Research Article Received: 21 October 2011
Revised: 22 November 2011
Accepted: 23 November 2011
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI 10.1002/jrs.3150
Carbon nanostructures produced by pyrolysis under high pressure inside a nanosize silica matrix Antonio E. L. Villanueva,a Naira M. Balzarettia* and Joao A. H. da Jornadaa,b In this work, the pyrolysis under high pressure of hydrocarbons dispersed inside a nanosized silica matrix (Aerosil) was investigated. The samples consisted of hydrophobic nanometric silica powder terminated by methyl groups with carbon contents ranging from 0.7 to 4 wt%. The pyrolysis was carried out in the temperature range from 1000 to 1600 C under high pressure (1.25 up to 7.7 GPa) to keep the two-dimensional distribution of carbon atoms originally at the silica grain boundaries. Evidences from Raman spectroscopy and transmission electron microscopy suggested that the resulting carbon nanostructures were actually graphene-like nanoflakes. The size of the nanostructures calculated from the ID/IG ratio increased from 6 to 30 nm for processing temperatures increasing from 1000 to 1600 C under pressure, respectively. The results revealed that the very good dispersion of the methyl groups inside the nanosize silica matrix, and the confinement under high pressure during the pyrolysis, played both a relevant role in the resulting carbon nanostructures. Copyright © 2012 John Wiley & Sons, Ltd. Keywords: Raman spectroscopy; nanocarbon; nanocomposite
Introduction Graphene has been extensively studied because of its twodimensional (2D) structure, consisting of a layer of carbon atoms in a hexagonal arrangement.[1] The behavior of the electrons on the graphene sheet resembles a two-dimensional gas of massless Dirac fermions with very interesting properties.[2,3] Recently, Liu et al.[4] presented graphene as an outstanding candidate designed for electrode material for electrical double-layer supercapacitors because of its exceptional thermal and electrical conductivity and, most importantly, because of its remarkably high specific surface area up to 2675 m2/g. Moreover, both surfaces of the graphene sheet are accessible for the electrolytes. One of the drawbacks of graphene, however, is the tendency of the sheets to bend or stack themselves to reduce the surface energy during the preparation steps following the several available experimental routes. This effect reduces the access of the electrolytes to both surfaces and, therefore, the formation of double-layer electrical charges. A possible way to stabilize the graphene sheet is the full hydrogenation of the surfaces to form a stable two-dimensional hydrocarbon, called graphane.[5] Another possibility, which will be investigated in the present work, would be to clutch graphene sheets inside a nanosized inert matrix, avoiding the stacking of the layers. The pyrolysis of organic precursors is a standard method to produce different carbon phases. Accordingly, in this work we investigated an alternative route to produce 2D carbon nanostructures with controlled size, dispersed, and embedded in a matrix of fumed silicon dioxide. The carbon nanostructures were formed during the pyrolysis under high pressure of the CH2 and CH3 termination groups at the surface of the nanoparticles of hydrophobic fumed silica Aerosil© (Evonik Inc). The partial or total elimination of the hydrogen atoms would be promoted by the high temperature processing of the samples confined under pressure. The
J. Raman Spectrosc. (2012)
carbon atoms will probably remain at the grain boundaries because of the decrease of the available free space for diffusion induced by high pressure. In this way, the pyrolysis inside the silica matrix carried out under high pressure would induce the formation of a 2D carbon nanostructure network creating a graphene–silica composite, as depicted in Fig. 1. In this work Raman spectroscopy and transmission electron microscopy (TEM) were used to characterize the carbon nanostructures produced after the pyrolysis under pressure[6–13] of Aerosil with carbon contents varying from 0.7 to 4 wt%. The processing conditions investigated ranged from 1.25 to 7.7 GPa in the temperature interval between 1000 and 1600 C.
Experimental Table 1 presents the properties of the Aerosil© samples used in this work. The carbon content in the methyl groups ranged from 0.7 to 4 wt%, and the average size of the silica grains was from 7 to 12 nm. Before the pyrolysis under high pressure, the powdered samples were precompacted at 0.25 GPa into cylinders of 3 mm diameter and 2 mm height. The cylinders were then placed inside a hBN container, which acted as a soft pressure transmitting medium. The container with the sample was placed inside a graphite cylinder used as the heating element of the high pressure configuration system. A toroidal high pressure apparatus was used
* Correspondence to: Naira M. Balzaretti, Instituto de Física, UFRGS, Av. Bento Gonçalves, 9500, 91501–970, Porto Alegre, RS, Brazil. E-mail:
[email protected] a Instituto de Física, UFRGS, Av. Bento Gonçalves, 9500, 91501-970, Porto Alegre, RS, Brazil b INMETRO, Rio de Janeiro, Brazil
Copyright © 2012 John Wiley & Sons, Ltd.
