E.G. Wang, Appl. Phys. Lett. 81, 1690 (2002) 16. E. Ando, M. Miyazaki, Thin Solid Films 516, 4574 (2008) 17. M. Sveningsson, R.E. Morjan, O.A. Nerushev, Y. Sato, J.
Ba¨ckstro¨m, E.E.B. Campbell, F. Rohumund, Appl. Phys. A 73, 409 (2001)
18. D.S. Knight, W.B. White, J. Mater. Res. 4, 385 (1989) 19. M. Yoshikawa, G. Katagiri, H. Ishida, A. Ishitani, J. Appl. Phys.
64, 6464 (1988)
20. T. Young, Philos. Trans. R. Soc. Lond. 95, 65 (1805) 21. C.P. Ho, H. Yasuda, J. Appl. Polym. Sci. 39, 1541 (1990) 22. T. Yasuda, T. Okuno, K. Yoshida, H. Yasuda, J. Sci. Polym.
Chem. Ed. 26, 1781 (1988)
23. B.N. Khare, P. Wilhite, R.C. Quinn, B. Chen, R.H. Schingler, B. Tran, H. Imanaka, C.R. So, C.W. Bauschlicher, M. Meyyappan, J. Phys. Chem. B 108, 8166 (2004)
24. P. Chevallier, M. Castonguay, S. Turgeon, N. Dubrulle, D. Mantovani, P.H. McBreen, J.C. Wittmann, G. Laroche, J. Phys. Chem. B 105, 12490 (2001)
25. R. Soto, P. Gonza´lez, F. Lusquin˜os, J. Pou, B. Leo´n, M. Pe´rez-Amor, Carbon 36, 781 (1998)
26. J.Y. Yook, J. Jun, S. Kwak, Appl. Surf. Sci. 256, 6941 (2010) 27. M. Kra´l’, A. Ogino, M. Nagatsu, Jpn. J. Appl. Phys. 47, 7346
(2008)
28. B. Stoehr, H.P. Boehm, R. Schloegl, Carbon 29, 707 (1991) 29. J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, in
Hand-book of X-ray Photoelectron Spectroscopy, Physical Electronics Division (Perkin-Elmer Corporation, Eden Prairie, MN, 1992) 30. R. Hauert, A. Glisenti, S. Metin, Thin Solid Films 268, 22 (1995) Table 5 N/C and O/C ratios, for MWCNTs samples with different
durations of treatment Samples N/C O/C Untreated MWCNTs 0 0.030 C-NH3-1 (1 min) 0.029 0.090 C-NH3-3 (3 min) 0.036 0.177 C-NH3-5 (5 min) 0.048 0.265
