CNOs - Wiley Online Library

DOI: 10.1002/asia.200600426

Reactivity Differences between Carbon Nano Onions (CNOs) Prepared by Different Methods Amit Palkar,[a] Frederic Melin,[a] Claudia M. Cardona,[a] Bevan Elliott,[a] Amit K. Naskar,[b] Danny D. Edie,[b] Amar Kumbhar,[c] and Luis Echegoyen*[a] Abstract: The carbon nanoparticles obtained from either arcing of graphite under water or thermal annealing of nanodiamonds are commonly called carbon nano onions (CNOs), or spherical graphite, as they are made of concentric fullerene cages separated by the same distance as the shells of graphite. A more careful analysis reveals some dramatic differences between the parti-

cles obtained by these two synthetic methods. Physicochemical methods indicate that the CNOs obtained from nanodiamonds (N-CNOs) are smaller and contain more defects than the Keywords: annealing · fullerenes · nanodiamonds · nano onions · nanostructures

Introduction The discovery of fullerenes in 1985 by Kroto et al. marked the beginning of the current carbon nanoscience revolution.[1] The unique all-carbon structure of closed-cage molecules or fullerenes has attracted considerable attention from diverse scientific fields ranging from materials science to medicinal chemistry because of their electronic properties and nanometer dimensions.[2] Carbon nanotubes (CNTs) discovered in the early 1990s have also received significant attention because of their interesting mechanical, chemical, electrical, thermal, and optical properties.[3] Although multi-

[a] A. Palkar, Dr. F. Melin, Dr. C. M. Cardona, B. Elliott, Prof. Dr. L. Echegoyen Department of Chemistry Clemson University, SC 29634 (USA) Fax: (+ 1) 864-656-6613 E-mail: [email protected] [b] Dr. A. K. Naskar, Prof. D. D. Edie Department of Chemical Engineering and Center for Advanced Engineering Fibers and Films 208 Earle Hall, Clemson, SC 29634-0910 (USA) [c] A. Kumbhar Advanced Materials Research Laboratories 91 Technology Dr., Anderson, SC 29625 USA Supporting information for this article is available on the WWW under http://www.chemasianj.org or from the author.

Chem. Asian J. 2007, 2, 625 – 633

CNOs obtained from arcing (ACNOs). These properties explain the enhanced reactivity of the N-CNOs in cycloaddition and oxidation reactions, as well as in reactions involving radicals. Given the easier functionalization of the N-CNOs, they are the most obvious choice for studying the potential applications of these multi-shelled fullerenes.

shelled fullerenes or carbon nano onions (CNOs) were discovered almost simultaneously with CNTs, these equally fascinating molecules have not received much attention yet. Their tribological properties have already been established by NASA to be better than other commonly used graphitic materials.[4] It is possible that their electronic and chemical properties may offer potential applications ranging from optical limiting and photovoltaics to catalysis and energy storage. Since Ugarte first observed CNOs while submitting carbon nanoparticles to an electron beam,[5] two main preparative methods have been described in the literature. Kuznetsov et al. reported the production of highly pure CNOs in high yields from annealing carbon nanodiamond particles at temperatures above 1200 8C.[6] However, this technique requires an oven operating under high vacuum, which is expensive. More recently, Sano et al. described a different method of producing CNOs that involves arcing between two graphite electrodes under water.[7] This method yields CNOs along with CNTs and other graphitic particles as side products. The annealing of the ultradispersed nanodiamond particles (5 nm average) under vacuum produces mainly small CNOs (N-CNOs) with 6–8 shells (5 nm). The arcing of graphite under water, however, leads to the formation of large CNOs (A-CNOs) with diameters in the range of 15– 25 nm (20–30 shells). The extreme insolubility of CNOs has precluded the investigation of their bulk properties in solution. Until now,

D 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

625

FULL PAPERS there have been only two reports of chemical functionalizations of CNOs that lead to modest solubility levels in common solvents.[8, 9] Both of these functionalizations were carried out only on large CNOs, which were produced by the arc-discharge method. Surprisingly, no functionalization of CNOs produced by any other method has ever been reported. N-CNOs have smaller radii and consequently higher curvature than A-CNOs. It is known that fullerene reactivity is partially dictated by the degree of surface strain: the higher the curvature, the greater the reactivity.[10] Therefore, the reactivity of small CNOs is expected to be greater than that of large CNOs. Herein we report a comparison of the physicochemical properties of CNOs obtained by the two methods, as well as the size-dependent differences in reactivity. The first functionalization procedures with CNOs obtained from the annealing of nanodiamond particles are also described.

