FULLERENES, NANOTUBES,AND

CARBON NANOSTRUCTURES

Volume 14, Numbers 2&3, 2006

Proceedings of the7TH BIENNIAL INTERNATIONAL WORKSHOP"FULLERENES AND ATOMIC CLUSTERS"

Guest Editors:VLADISLAV LEMANOVALEXANDER VUL 'Ioffe Physico-Technical Institute

JUNE 27 JULY 01, 2005

St. PETERSBURG, RUSSIA

St. Petersburg, Russia

–

PROCEEDINGS OF THE 7TH BIENNIALINTERNATIONAL WORKSHOP“FULLERENES AND ATOMIC

CLUSTERS”

JUNE 27–JULY 1, 2005St. PETERSBURG, RUSSIA

Guest Editors:

VLADISLAV LEMANOVALEXANDER VUL’

Ioffe Physico-Technical InstituteSt. Petersburg, Russia

This is a special issue of

Fullerenes, Nanotubes and Carbon Nanostructures

Volume 14, Numbers 2&3, 2006

Fluorination of Carbon Nanostructures andTheir Comparative Investigation by XPS

and XAES Spectroscopy

A. P. Dementjev, A. V. Eletskii, and K. I. Maslakov

RRC “Kurchatov Institute,” Moscow, Russia

E. G. Rakov

D.I. Mendeleev University of Chemical Technology of Russia,

Moscow, Russia

V. F. SukhoverhovN.S. Kurnakov Institute of Inorganic and General Chemistry RAS,

Moscow, Russia

A. V. Naumkin

Institute of Elemento-Organic Compounds, Russian Academy of

Sciences, Moscow, Russia

Abstract: X-ray photoelectron (XPS) and Auger electron (XAES) spectroscopies have

been used to investigate fluorine interaction with surfaces of nanocarbons. MWCNT,

SWCNT and Fiber were fluorinated in F2 and HOPG in ClF3 atmosphere at room temp-

erature. The performed measurements imply that the surface of the samples is inert to

the environmental before and after fluorination. The XPS spectra show that MWCNT

and Fiber after fluorination have similar composition, which includes CF, CF2 and

CF3 states, but MWCNT includes CF4 state as well. During defluorination of the fluori-

nated HOPG sample with initial C-F bonds the formation of CF2 and CF3 states has been

observed. XAES spectra show the absence of C-F bonding in the outer layers of fluori-

nated HOPG, MWCNT and Fiber, while C-F bonding exists in SWCNT.

Keywords: Nanocarbon, HOPG, fluorination, XPS, Auger

Address correspondence to A. P. Dementjev, RRC “Kurchatov Institute,” 1,

Kurchatov sq., Moscow 123182, Russia. E-mail: a.dem@ru.net

Fullerenes, Nanotubes, and Carbon Nanostructures, 14: 287–296, 2006

Copyright # Taylor & Francis Group, LLC

ISSN 1536-383X print/1536-4046 online

DOI: 10.1080/15363830600663990

287

INTRODUCTION

Fluorination of carbon nanotubes is one of the ways for deep functionalization

and opens both great possibilities to replace the fluorine atoms with various

functional groups and develop nanotube chemistry (1–3) for practical appli-

cations (4–6). Understanding the chemistry of carbon nanostructural

materials is a crucial step towards their ultimate practical use. X-ray photo-

electron spectroscopy (XPS) has been widely used to investigate the surface

chemical states of carbon atoms after fluorination of carbon black, highly

oriented pyrolytic graphite (HOPG), single-wall carbon nanotubes

(SWCNT), multi-wall carbon nanotubes (MWCNT) and carbon nanofiber

(CNF) (7–17). It has been found that CF, CF, CF2, CF3 groups may form

during fluorination of the above-mentioned materials. Both the number and

relative intensities of the functional groups depend on the method of fluorina-

tion. Unfortunately, to our knowledge, there are no data on depth distribution

in multilayered structures. A partial interest is the chemical state of the carbon

atoms in the top layer of any carbon nanostructure. It is supposed that the only

top layer of graphite is involved in the fluorination reaction at low tempera-

tures, and that the outer fluorinated surface hinders further migration of

fluorine to deeper layers (16).