A. E. L. Villanueva, N. M. Balzaretti and J. A. H. da Jornada
Figure 1. Schematic representation of the formation of the carbon nanostructures during the pyrolysis under high pressure of hydrophobic Aerosil. (a) Hydrophobic silica grain (nanometric size) with the methyl groups at the surface; (b) cluster of pristine silica grains; and (c) after the pyrolysis under high pressure, the growth of the silica grains is represented by the large white areas, while the small amount of remaining carbon atoms would remain at the grain boundaries forming nanostructures.
to perform the high pressure and high temperature processing from 1.25 to 7.7 GPa in the temperature range from 1000 to 1600 C. During the experiments, initially the sample was submitted to high pressure and, then, it was heated to the desired temperature during the selected time interval. After that, the heating system was turned off and the pressure was decreased. Details about the experimental setup are given elsewhere.[14] The Raman spectra were measured in a Horiba Jobin-Yvon HR320 spectrometer connected to an Olympus microscope. The excitation was provided by the 632.8 nm line of a 10 mW polarized
Table 1. Physico-chemical data of Aerosil© (Evonik, Ind.) samples
HeNe laser focused in a spot of approximately 2 mm. The scattered light was collected in back reflection geometry and filtered with a Super-Notch Plus filter (HSNF 633–1.0, Kaiser Optical Systems, Inc.). A Jeol JEM 1200 microscope was used to obtain transmission electron micrographs of the samples at 120 kV using copper grids. Figure 2 shows a picture of the samples before and after processing. As can be seen, the originally white Aerosil powder changed to a black and glassy sample after pyrolysis under high pressure. The black color indicates that graphitic or amorphous carbon remains in the sample.
Results and discussion
Sample
Aerosil© R974
Aerosil© R816
Aerosil© R812S
Effect of high pressure
Average particle size (nm) Carbon content (%) BET surface area (m2/g)
12 0.7 – 1.3 170 20
12 1.2 – 2.2 170 25
7 3.0 – 4.0 220 25
To investigate the effect of pressure during the pyrolysis, the Aerosil samples were first pyrolyzed under ambient pressure. Figure 3 shows the Raman spectrum of the R974 Aerosil sample processed at different temperatures from 100 to 1100 C under ambient
Figure 2. (a) Hydrophobic Aerosil powder before high pressure - high temperature (HPHT) and the corresponding representation of the methyl group terminated silica grains; (b) picture of the cross-section of the cylinder containing the sample after the pyrolysis under high pressure and the corresponding representation of the silica grains and carbon nanostructures at the grain boundaries. The height of the sample is ~2 mm and the width is ~3 mm.
wileyonlinelibrary.com/journal/jrs
Copyright © 2012 John Wiley & Sons, Ltd.
J. Raman Spectrosc. (2012)
Carbon nanostructures produced by pyrolysis under high pressure inside a nanosize silica matrix amount of edges acting as defects with respect to the small crystallite area.[16] This is particularly evident for the samples processed at 1.25 and 2.5 GPa. At this pressure range, the level of plastic deformation of the silica grains was probably suitable for the elimination of hydrogen and oxygen under high temperature, leaving small amounts of carbon atoms in-between them. At 4.0 GPa, the width of both D and G peaks increased and the D´ peak is not clearly identified. The broad background observed at 7.7 GPa is probably related to amorphous carbon. These high pressure and temperature conditions induced the crystallization of the silica matrix, as indicated by the small peak at ~520 cm–1 corresponding to the coesite phase of SiO2, even for very short processing times. Effect of high temperature
Figure 3. Raman spectra of R974 aerosil sample after annealing from 100 to 1100 C under argon atmosphere during 15 min at ambient pressure.