Results and Discussion The original annealing procedure described by Kuznetsov et al. involved heating the nanodiamond particles under high vacuum.[6] However, in the course of our experiments, we observed that highly pure samples of onions with the same morphology and size could also be obtained by using a slightly positive pressure of helium. This modification led to a more practical synthetic method that eliminates completely the need for high-vacuum systems.

Characterization of the CNOs Transmission Electron Microscopy (TEM) Figure 1 a and b shows high- and low-resolution TEM images of the material collected either on the top or the bottom of the water vessel after arcing of graphite under water and annealing under air for one hour at 400 8C to remove amorphous carbon. The sample contains CNOs with an average diameter of 20 nm (around 25 layers) as well as carbon nanotubes, as Sano et al. also reported.[7] Most of the CNOs show a core size greater than 1 nm, thus suggesting that the nucleus of the onion is a bigger fullerene than C60. Figure 1 c and d shows the TEM images of the material obtained from annealing nanodiamond particles at 1650 8C under helium. As clearly seen from the images, the sample contains exclusively CNOs. These CNOs have an average size of 5 nm and consist of 6–8 shells. This is consistent with the initial nanodiamond particle size employed (average of 5 nm) and is comparable to the CNOs obtained by Kuznetsov et al. under high vacuum.[6] The core of these CNOs also appears to be larger than 1 nm. The CNOs from both samples show interlayer distances of 3.3 G, which corresponds to the distance between graphene sheets. These two samples of CNOs offer an unprecedented opportunity to compare the properties and reactivities of CNOs with contrasting diameters and prepared by entirely different methodologies.

626

www.chemasianj.org

Figure 1. a) High- and b) low-resolution TEM images of CNOs from arcing under water (A-CNOs). c) High- and d) low-resolution TEM images of CNOs from nanodiamonds (N-CNOs). TEM was performed with lacey-carbon grids. Scale bars represent 20 nm and 5 nm in the lowand high-resolution images, respectively.

X-ray Diffraction The interlayer distance was further confirmed by X-ray powder diffraction. As seen in Figure 2, A-CNOs show a d spacing of approximately 3.4 G, similar to the interlayer distance in graphite. In contrast, nanodiamonds have a d spacing of 2.1 G, which becomes 3.4 G after the annealing process (Figure 2). These measurements are consistent with the CNOs obtained from annealing nanodiamonds under high vacuum as previously described.[11]

Figure 2. Powder X-ray diffraction spectra of nanodiamonds (b), N-CNOs (c), and A-CNOs (g).


Chem. Asian J. 2007, 2, 625 – 633

Reactivity Differences of Carbon Nano Onions

NMR Spectroscopy High-resolution magic-angle spinning (HR-MAS) solid-state NMR spectroscopy was performed on both N-CNOs and ACNOs. The comparative spectra obtained (Figure 3) clearly show resonances above 100 ppm, which substantiates the presence of a CACHTUNGRE(sp2) framework in these nanoparticles. Similar broad spectra were reported for nanotubes with comparable chemical shifts.[12]

The enhanced stability of the N-CNOs may be attributed to the fact that they were heated for 1 h (1650 8C) and were subsequently cooled very slowly. The slow cooling process may allow for structural reorganization to a thermodynamically more stable structure. In contrast, the A-CNOs were rapidly cooled as they were produced in the surrounding solvent medium (water). To verify this hypothesis, the A-CNOs were annealed under helium followed by a period of slow cooling. Indeed, the thermal stability of these CNOs significantly increased after annealing at 1650 8C for an hour (Figure 5). Annealing

Figure 3. HR-MAS 13C NMR spectra of A-CNOs (bottom) and N-CNOs (top).

Thermal Gravimetric Analysis (TGA) Thermal degradation under aerobic conditions is a useful tool for studying the stability of CNOs. Figure 4 compares the thermal stability of the two CNO samples in air. Remarkably, N-CNOs show noticeably higher stability. The decomposition of these onions starts at 700 8C.[13] In contrast, the larger A-CNOs show lower thermal stability and undergo decomposition at 500 8C.

Figure 4. TGA plots of A-CNOs (b) and N-CNOs (c) in air.