The information depth (ID) of C 1s XPS spectroscopy with Al Ka exci-

tation is about 10–12 monolayers (18). So the chemical states of carbon atoms

in the top layer can be hardly identified using these excitation sources. Another

problem concerns the possibility for detecting contaminations of the surface

with adventitious carbon (AC). This problem is related to the question: Are

the surfaces of the fluorinated carbon nanostructures inert to the environ-

mental? Since ID of C KVV Auger spectroscopy in graphite is about two

monolayers (18), these problems can be resolved using a combination of

XPS and X-ray Auger electron spectroscopy (XAES). The present paper

describes an approach to resolve the above-mentioned problems and

presents some experimental data.

EXPERIMENTAL

CNF used in this study were prepared by pyrolysis of methane using

Ni-containing catalyst (5). CNF have metal (Ni, Fe, Co) particles on the

tips, which are washed out with weak hydrochloric acid solution. MWCNT

were synthesized by pyrolysis of methane at an elevated temperature with

the presence of a metal catalyst. SWCNT of 1.2–1.5 nm in diameter bound

into bundles of about 10 nm in diameter were produced by the standard elec-

trical arc method with the presence of Ni/Cr catalyst (6). Both carbon

nanotubes and CNF were fluorinated with 1 atm F2 gas at room temperature

for 48 hours. The HOPG sample with dimensions 10 � 10 � 1 mm3 was

fluorinated with 1 atm ClF3 gas at room temperature for 60 hours. It was

A. P. Dementjev et al.288

found that the thickness of the sample increased from 1 mm up to 15 mm as a

result of fluorination. The central part of the fluorinated sample remained hard

after fluorination. We analyzed the top layer (region I), the middle of expanded

part (region II) and the central hard part (region III) of the HOPG sample. The

fluorinated HOPG sample was defluorinated at 2708C in the preparation

chamber of the spectrometer without contact with environment before XPS

study.

The experimental data were obtained using a MK II VG Scientific spec-

trometer. The C 1s and F 1s photoemission and C KVV Auger emission were

excited using an Al Ka source with photon energy of 1486.6 eV. The pressure

in both the analytical and preparation chamber was kept at 5 � 10210 mbar.

Spectra were collected in the constant analyzer energy mode, with a pass

energy of 20 eV and 0.1 eV step size. The energy resolution determined as

full width at half-maximum of the Au 4f7/2 photoelectron line is 1.1 eV.

We used an energy of 284.4 eV for charge reference of HOPG (19). The

value of 285.0 eV was used for charge referencing of non-fluorinated

nanotubes and CNF C 1s spectra.

RESULTS AND DISCUSSION

It is well known that metal and oxide surfaces adsorb carbon-containing

molecules or in other words adventitious carbon (AC). It is quite easy to

recognize them at metal or oxide surfaces, while in the case of carbon nano-

materials this may be a serious problem. To determine the degree of contami-

nation of pure and fluorinated carbon surface by environmental AC we used

XAES, which is highly sensitive to the state of the outer surface layer. We

studied the C KVV spectra for AC at surfaces of different metals and their

oxides and found that the C KVV line shape is independent of substrate,

with carbon atoms having sp3–bonds (20). Comparison between the

C KVV spectra of “reference AC” and studied samples (Figure 1) allows us

to conclude that their surfaces are free of AC.

The survey XPS spectra of SWCNT, MWCNT and Fiber before and after

fluorination show the presence of oxygen and metal impurities (about 1%).

Figure 2 shows the C 1s XPS spectra for fluorinated SWCNT, MWCNT

and Fiber. The C 1s spectra of the MWCNT and Fiber have similar line

shapes excluding for the high-energy region. Fluorine content in MWCNT

and Fiber is practically the same. The C 1s XPS spectra of SWCNT,

MWCNT and Fiber were deconvoluted. The results are submitted in

Table 1. The carbon states were identified in accordance with the relevant

data for fluorinated polymers and fluorocarbons (17, 21). However, it was

difficult to find correlation between the chemical states of fluorinated

carbons and fluorine because of approximately equal intensities of CF, CF2,

and CF3 states in the C 1s spectrum while the F 1s spectrum has only one

state with an enhanced intensity. At the same time the peaks with the

Fluorination of Carbon Nanostructures 289

largest binding energies in the C 1s and F 1s spectra may be assigned to CF4

gas-like phase intercalated into MWCNT. The similar states were observed in

fluorographite (17).