pressure during 15 min under argon atmosphere. Similar results were obtained for R812S and R816 samples. As can be seen, there was no indication of remaining graphitic carbon phases after the pyrolysis under ambient pressure, except for a large band centered at ~1400 cm–1 compatible with amorphous carbon. Figure 4 shows the typical Raman spectrum for the samples processed at 1000 C during 10 s in the pressure range from 1.25 to 7.7 GPa. A comparison with the results shown in Fig. 3 clearly demonstrates the remarkable effect of high pressure on the resulting carbon nanostructures produced after the pyrolysis of the Aerosil samples even for a very short processing time of 10 s. The observed peaks are related to vibration modes of sp2-hybridized carbon atoms.[15] The G band at ~1580 cm–1 is related to the E2g phonons at the center of the Brillouin zone. The D band at ~1330 cm–1 is related to breathing-like modes activated by defects. The D´ band at ~1620 cm–1 is also related to the presence of defects. The anomalous high intensity of the D and D´ bands, and, the very small full-width at half-maximum (FWHM) of these bands observed for the samples processed at 1.25 and 2.5 GPa, may be related to very small but highly crystalline carbon structures formed at the interfaces among silica grains, containing a large
Figure 4. Raman spectra of the R974 sample processed at 1000 C during 10 s at 1.25, 2.5, 4 and 7.7 GPa. The spectra were vertically displaced for clarity.
J. Raman Spectrosc. (2012)
Figure 5 shows the Raman spectra of the Aerosil© R812S processed at 2.5 GPa during 10 s at different temperatures. As can be seen, the intensity of the D and D´ peaks decreased as the temperature increased. At higher temperatures the larger diffusion of the carbon atoms would induce the formation of larger graphene-like sheets, decreasing the contribution of edge effects in the Raman spectra. Figure 6 shows the ID/IG ratio as a function of the processing temperature for the samples with different carbon contents (Table 1) submitted to 2.5 GPa during 10 s. In this case, ID and IG are the intensity of the D and G peaks, respectively. As can be seen, this ratio is unusually high, close to 6, for samples processed at 2.5 GPa and 1000 C, corroborating the idea of the presence of very small but highly crystalline carbon structures at the grain boundaries. This ratio can, in fact, be related to the size La of the sp2-hybridizated carbon clusters according to:[6,11] La ¼ 1:8 0:5 10
9
l4laser
1 ID IG
(1)
where llaser is the wavelength of the excitation laser. According to Eqn (1), for samples processed at 1000 C, the calculated size of the carbon crystallites was in the range between 6 and 10 nm, of the same order of magnitude as the grain size of the silica matrix, according to Table 1, suggesting that the carbon atoms remained close to the surface of the grains during the pyrolysis at these temperature and pressure conditions. The pyrolysis induced by high temperature would promote the partial
Figure 5. Raman spectra for Aerosil© R812S processed at 2.5 GPa during 10 s as a function of temperature. The spectra were vertically displaced for clarity.
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jrs
A. E. L. Villanueva, N. M. Balzaretti and J. A. H. da Jornada 1000 C during 10 s are related to the formation of graphane because of the hydrogenation of the graphene-like sheets because of the OH termination groups of silica, the samples were submitted to a subsequent thermal annealing, under ambient pressure, and inert atmosphere to eliminate any residual hydrogen, following the experimental procedure proposed by Srinivasan and Saraswathi.[9] The Aerosil R812S sample initially processed at 2.5 GPa, 1000 C during 10 s was, then, annealed at different temperatures under argon atmosphere (50 ml/min) during 15 min heated at 15 C/min up to 700 C. The behavior of the Raman spectra remained the same up to 700 C, indicating that there was no residual hydrogen in the samples after the pyrolysis under high pressure and, therefore, the Raman spectra would not be related to the formation of graphane. Effect of processing time Figure 6. ID/IG ratio as a function of temperature for samples processed at 2.5 GPa during 10 s. The carbon content of each sample is given in Table 1.