Chem. Asian J. 2007, 2, 625 – 633

Figure 5. TGA plots of pristine A-CNOs (c) and A-CNOs annealed at 1650 8C (b) and 2300 8C (g) under helium atmosphere, showing the increase in thermal stability of the A-CNOs in air after annealing.

of these CNOs at 2300 8C for one hour eventually resulted in a thermal stability comparable to that of the N-CNOs. Raman Spectroscopy The Raman spectrum of single crystals of graphite exhibits a narrow band at 1575 cm 1 originating from the vibrations of the CACHTUNGRE(sp2) framework. The Raman spectrum of all the other carbon materials shows an additional band at around 1300– 1350 cm 1. The former is called the G band and the latter the D band. The intensity of the D band increases as the average size of the crystallites in the surface layer decreases. Therefore, the intensity of this D band indicates the number of defects (sp3-hybridized carbon atoms) in the surface layer of graphite. Hence, the amount of defect in the graphitic layers can be estimated from the ratio of the intensity of the D and G bands.[14] The Raman spectrum of the CNOs from both sources exhibit D and G bands.[15, 16] The D band appears at around 1313 cm 1 and the G band at 1581 cm 1 for the A-CNOs (Figure 6). On the other hand, the corresponding bands appear at 1307 and 1593 cm 1, respectively, in case of the NCNOs. The D/G ratio is 0.8 for A-CNOs and 1.4 for NCNOs. This indicates that the outermost graphene sheet in the small N-CNOs contains more defects than in the large


www.chemasianj.org

627

L. Echegoyen et al.

FULL PAPERS

do not seem to have the paramagnetic properties of their small counterparts. The presence of carbon radicals on the surface also suggests that the small N-CNOs may be more susceptible to radical reactions than the larger A-CNOs. Reactivity of CNOs

Figure 6. Raman spectra of A-CNOs (b) and N-CNOs (c).

A-CNOs. This is also confirmed from the width at half maximum of the G band, which is broader for the N-CNOs than for the A-CNOs.[16] Interestingly, Raman and TGA studies indicate that the structure of the N-CNOs is more defective and simultaneously more thermodynamically stable than the structure of the A-CNOs. This is another example of defects leading to stability, presumably for entropic reasons. Electron Paramagnetic Resonance (EPR) Spectroscopy EPR spectroscopy has been employed frequently to study radicals in carbon materials. Tomita et al. reported that both nanodiamond particles as well as CNOs obtained from these particles by irradiation under an electron beam show a narrow EPR signal with a g value of 2.002 and a linewidth of 8.5 G.[17] This signal was attributed to the presence of dangling bonds associated with radicals on the outermost shell of the CNOs. We observed an analogous signal with a g value of 2.001 and a linewidth of 16.5 G (Figure 7), thus confirming that the CNOs formed by annealing under helium are similar to those formed by annealing nanodiamond particles under an electron beam. Interestingly, the CNOs obtained from arcing do not display an EPR signal; thus, they

Figure 7. Powder X-band EPR spectrum of N-CNOs.

628

www.chemasianj.org

Based on the different properties of the CNOs obtained from the two sources, we assumed that their reactivities are remarkably different. The presence of a large number of sp2-hybridized carbon atoms, as observed in the NMR spectra, should render them susceptible to addition reactions. Furthermore, the Raman spectroscopic studies indicate that the N-CNOs have a larger number of surface defects than the A-CNOs. Therefore, the former are expected to be more susceptible to reactions occurring at defect sites on the surface than the latter. Finally, the EPR spectroscopic results suggest that N-CNOs may readily undergo radical-addition reactions. To confirm these expectations, three reaction types were attempted with CNOs obtained from the arcing and nanodiamond-annealing processes. Cycloaddition Reaction Many addition reactions have been successfully employed in the functionalization of fullerenes. Among these reactions, the [2 + 1] Bingel–Hirsch cyclopropanation has often been used with C60. Here, bromomalonate reagents (prepared in situ with the malonate species and CBr4) in the presence of 1,8-diazabicycloACHTUNGRE[5.4.0]undec-7-ene (DBU) undergo multicycloaddition at 6,6-double-bond sites to give rise ultimately to the hexasubstituted fullerene product.[18, 19] Based on the results of NMR spectroscopy, we expected this reaction to lead to more dispersable CNOs in common organic solvents. With this aim, dodecyl malonate ester was prepared by following published procedures.[20] CNOs obtained through both methodologies were treated with this malonate, carbon tetrabromide, and DBU in o-dichlorobenzene (ODCB) for 24 h (Scheme 1). The mixture was then centrifuged, and the collected solid was washed several times with ODCB and dichloromethane to remove any trace of unreacted reagents. The residues obtained were characterized with TGA and Raman spectroscopy, as well as attenuated total reflection infrared (ATR-IR) spectroscopy and TEM (see Supporting Information). The ATR-IR spectrum of the functionalized product shows a broad band at around 1705 cm 1 that can be attributed to the C=O stretching mode. TGA performed in air (Figure 8 a) of the product obtained after reaction in the case of the N-CNOs shows the removal of the malonate groups attached to the surface of the CNOs at around 250 8C, as well as the subsequent oxidative degradation of the N-CNOs. In contrast, the product obtained from the reaction with the A-CNOs did not show any defunctionalization; the TGA plot resembles that of the pristine A-CNOs, which suggests that the reaction did not proceed with these CNOs under the experimental conditions used. The Raman spectrum of the derivatized N-CNOs (Figure 8 b) confirm the successful functionalization. An increase


