www.elsevier.com/locate/diamond

Diamond & Related Materia

FTIR studies of nitrogen doped carbon nanotubes

Abha Misra, Pawan K. Tyagi, M.K. Singh, D.S. Misra *

Department of Physics, Indian Institute of Technology, Bombay, Mumbai-400076, India

Available online 13 October 2005

Abstract

Purified and defect free carbon nanotubes have great potential for applications in electronic, polymer composites and biological

sciences. The removal of impurities (carbon nanoparticles and amorphous carbon) is an important step before the CNT applications can be

realized. We report the results of FTIR and TGA/DTA studies of the impurities present in the carbon nanotubes. The multiwalled CNTs

were grown using Microwave Plasma Chemical Vapor Deposition (MPCVD) technique. Fourier transform infrared (FTIR) spectroscopy

was carried out in the range of 400–4000 cm�1 to study the attachment of the impurities on carbon nanotubes. FTIR spectra of the as-

grown MWCNTs show dominant peaks at 1026, 1250, 1372, 1445, 1736, 2362, 2851, 2925 cm�1 that are identified as Si–O, C–N, N–

CH3, CNT, C–O, and C–Hx respectively. The peaks are sharp and intense showing the chemisorption nature of the dipole bond. The

intensity of the peaks due to N–CH3, C–N and C–H reduces after annealing and the peaks vanish on annealing at high temperature (900

-C). The presence of C–N peak may imply the doping of the CNTs with N in substitution mode. TGA/DTA measurements, carried out

under argon flow, show that the dominant weight loss of the sample occurs in the temperature range 400–600 -C corresponding to the

removal of the impurities and amorphous carbon.

D 2005 Elsevier B.V. All rights reserved.

Keywords: MWCNTs; FTIR spectroscopy; TGA/DTA

1. Introduction

Carbon nanotubes [1] (CNTs) are found attractive on

account of their potential for applications in hydrogen

storage, nanoscale devices and sensor [2–6]. High quality

and well-aligned carbon nanotubes are essential for the

applications in the field of nanoelectronics and many of

these applications are dependent upon the chirality and

diameter of CNTs [7]. The nanotubes can be either metallic

or semiconducting depending upon their diameter and

chirality. The presence of defects and impurities that are

electronically and chemically active can change these

properties. Therefore, the control of the defects and

impurities has become important for many applications of

CNTs. Doping is a practical way to tailor the electrical

properties of the CNTs. The nitrogen doped CNT shows n-

type semiconducting behavior regardless of tube chirality

[8].

0925-9635/$ - see front matter D 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.diamond.2005.08.013

* Corresponding author.

E-mail address: nirdesh@phy.iitb.ac.in (D.S. Misra).

A great deal of attention has been devoted to the N-doped

CNTs. Several groups have produced carbon nanotubes

containing nitrogen or nitrogen and boron. The specific

interest in CNx tubes is due to the fact that substituted N

makes exclusively semiconducting tubes. Many groups have

proposed that the electronic structure of N doped pyridine-

like, pyrolic, and graphite like structure (Carbon atom has

been replaced by N atom in graphite layers) is similar to

triple-bonded CN [9–11]. Another group has proposed that N

atoms might exist mainly inside the inner core of CNTs [12].

In this paper we investigate the nitrogen doped multiwalled

carbon nanotubes using Fourier transform infrared (FTIR)

spectroscopy and thermo gravimetric analysis (TGA).

2. Experimental

CNTs were grown onto nickel electroplated copper foil

substrates. The commercial grade copper foils (purity, 98%),

polycrystalline in nature, were used without any special pre-

treatment to improve their smoothness. The electroplating of

ls 15 (2006) 385 – 388

A. Misra et al. / Diamond & Related Materials 15 (2006) 385–388386

Ni on copper foil was performed in an electroplating bath.

The copper substrate (1 cm�1 cm�0.1 mm) was used as

cathode and a thin platinum wire was used as the anode.