Figure 3 shows F 1s XPS spectra for fluorinated SWCNT, MWCNT,

HOPG and Fiber. The widths of SWCNT, MWCNT and Fiber spectra are

considerably differ from that of HOPG. Obviously these effects are caused

by significant differences in a chemical environments of F-atoms.

Figure 1. C KVV spectra for pure HOPG, SWCNT pure and fluorinated MWCNT

and reference AC. Dissimilarity between the spectra of the samples and that of refer-

ence AC allows one to conclude that the samples do not contain AC.

Figure 2. C 1s XPS spectra for fluorinated Fiber, MWCNT and SWCNT.

A. P. Dementjev et al.290

In distinction to fluorinated MWCNT, in fluorinated SWCNT only one

chemical state of fluorine appeared in the F 1s spectrum, which was identified

as the singlet state. A similar spectrum was observed by Pehrsson et al. (7),

however the most intense state was assigned to CF state.

Figure 4 shows HOPG C 1s spectra after fluorination. The C 1s spectrum

of HOPG has two components as in (11) with a peak separation of about

2.5 eV. These peaks are assigned to C-CF and C-F bonds. This separation is

more than that observed in polymers between peaks corresponding to C-CF

and C-F bonds (21). The low-energy peak was shifted to 285.75 eV in accord-

ance with energy for C-CF state in poly(vinyl fluoride) (21). Measurements

were carried out on three sections of the extended sample. The relation of

C-CF/C-F for those sections has changed insignificantly. It testifies that fluor-

ination of the top layer did not retard penetration of F-atoms from a surface

into deeper layers.

Table 1. Deconvolution parameters of C 1s spectra for MWCNT, Fiber and SWCNT

after fluorinaton

Component

Eb(C 1s),

eV

FWHM,

eV

Relative

intensity

Functional

group

MWCNTþ F

C1 285.6 1.2 4.4 C-CF

C2 286.9 1.5 1.5 C-CF2

C3 288.2 1.5 1.8 CF

C4 289.6 1.5 4.2 CF-CFn

C5 291.4 1.6 2.2 CF2

C6 293.3 1.2 1.7 CF3

C7 295.9 2.6 2.0 CF4

Fiberþ F

C1 285.0 1.2 14.1 C-C

C2 285.6 1.4 3.2 C-CF

C3 286.9 1.4 3.7 C-CF2

C4 288.3 1.4 2.9 CF

C5 289.6 1.5 5.2 CF-CFn

C6 291.4 1.6 2.9 CF2

C7 293.2 1.7 1.2 CF3

SWCNTþ F

C1 284.8 1.7 0.72 C-C

C2 285.6 1.7 1.1 C-CF

C3 286.9 1.6 2.5 C-CF2

C4 288.2 1.8 2.8 CF

C5 289.6 1.7 4.0 CF-CFn

C6 291.4 1.8 3.6 CF2

C7 293.3 1.8 1.6 CF3

C8 295.9 1.3 0.25 CF4

Fluorination of Carbon Nanostructures 291

The C 1s spectra for HOPG after defluorination are presented in Figures 5

and 6. As a consequence of defluorination, one can observe the presence of

C-C-C state as well as CF2 and CF3 states of carbon atoms.

The C KVV Auger spectra in Figure 7 allow us to obtain information

about modification of chemical state of C-atoms on the HOPG surface after

defluorination. After heating for 4 minutes, the C KVV spectrum was not

changed and after 10 minutes it was transformed to HOPG-like. Such

spectrum we observed on surface of HOPG after ion sputtering. Apparently

the process of defluorination results in amorphous carbon layer on the surface.

Figure 3. F 1s XPS spectra for fluorinated Fiber, MWCNT, SWCNT and HOPG.

Figure 4. C 1s XPS spectra for 3 sections of fluorinated HOPG as shown in framing.

A. P. Dementjev et al.292

The carbon atoms in HOPG, SWCNT, MWCNT are in an sp2-configuration.