Chem. Asian J. 2007, 2, 625 – 633


Scheme 1. Cyclopropanation of CNOs.

Scheme 2. Addition of phenyl radicals on CNOs.

Figure 8. a) TGA plots of cyclopropanated A-CNOs (b) and N-CNOs (c) in air. b) Raman spectra of crude N-CNOs (c) and N-CNOs after cyclopropanation (b).

in the intensity of the D band at 1307 cm 1, explained by the formation of new sp3-hybridized carbon atoms, is clearly observed. Although CNOs from both sources have sp2-hybridized carbon atoms as shown by the NMR spectra, the [2 + 1] cycloaddition only took place with the N-CNOs. This is, to the best of our knowledge, the first cyclopropanation reaction reported on the surface of CNOs. We believe that the reasons for the higher reactivity could be the larger curvature as well as the higher surface-to-volume ratio of the NCNOs. Free-Radical Addition The free-radical-addition reaction selected was previously employed with single-walled CNTs (SWCNTs) by Liang et al.[21] By following this procedure, CNOs were treated with benzoyl peroxide as the source of phenyl radicals (Scheme 2). After completion of the reaction, the mixture was centrifuged, and the residue was washed with toluene several

Chem. Asian J. 2007, 2, 625 – 633

Figure 9. a) TGA plots of A-CNOs (c) and N-CNOs (b) after phenyl functionalization. b) Raman spectra of crude N-CNOs (c) and N-CNOs after radical addition (b).

times and dried. Comparative TGA studies performed in air (Figure 9 a) revealed that in the case of the N-CNOs, an initial 25 % mass loss was observed, which was most likely due to the thermal defunctionalization followed by the degradation of the CNOs. Once again, the residue from the reaction with the larger A-CNOs showed absolutely no mass loss, which suggests that these CNOs had not reacted. The increase in the Raman D band (Figure 9 b) in the case of the N-CNOs further confirms their functionalization. EPR spectroscopic studies were performed on the NCNOs after functionalization to determine the fate of the radicals present on the surface of the pristine CNOs (Fig-


www.chemasianj.org

629

L. Echegoyen et al.

FULL PAPERS

Figure 11. Low- (left) and high-resolution (right) TEM images of N-CNOs after phenyl sulfonation. Scale bars represent 20 nm and 5 nm in the low- and high-resolution images, respectively.

Figure 10. Powder X-band EPR spectra of N-CNOs before (c) and after (b) derivatization with phenyl radicals.

ure 10 a). The absence of an EPR signal after reaction with phenyl radicals (Figure 10 b) confirms the assumption of Tomita et al[17] that all unpaired spins of the N-CNOs are present on the surface and corroborates the success of the functionalization reaction. As the phenylated CNOs were insoluble in organic solvents such as toluene or chloroform, we decided to sulfonate the phenyl residues by dispersing and heating them under

Figure 12. EDS spectrum of sodium phenyl sulfonate N-CNOs. Magnesium, copper, aluminium, and silicon are present as constituents of the TEM grid.