Commercial grade nickel sulfate (NiSO4I7H2O) (275 g/l)

and nickel chloride (NiCl2I6H2O) (60 g/l) were used as

electrolytes in a dc electroplating process. The bath

temperature and current were maintained at 60 -C and 80

mA, respectively. The nickel layers having a thickness of

5–10 Am were treated in the plasma of ammonia (NH3) gas

for 10 s. Ammonia plasma treatment was performed for 10 s

in the temperature range of 300–750 -C at 10 Torr for the

formation of nano-sized nickel catalyst, which was the

precursor for subsequent CNTs growth using chemical

vapor deposition (CVD). The flow rate of ammonia was

kept at 180 sccm. Methane (CH4) and hydrogen (H2) were

used with respective flow rates of 6 and 40 sccm. The

temperature of the substrate was maintained at 820T20 -Cat deposition pressure ;40 Torr. The deposition time was 5

min. Nitrogen doping was achieved by adding ammonia gas

with flow rate of 180 sccm in the gas mixture at the time of

deposition.

Thermal Gravimetric Analysis (TGA) were carried out

on Mettler–Toledo (TGA/SDTA851) thermal analyzer

under argon flow at the heating rate of 5-/min to obtain

information on the decomposition and the burning proper-

ties of carbon nanotubes and impurities present in it. The

temperature of the sample was varied from room temper-

Fig. 1. (a) SEM micrograph of a sample of multiwalled carbon nanotubes deposi

TEM images of the multiwalled carbon nanotubes free of encapsulated and wit

Transmission Electron Microscopy (HRTEM) image of graphitic walls of a typic

ature to 900 -C. The Fourier transform infrared (FTIR)

spectrometer were recorded on Nicolet Magna 550 FT-IR

Spectrum. The samples for FTIR studies were prepared by

suspending approximately 6 mg of MWCNT materials in

¨15 ml isopropyl alcohol by sonication with an ultra sonic

probe for several minutes. One drop of this solution was

sprayed onto silicon wafer and a uniform thin MWNT film

on the IR transparent silicon substrate was thus obtained.

FTIR studies were carried out in the range of 400–4000

cm�1 in the absorbance mode. The FTIR results with the

support of TGA results give a reasonably good picture of the

attachment on the carbon nanotubes.

3. Results and discussion

Fig. 1(a) shows the SEM micrograph of high density

MWCNTs grown on nickel electroplated copper substrate

after ammonia plasma treatment for 2 min. The average

length and diameter of the tubes are 100 Am and 20 nm

respectively. Fig. 1 (b–c) show TEM images of the

multiwalled carbon nanotubes free of encapsulated and with

encapsulated catalytic nickel particles, respectively. The

inset in Fig. 1(c) shows that the inner and outer diameter of

the CNT is 12.84 and 18.43 nm respectively and the tips of

the CNTs are closed. Fig. 1(d) is the High-Resolution

Transmission Electron Microscopy (HRTEM) image of

ted on a copper substrate. The high density of tubes is noteworthy. (b)– (c)

h encapsulated catalytic nickel particles, respectively. (d) High-Resolution

al MWCNT.

-200 200 400 600 800 1000 1200-20

0

20

40

60

80

100

120

DTA

TGA

Temperature (0C)

Wei

gh

t Per

cen

t

0

Fig. 3. TG-DTA results for MWNTs.

A. Misra et al. / Diamond & Related Materials 15 (2006) 385–388 387

graphitic walls of a MWCNT. The separation between two

consecutive walls is 3.22 A.

FTIR is used to characterize the functional elements

absorbed by carbon nanotubes. Fig. 2 (a–d) shows the FTIR

spectra in the range of 400–4000 cm�1 a) of as-grown

tubes, b) of the tubes collected after TGA at 400 -C, c) ofthe tubes annealed at 600 -C, and d) of the tubes collected

after TGA at 900 -C. FTIR of a) shows dominant peaks at

1026, 1250, 1372, 1445, 1736, 2362, 2851, 2925 cm�1

which corresponds to Si–O, C–N, N–CH3, CNT, C–O,

C–Hx respectively. The infra-red absorbance at 1026 cm�1

is consistent with Si–O stretching vibrations due to slightly

different concentrations in the native oxide layer of the

silicon before and after coating with MWCNTs film [13].