or total elimination of the hydrogen atoms and the remaining carbon atoms would probably add up to a graphene or graphane-like coating of the silica grains. As the temperature increases, the diffusion increases, and the size of the carbon crystallites, reaching ~30 nm for 1500 C according to Eqn (1). The final structure would be a composite containing silica grains of nanometric size coated by a graphene-like carbon layer, as depicted in Fig. 1. The very short processing time under high temperature and high pressure would limit the growth of the silica grains. To investigate if the anomalously high intensity of D and D´ Raman peaks, and the very small value of the FWHM of the D peaks observed for samples processed at 1.25 and 2.5 GPa and
Figures 7(a) and (b) show the Raman spectra, including the 2D band, of the R812 sample processed at 2.5 GPa, 1000 and 1200 C, respectively, for different processing times. Table 2 shows the Raman parameters obtained from these spectra. The effect of longer processing times at 2.5 GPa is similar to the effect of higher temperatures, corresponding to a decrease of the AD/AG ratio, where AD and AG are the areas of the Raman peaks. This is probably related to the increase of the size of the carbon nanostructures because of the longer time/temperature for diffusion of the carbon atoms inside the silica matrix. Figures 7(c) and (d) show the Raman spectra of the R812 sample processed at 7.7 GPa, at 1000 and 1200 C, respectively, for different processing times and the corresponding Raman parameters are also presented in Table 2. As can be seen, at 7.7 GPa the result of the pyrolysis is completely different. For longer processing times the Raman spectra show only a large band in the region corresponding to amorphous carbon. The shear
Figure 7. Raman spectra of the R812 Aerosil sample processed at (a) 2.5 GPa and 1000 C; (b) 2.5 GPa and 1200 C; (c) 7.7 GPa and 1000 C; and (d) 7.7 GPa and 1200 C at different processing times: 30 s, 2 min, and 10 min.
wileyonlinelibrary.com/journal/jrs
Copyright © 2012 John Wiley & Sons, Ltd.
J. Raman Spectrosc. (2012)
Carbon nanostructures produced by pyrolysis under high pressure inside a nanosize silica matrix Table 2. Raman parameters for R812 Aerosil sample processed at 2.5 and 7.7 GPa, and 1000 and 1200 C, for different processing times (30 s, 2 min, and 10 min) P (GPa)
T ( C)
t (s)
D peak (cm–1)
G peak (cm–1)
2D peak (cm–1)
FWHM D (cm–1)
FWHM G (cm–1)
AD/AG
A2D/AG
2.5
1000 — — 1200 — — 1000 — — 1200 — —
30 120 600 30 120 600 30 120 600 30 120 600
1333.0 1330.0 1334.8 1334.1 1334.0 1338.6 1338.5 1339.8 — 1353.0 1354.5 —
1587.8 1585.5 1582.4 1581.6 1582.7 1583.8 1598.0 1588.6 — 1591.7 1593.5 —
2636.5 2645.9 2661.4 2659.0 2659.7 2675.2 2657.1 2653.7 — 2699.4 — —
32 31 32 28 27 31 46 — — 38 — —
30 28 21 23 20 31 34 — — 18 — —
7.5 7.3 5.4 4 4.5 1 4.9 — — 2.1 — —
0.6 1.1 2.3 3.5 2.5 1.6 0.7 — — 1.2 — —
2.5
7.7
7.7
Figure 8. TEM images of Aerosil R812S samples processed at 2.5 GPa, 1000 C during 10 s after etching the silica matrix.
stresses induced during the phase transformation would probably hinder the formation of graphene-like carbon structures in between the grains for longer processing times. It is interesting to point out that the FWHM of the D peak for the samples processed at 2.5 GPa and 1200 C during 30 s and 2 min was smaller whereas the A2D/AG was larger than the values measured for the other processing conditions for hydrophobic Aerosil, compatible with very small graphene-like clusters.
Transmission electron microscopy Figure 8 shows transmission electron microscopy (TEM) images of the Aerosil R812S samples processed at 2.5 GPa, 1000 C during 2 min. The silica matrix was eliminated by a normal solution of fluoridric acid (HF) diluted in 50% water during 96 h and then washing it with isopropyl alcohol. A drop of the alcohol containing the dispersed carbon clusters was analyzed on top of a copper grid after the evaporation of the alcohol. The size of the carbon structures shown by TEM were of the same order of magnitude of the values calculated using Eqn (1). The Raman spectrum of the sample after etching the silica matrix remained unchanged, revealing that there was no effect of the matrix on the stabilization of the carbon structure other than the confinement in two dimensions in-between the grains.