Oxidation of CNOs

Scheme 3. Sulfonation of the phenylated N-CNOs.

reflux for 2 h in oleum (30 % SO3 in H2SO4) followed by further treatment with NaOH (Scheme 3). The reaction mixture was filtered, and the solid obtained was washed several times with deionized water to ensure complete removal of the excess of sodium ions. The residual solid obtained after the washings was highly dispersible in water and ethanol and resulted in a solution that was stable for months, similar to that observed in the case of SWCNTs.[21] TEM micrographs (Figure 11) of the sulfonated product confirmed the presence of a large concentration of small CNOs in these aqueous solutions. Energy dispersive spectroscopy (EDS) performed during the TEM experiments confirmed the presence of sodium and sulfur in the sample (Figure 12). These results demonstrate that radical-addition reactions are possible only with the N-CNOs.

630

www.chemasianj.org

The oxidation of surface defects to carboxylic acid groups has already been described for SWCNTs[22] as well as for ACNOs.[9] As we have already shown that N-CNOs contain a larger number of defective sites on the surface (Figure 6) than the larger A-CNOs, we decided to oxidize these sites to introduce carboxylic acid functional groups, which can be subsequently employed to introduce other functionalities. TGA and Raman spectroscopic studies (Figure 13) reveal that the defective sites on the N-CNOs were oxidized after heating under reflux in 3.0 m nitric acid for 48 h (Scheme 4). TGA analysis performed in air (Figure 13 a) shows the thermal removal of the carboxy groups followed by the decomposition of the CNOs in air. Not only were the N-CNOs more dispersible in water after oxidation, but the relative ratio of the D to G bands in the Raman spectrum shows a significant increase, indicating an increase in the number of surface defects. On the other hand, the A-CNOs did not undergo any oxidation when treated under the same experimental conditions (Figure 13 a). More aggressive conditions were necessary to oxidize the surface of the larger A-CNOs. The required treatment involved heating the CNOs under


Chem. Asian J. 2007, 2, 625 – 633


Figure 13. a) TGA plots of A-CNOs (c) and N-CNOs (b) after treatment with 3.0 m nitric acid. b) Raman spectra of crude N-CNOs (c) and N-CNOs after oxidation (b).

Figure 14. a) TGA plots in air and b) Raman spectra of crude A-CNOs (c) and A-CNOs after treatment with a 1:1 mixture of HNO3 and H2SO4 (b).

Scheme 4. Oxidation of CNOs.

reflux for 48 h in a 1:1 mixture of concentrated nitric and sulfuric acids. Clear evidence of the oxidation of the A-CNOs under these experimental conditions was obtained from TGA and Raman spectroscopy (Figure 14). The thermal removal of the carboxy groups prior to the degradation of the CNOs, as well as the increase in the Raman D band, is observed. Under these aggressive conditions, the N-CNOs from nanodiamonds were completely destroyed, but not the A-CNOs, as seen in the TEM images (Figure 15).

Conclusions We have synthesized two types of CNOs. Large CNOs (20– 30 shells, 15–20 nm) are produced by arcing graphitic rods under water, whereas annealing of nanodiamond particles (5 nm average size) produces small CNOs (6-8 shells, 5 nm). Analysis of their physical properties (NMR, Raman, and EPR spectroscopy, TGA, powder X-ray diffraction, and TEM) showed pronounced differences. After conducting [2 + 1] cycloaddition, free-radical addition, and oxidative reactions, we concluded that the N-CNOs are more reactive than the A-CNOs, even though the N-CNOs are thermodynamically more stable. This suggests that the reactions stud-

Chem. Asian J. 2007, 2, 625 – 633

Figure 15. TEM image of A-CNOs after heating under reflux in a 1:1 mixture of conc. HNO3/H2SO4. N-CNOs from nanodiamonds did not survive this treatment. Scale bar represents 5 nm.

ied are under kinetic control. The higher degree of disorder along with the higher curvature of the surface of the NCNOs probably leads to a lower activation barrier for the reactions than for the A-CNOs. Other functionalizations are underway in our laboratory to make these materials soluble in organic media.