Out of the above identified bonds, the presence of C–N and

N–CH3 bonds at 1250 and 1372 cm�1 is most interesting.

The strong peaks at 1250 and 1372 cm�1 are consistent with

C–N and N–CH3 stretching vibrations attributed by Choi et

al. [14] to the presence of intercalated N atoms between the

graphite layers at the inner part of the nanotube walls.

However, in our opinion the intercalated nitrogen atoms in

between the graphite walls may not be strongly IR active.

The halfwidths of the C–N and N–CH3 peaks are 36.72 and

18.53 cm�1 respectively, and suggest that the chemisorption

process may be dominant implying chemical bonding

between carbon and nitrogen atoms. It has been pointed

out [15–18] that substituting a N atom in place of a C atom

in a sp2 bonded carbon network will induce strong IR

activity; consequently the absorption in the 1200–1600

cm�1 region is expected if the N atoms are bonded into the

carbon network. We therefore strongly believe that N

doping of graphene sheets may be taking place and a C–

N bond identical to the sp3 bonded carbon nitride sample

may be forming. Features at 1445 and 1736 cm�1 attributed

to MWCNT vibrational modes are also apparent [19]. The

peak at 2362 cm�1 corresponds to the C–O bonds and the

features between 2851 and 2925 cm�1 are consistent with

500 1000 1500 2000 2500 3000 3500

CNT

CNT

N-CH3

C-N

Si-O

(d)(c)

(b)

(a) C-Hx

Ab

sorb

ance

(ar

b. u

nit

s)

Wave number (cm-1)

Fig. 2. FTIR spectra of (a) as-grown MWNTs, (b) after TGA at 400 -C, (c)

after annealing at 600 -C, and (d) after TGA at 900 -C.

C–Hx stretching vibrations of chemisorbed hydrogen of

various types presents in all carbon films [20,21].

The intensity of the peaks due to N–CH3, C–N and C–

H are gradually suppressed, after annealing at 400 -C. Thesuppression of the peaks would be related to the gradual

removal of nitrogen and hydrogen bonded species as

impurity. The bonding of N to carbon atom in graphene

sheet will give rise to a defect in structure, which would get

annealed, at high temperature. This results in the reduction

of the intensity of the IR peaks with annealing. Further

annealing of the sample at 600 -C does not induce any

appreciable change. It is interesting to note that these peaks

vanish on annealing at higher temperature (900 -C)implying that at high temperature the defects inside the

tubes are getting strongly mobile leading to the destruction

of the attachments along with the tubes.

TGA/DTA curve in Fig. 3 reveals a small weight loss due

to water removal around 80 -C. The dominant weight loss

steps are due to the removal of carbon materials and also

due to the decomposition of nanotubes [22–24] taking place

in the temperature range between 400 and 600 -C. The

weight loss starts near 400 -C and the process completes by

600 -C. This result agrees with the FTIR spectra in Fig.

1(b–c) discussed above where we have shown a significant

decrease in the intensity of IR peaks upon annealing to 400

-C. The MWCNTs are completely destroyed at 900 -C,suggesting that at this temperature amorphous carbon as

well as MWCNTs converts to gaseous form. The complete

removal of all the dominant peaks from the FTIR spectra in

Fig. 1(d) confirms the burning of MWCNTs at 900 -Cresulting in the change of the color of the sample.

4. Conclusion

FTIR technique gives information about gaseous ele-

ments attached to the tube. Nitrogen attachment is evident

from strong and intense IR peaks at 1250 and 1372 cm�1.