J. Raman Spectrosc. (2012)
Conclusions The results shown in this work revealed that very small twodimensional carbon structures can be created under high pressure and remain stable inside the silica matrix. The results obtained from Raman spectroscopy and TEM were consistent with this picture. The size of the carbon clusters calculated through the ID/IG ratio was of the same order of magnitude as the silica grain size, below 10 nm, for samples processed at 1–2.5 GPa and 1000 C. These results demonstrate that the pyrolysis of carbon precursors under high pressure could be a promising new route to produce graphene and graphene-like structures of very small dimensions in a controlled and confined way. Hydrophobic Aerosil samples showed to be very suitable for creating these structures because they have controlled and well dispersed amounts of carbon chemically connected to the silica matrix and the matrix itself does not interact with the carbon structures during the pyrolysis. After the pyrolysis, the silica matrix can be easily removed from the carbon nanostructures by fluoridric acid (HF) chemical etching. On the other hand, the procedure developed in this work represents a practical way to produce a graphene-silica composite that can be suitable for technological applications such as in supercapacitors. The most interesting results, corresponding to unusual small FWHM of the D peak related to very small graphene-like clusters of carbon
Copyright © 2012 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jrs
A. E. L. Villanueva, N. M. Balzaretti and J. A. H. da Jornada atoms, were obtained at relatively mild conditions, in the pressure range between 1 and 2.5 GPa at temperatures close to 1200 C. Acknowledgements The authors thank Evonik Inc for the Aerosil samples and for the Brazilian agencies CAPES, CNPq and Fapergs for financial support. The authors also thank the Center of Electron Microscopy (CME/ UFRGS) for helping with the transmission electron microscopy analysis.
References [1] K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dobonos, I. V. Grigorieva, A. A. Firsov, Science 2004, 306, 666. [2] K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, M. I. Katsnelson, I. V. Grigorieva IV, S. V. Dubonos, A. A. Firsov, Nature 2005, 438, 197. [3] M. I. Katsnelson, Mater. Today 2007, 10, 20. [4] C. Liu, Z. Yu, D. Neff, A. Zhamu, Z. J. Jang, Nano Lett. 2010, 10, 4863. [5] J. O. Sofo, A. S. Chaudhari, G. D. Barber, Phys. Rev. 2007, 75, 153401.
wileyonlinelibrary.com/journal/jrs
[6] G. Compagnini, F. Giannazzo, S. Sonde, V. Rainieri, E. Rimimi, Carbon 2009, 47, 3201. [7] A. Das, B. Chakraborty, A. K. Sood, Bull. Mater. Sci. 2008, 31, 579. [8] I. Calizo, A. A. Balandin, W. Bao, F. Miao, C. N. Lau, Nano Lett. 2007, 7, 2645. [9] C. Srinivasan, R. Saraswathi, Curr. Science 2009, 97, 302. [10] D. Graf, F. Molitor, K. Ensskin, C. Stamper, A. Jungen, C. Hierold, L. Wirtz, Nano Lett. 2007, 7, 238. [11] L. G. Cançado, A. Jorio, E. H. Martins Ferreira, F. Stavale, C. A. Achete, R. B. Capaz, M. V. O. Moutinho, A. Lompardo, T. S. Kulmala, A. C. Ferrari, Nano Lett. 2011, 11, 3190. [12] J. S. Park, A. Reina, J. Kong, G. Dresselhaus, M. S. Dresselhaus, Carbon 2009, 47, 1303. [13] D. C. Elias, R. R. Nair, T. M. G. Mohiudiin, S. V. Morozov, P. Blake, M. P. Halsall, A. C. Ferrari, D. W. Boukhvalov, M. I. Katsnelson, A. K. Geim, K. S. Novoselov, Science 2009, 323, 610. [14] W. F. Sherman, A. Stadmuller. Experimental Techniques in High Pressure Research, Wiley, London, 1987. [15] A. C. Ferrari, Solid State Commun. 2007, 143, 47. [16] J. Campos-Delgado, J. M. Romo-Herrera, X. Jia, D. A. Cullen, H. Muramatsu, Y. A. Kim, T. Hayashi, Z. Ren, D. J. Smith, Y. Okuno, T. Ohba, H. Kanoh, K. Kaneko, M. Endo, H. Terrones, M. S. Dresselhaus, M. Terrones. Nano Lett. 2008, 8(9), 2773.
Copyright © 2012 John Wiley & Sons, Ltd.
J. Raman Spectrosc. (2012)