Their C KVV Auger spectra have similar line shape excluding for the high

energy part, which reflects, on our opinion, the interlayer interaction which is

present in HOPG and MWCNT and absent in SWCNT and C60 (22). The

observed distinction of C KVV spectra for HOPG and MWCNT may be due

to the curvature of graphene sheets in nanotubes.

Figure 5. C 1s XPS spectra for HOPG after in situ defluorination at 2708C during

4 min.

Figure 6. C 1s XPS spectra for HOPG after in situ defluorination at 2708C during

10 min.

Fluorination of Carbon Nanostructures 293

In view of these data it is probably that coincidence of fluorinated

MWCNT spectrum with that for non-fluorinated SWCNT (Figure 1) and

HOPG spectrum after fluorination (Figure 8) is caused by disappearance of

interlayer interaction.

The absence of strong shift for fluorinated HOPG and MWCNT spectra

(Figure 8) in comparison with the shift for SWCNT spectrum allows

Figure 7. Modification of C KVV Auger spectra for HOPG during fluorination and

defluorination.

Figure 8. C KVV Auger spectra for fluorinated MWCNT, SWCNT and HOPG, and

pure HOPG and MWCNT. The spectra of Fiber (not shown) before and after fluorina-

tion coincide with that of MWCNT. The spectrum of MWCNT after fluorination

coincides with that of SWCNT.

A. P. Dementjev et al.294

speculation that the top layer of HOPG and MWCNT do not interact with

F-atoms. Obviously the F-atoms are in deeper layers and only change the

interaction between the uppermost layers.

CONCLUSIONS

The performed measurements imply that the HOPG, SWCNT, MWCNT and

Fiber samples before and after fluorination are inert to the environmental. The

XPS spectra show that MWCNT and Fiber after fluorination have similar

composition, which includes CF, CF2 and CF3 states, but MWCNT also

include CF4. Fluorine atoms have four chemical states in fluorinated

MWCNT and one state in HOPG, SWCNT and Fiber. It was observed that

the degree of fluorination in HOPG changes in depth. During defluorination

of the fluorinated HOPG sample with initial C-F states the formation of CF2

and CF3 states has been observed. XAES spectra show the absence of C-F

bonding in the outer layers of fluorinated HOPG, MWCNT and Fiber while

C-F bonding exists in the outer layer SWCNT.

REFERENCES

1. Mickelson, E.T., Huffman, C.B., Rinzler, A.G., Smalley, R.E., Hauge, R.H., andMargrave, J.L. (1998) Fluorination of single-wall carbon nanotubes. Chem.Phys. Lett., 296: 188–194.

2. Boul, P.J., Liu, J., Mickelson, E.T., Huffman, C.B., Ericson, L.M., Chiang, I.W.,Smith, K.A., Colbert, D.T., Hauge, R.H., Margrave, J.L., and Smalley, R.E.(1999) Reversible sidewall functionalization of Buckytubes. Chem. Phys. Lett.,310: 367–372.

3. Mickelson, E.T., Chiang, I.W., Zimmerman, J.L., Boul, P.J., Lozano, J., Liu, J.,Smalley, R.E., Hauge, R.H., and Margrave, J.L. (1999) Solvation of fluorinatedsingle-wall carbon nanotubes in alcohol solvents. J. Phys. Chem. B, 103:4318–4322.

4. Peng, H., Gu, Z., Yang, J., Zimmerman, J.L., Willis, P.A., Bronilowski, M.J.,Smalley, R.E., Hauge, R.H., and Margrave, J.L. (2001) Fluorotubes as cathodesin lithium electrochemical cells. Nano Lett., 1: 625–629.

5. Rakov, E.G., Blinov, S.N., Ivanov, I.G., Rakova, E.B., and Digurov, N.A. (2004)Continuous process for obtaining carbon nanofibers. Russ. J. Appl. Chem., 77 (2):187–191.

6. Bezmelnitsyn, V.N., Domantovskii, A.G., Eletskii, A.V., Obraszova, E.D.,Pernbaum, A.G., Prichod’ko, K.E., and Terekhov, S.V. (2002) Preparation ofsingle-walled nanotubes with the help of a Ni/Cr-based catalyst. Russ. J. Phys.Solids, 44 (4): 656–658.