www.chemasianj.org

631

L. Echegoyen et al.

FULL PAPERS Synthesis of Arylsulfonated CNOs

Experimental Section Materials High-purity ultradispersed nanodiamonds (purchased from www.nanodiamond.com) were used as obtained. Copper grids coated with formvar stabilized with holey/lacey carbon purchased from Ted Pella, Inc. were used for TEM studies. The chemicals were used without any purification. Instrumentation Solid-state MAS NMR spectroscopy was performed on a Bruker Avance 500-MHz instrument fitted with a 4-mm magic-angle-spinning probe head. Low- and high-resolution TEM images and EDS spectra were obtained with a Hitachi Model 7600 (120 kV) and Model 9500 (300 kV) microscopes, respectively. The samples were prepared from either solution or dispersion followed by deposition on the copper grids. TGA was performed in an alumina pan on a Mettler TGA/SDTA85i instrument with a typical sample size of around 2 mg. The maximum temperature was set at 850 8C with the heating ramp set at 20 8C min 1. Raman spectra were recorded on a Renishaw 1000 Raman spectrometer. The excitation source was the 785-nm emission line of a near-IR laser. Powder X-ray diffraction was performed on a Scintag 2000 system with a germanium detector. Powder X-band EPR spectra were recorded at room temperature with a Bruker EMX spectrometer. Synthesis of Pristine CNOs Commercially available nanodiamonds ( 1 g) were placed in a graphite crucible and transferred to an Astro carbonization furnace. The air in the furnace was removed by applying vacuum followed by purging with helium. The process was repeated twice to ensure complete removal of air. The nanodiamonds were then heated to 1650 8C under a helium atmosphere with a heating ramp of 20 8C min 1. The final temperature was maintained for one hour, then the material was slowly cooled to room temperature over a period of one hour. The furnace was opened, and the transformed CNOs were annealed in air at 400 8C to remove any amorphous carbon that might have been present in the sample. About 950 mg of pure N-CNOs was obtained (95 % yield). For comparison purposes, 200 mg of A-CNOs were also synthesized by arcing graphite under water, as previously described.[4] Dodecylmalonate ACHTUNGREdescribed.[20]

was

obtained

following

a

procedure

already

The phenylated N-CNOs (30 mg) were dispersed in fuming H2SO4 (20 mL) containing 30 % SO3 by using a sonicator. The dispersion was heated under reflux for 2 h and then poured over ice. After centrifugation, the obtained solid was washed several times with water and dispersed in aqueous sodium hydroxide (1 m, 30 mL) by using a sonicator. The mixture was heated under reflux overnight and centrifuged again. The recovered solid was washed several times with water and dried overnight in a vacuum oven at 60 8C. Sonication of this material in water for 30 s led to very stable dispersions (no settling over several weeks). The product yield based on the mass of starting CNO was 64 %. Synthesis of Carboxylated CNOs A-CNOs or N-CNOs (50 mg) were dispersed in aqueous nitric acid (3 m, 30 mL) by using a sonicator. After heating at reflux for 48 h, the dispersion was centrifuged, and the recovered solid was washed several times with water and dried overnight in a vacuum oven at 100 8C. No loss of mass was observed during this process. More aggressive conditions were also attempted with both types of CNOs. The CNOs (50 mg) were dispersed in a 1:1 mixture of concentrated HNO3 and H2SO4 and heated at reflux for 48 h under argon. The crude mixture was then poured over ice and centrifuged. No solid was recovered in the case of the CNOs from nanodiamonds. The recovered solid in the case of CNOs from arcing (25 mg) was washed several times with water and dried overnight in a vacuum oven at 100 8C.

Acknowledgements Prof. Alex Kitaygorodskiy is gratefully acknowledged for the solid-state MAS NMR spectra. Financial support from the National Science Foundation (Grant number CHE-0509989) is greatly appreciated. F.M. thanks the French Ministry of Foreign Affairs for a Lavoisier Fellowship. This material is based on work supported by the National Science Foundation while L.E. was working there. All opinions, findings, conclusions, or recommendations expressed herein are those of the authors and do not necessarily reflect the views of the National Science Foundation.

Synthesis of Cyclopropanated CNOs A-CNOs or N-CNOs (15 mg) were dispersed under argon in anhydrous ODCB (20 mL) by using a sonicator. Dodecylmalonate (60 mg, 0.14 mmol), carbon tetrabromide (135 mg, 0.41 mmol), and DBU (0.15 mL, 1.00 mmol) were added, and the mixture was stirred for 24 h under argon. After centrifugation, the recovered solid was washed several times with dichloromethane and dried in a vacuum oven at 50 8C. In the case of CNOs from the nanodiamonds, 18 mg of material was obtained (20 % increase in mass). No increase in mass was observed in the case of A-CNOs. Synthesis of Phenylated CNOs A-CNOs or N-CNOs (30 mg) were dispersed under argon in anhydrous toluene (30 mL) by using a sonicator. Benzoyl peroxide (100 mg) was then added, and the mixture was heated at reflux for 2 h under argon. After cooling, more benzoyl peroxide (100 mg) was added, and the mixture was heated at reflux for two more hours. Benzoyl peroxide (100 mg) was again added, and the mixture was kept under reflux overnight. The mixture was centrifuged, and the recovered solid was washed several times with dichloromethane and dried for 4 h in a vacuum oven at 50 8C. The mass of solid obtained in the case of CNOs from nanodiamonds was 33 mg (10 % increase in mass). No increase in mass was observed with the CNOs from arcing.