From the intensity and halfwidth of these peaks we

conclude that the N atom is incorporated at substitutional

A. Misra et al. / Diamond & Related Materials 15 (2006) 385–388388

sites in the carbon network. TGA/DTA has been used to

study the weight loss of carbon nanotube as a function of

temperature. We found that the dominant weight loss

occurs the between 400 and 600 -C range and agrees well

with disappearance of IR peaks corresponding to N-

attachment. TGA can be used to get the information about

the weight loss of the sample.

Acknowledgements

The authors would like to thank Dr. P.V. Satyam of

Institute of Physics Sachivalaya Marg, Bhubaneshawar for

high-resolution transmission electron microscopy (HRTEM)

measurements.

References

[1] S. Ijima, Nature (London) 354 (1991) 56.

[2] S.J. Tans, C. Dekker, Nature (London) 404 (2000) 834.

[3] J.T. Hu, O.Y. Min, P.D. Yang, C.M. Laiber, Nature (London) 399

(1999) 48.

[4] R.D. Antonov, A.T. Johnson, Phys. Rev. Lett. 83 (1999) 3274.

[5] M.S. Fuhrer, J. Nygard, L. Shih, M. Forero, Y.-G. Yoon, M.S.C.

Mazzoni, H.J. Choi, J. Ihm, S.G. Louie, A. Zettl, P.L. Mceuen,

Science 288 (2000) 494.

[6] J. Kong, N.R. Franklin, C. Zhou, M.G. Chapline, S. Peng, K. Cho, H.

Dai, Science 287 (2000) 622.

[7] A. Hassanien, M. Tokumoto, Y. Kumazawa, H. Kataura, Y. Maniwa,

S. Suzuki, Y. Achida, Appl. Phys. Lett. 73 (1998) 3839.

[8] R. Czerw, et al., Nano Lett. 1 (2001) 457.

[9] J. Casanovas, J.M. Ricart, J. Rubio, F. Illas, J.M. Jimenez-Mateos, J.

Am. Chem. Soc. 118 (1996) 8071.

[10] M.C. dos Santos, F. Alvarez, Phys. Rev., B 58 (1998) 13918.

[11] I. Shimoyama, G. Wu, T. Sekiguchi, Y. Baba, Phys. Rev., B 62 (2000)

R605.

[12] M. Terrones, R. Kamalakaran, T. Seeger, M. Ruhle, Chem. Commun.

(Cambridge) (2000) 2335.

[13] S.L. Sung, C.H. Tseng, F.K. Chiang, X.J. Guo, X.W. Liu, Thin Solid

Films 340 (1999) 169.

[14] Hyun Chul Choi, Seung Yong Bae, Jeunghee Park, Appl. Phys. Lett.

85 (2004) 5742.

[15] M.R. Wixom, J. Am. Ceram. Soc. 73 (1990) 1973.

[16] J.H. Kaufman, S. Metin, D.D. Saperstein, Phys. Rev., B 39 (1989)

13053.

[17] Y.K. Yap, S. Kida, T. Aoyama, Y. Mori, T. Sasaki, Appl. Phys. Lett. 73

(1998) 915.

[18] S.H. Lai, et al., Thin Solid Films 444 (2003) 38.

[19] A.C. Dillon, T. Gennett, J.L. Alleman, K.M. Jones, Proceedings of the

2000 DOE/NREL Hydrogen Program Review, 2000 (May 8–10).

[20] G.-Q. Yu, et al., Diamond Relat. Mater. 11 (2002) 1633.

[21] J. Ristein, R.T. Stief, L. Ley, W. Beyer, JAP 84 (1998) 3836.

[22] D.G. McCulloch, E.G. Gerstner, D.R. McKenzie, S. Prawer, R. Lalish,

Phys. Rev., B 52 (1995) 850.

[23] M. Zhang, M. Yudasaka, S. Iijima, Chem. Phys. Lett. 364 (2002) 42.

[24] M. Zhang, M. Yudasaka, S. Bandow, S. Iijima, Chem. Phys. Lett. 369

(2003) 680.