7. Pehrsson, P.E., Zhao, W., Baldwin, J.W., Song, C., Liu, J., Kooi, S., and Zheng, B.(2003) Thermal fluorination and annealing of single-wall carbon nanotubes.J. Phys. Chem. B, 107: 5690–5695.

Fluorination of Carbon Nanostructures 295

8. Kawasaki, S., Komatsu, K., Okino, F., Touhara, H., and Kataura, H. (2004)Fluorination of open- and closed-end single-walled carbon nanotubes PCCP.Phys. Chem. Chem. Phys., 6: 1769–1772.

9. Marcoux, P.R., Schreiber, J., Batail, P., Lefrant, S., Renouard, J., Jacob, G.,Albertini, D., and Mevellec, J.-Y. (2002) A spectroscopic study of the fluorinationand defluorination reactions on single-walled carbon nanotubes PCCP. Phys.Chem. Chem. Phys., 4: 2278–2285.

10. Hattori, Y., Watanabe, Y., Kawasaki, S., Okino, F., Pradhan, B.K., Kyotani, T.,Tomita, A., and Touhara, H. (1999) Carbon-alloying of the rear surfaces ofnanotubes by direct fluorination. Carbon, 37: 1033–1038.

11. Okotrub, A.V., Yudanov, N.F., Chuvilin, A.L., Asanov, I.P., Shubin, Yu.V.,Bulusheva, L.G., Gusel’nikov, A.V., and Fyodorov, I.S. (2000) Fluorinated cagemultiwall carbon nanoparticles. Chem. Phys. Lett., 322: 231–236.

12. Bismark, A., Tahhan, R., Springer, J., Schulz, A., Klapotke, T.M., Zell, H., andMichaeli, W. (1997) Influence of fluorine on the properties of carbon fibers.J. Fluorine Chem., 84: 123–134.

13. Tressaud, A., Chambon, M., Gupta, V., Flandrois, S., and Bahl, O.P. (1995)Fluorine-intercalated carbon fibers. Carbon, 33: 1339–1345.

14. Kudin, K.N., Bettinger, H.F., and Scuseria, G.E. (2001) Fluoriented single-wallcarbon nanotubes. Phys. Rev. B, 63: 045413–8.

15. Lee, Y.-S., Cho, T.H., Lee, B.K., Rho, J.S., An, K.H., and Lee, Y.H. (2003)Surface properties of fluorinated single-walled carbon nanotubes. J. Fluor.Chem., 120: 99–104.

16. Watanabe, N., Touahra, H., Bartlett, N., and Mallouk, T. (1985) Fluorine intercala-tion compounds of graphite. In Inorganic Solid Fluorides: Chemistry and Physics;Hagenmuller, P. (ed.); Academic Press: New York, 331–369.

17. Nanse, G., Papirer, E., Fioux, P., Moguet, F., and Tressaud, A. (1997) Fluorinationof carbon blacks: An X-ray photoelectron spectroscopy study: I. A literaturereview of XPS studies of fluorinated carbons. XPS investigation of somerefrences compounds. Carbon, 35: 175–194.

18. Powell, C.J. and Jablonski, A. (2000) NIST Electron Inelastic-Mean-Free-PathDatabase, Version 1.1; National Institute of Standards and Technology:Gaithersburg, MD.

19. C.D. Wagner, A.V., Naumkin, A., Kraut-Vass, J.W., Allison, C.J., Powell, J.R.,and Rumble Jr. (2001) NIST Standard Reference Database 20, Version 3.1(Web Version).

20. Dementjev, A.P. (2001) Electron spectroscopy characterization of carbon surfaces.New Diamond Frontier Carbon Technol., 11 (1): 37–51.

21. Beamson, G. and Briggs, D. (1992) High Resolution XPS of Organic Polymers:The Scienta ESCA 300 Database; Wiley: Chichester.

22. Dementjev, A.P., Maslakov, K.I., and Naumkin, A.V. (2005) Relation between theC KVV Auger line shape and layered structure of graphite. Appl. Surf. Sci., 245:128–134.

A. P. Dementjev et al.296