632

www.chemasianj.org

[1] H. W. Kroto, J. R. Heath, S. C. OLBrien, R. F. Curl, R. E. Smalley, Nature 1985, 318, 162 – 163. [2] N. Martin, Chem. Commun. 2006, 2093 – 2104. [3] R. H. Baughman, A. A. Zakhidov, W. A. de Heer, Science 2002, 297, 787 – 792. [4] K. W. Street, M. Marchetti, R. L. Vander Wal, A. J. Tomasek, Tribol. Lett. 2004, 16, 143 – 149. [5] D. Ugarte, Nature 1992, 359, 707 – 709. [6] V. L. Kuznetsov, A. L. Chuvilin, Y. V. Butenko, I. Y. Malkov, V. M. Titov, Chem. Phys. Lett. 1994, 222, 343 – 348. [7] N. Sano, H. Wang, M. Chhowalla, I. Alexandrou, G. A. J. Amaratunga, Nature 2001, 414, 506 – 507. [8] V. Georgakilas, D. M. Guldi, R. Signorini, R. Bozio, M. Prato, J. Am. Chem. Soc. 2003, 125, 14 268 – 14 269. [9] A. S. Rettenbacher, B. Elliott, J. S. Hudson, A. Amirkhanian, L. Echegoyen, Chem. Eur. J. 2006, 12, 376 – 387. [10] A. Hirsch, Fullerenes Relat. Struct. 1999, 199, 1 – 65. [11] S. Tomita, A. Burian, J. C. Dore, D. LeBolloch, M. Fujii, S. Hayashi, Carbon 2002, 40, 1469 – 1474. [12] L. S. Cahill, Z. Yao, A. Adronov, J. Penner, K. R. Moonoosawmy, P. Kruse, G. R. Goward, J. Phys. Chem. B 2004, 108, 11 412 – 11 418. [13] M. Zhang, M. Yudasaka, S. Bandow, S. Iijima, Chem. Phys. Lett. 2003, 369, 680 – 683. [14] D. Roy, M. Chhowalla, H. Wang, N. Sano, I. Alexandrou, T. W. Clyne, G. A. J. Amaratunga, Chem. Phys. Lett. 2003, 373, 52 – 56. [15] E. D. Obraztsova, M. Fujii, S. Hayashi, V. L. Kuznetsov, Y. V. Butenko, Carbon 1998, 36, 821 – 826.


Chem. Asian J. 2007, 2, 625 – 633


[16] F. Tunistra, J. L. Koenig, J. Chem. Phys. 1970, 53, 1126 – 1130. [17] S. Tomita, T. Sakurai, H. Ohta, M. Fujii, S. Hayashi, J. Chem. Phys. 2001, 114, 7477 – 7482. [18] C. Bingel, Chem. Ber. Recl. 1993, 126, 1957 – 1959. [19] A. Hirsch, I. Lamparth, T. Grosser, H. R. Karfunkel, J. Am. Chem. Soc. 1994, 116, 9385 – 9386. [20] M. Braun, X. Camps, O. Vostrowsky, A. Hirsch, E. Endress, T. M. Bayerl, O. Birkert, G. Gauglitz, Eur. J. Org. Chem. 2000, 1173 – 1181.

Chem. Asian J. 2007, 2, 625 – 633

[21] F. Liang, J. M. Beach, P. K. Rai, W. H. Guo, R. H. Hauge, M. Pasquali, R. E. Smalley, W. E. Billups, Chem. Mater. 2006, 18, 1520 – 1524. [22] Y. P. Sun, K. F. Fu, Y. Lin, W. J. Huang, Acc. Chem. Res. 2002, 35, 1096 – 1104.


Received: December 21, 2006 Published online: March 29, 2007

www.chemasianj.org

633