Carbon nanotubes growth on silicon nitride substrates

Share Embed


Descrição do Produto

Materials Letters 65 (2011) 1479–1481

Contents lists available at ScienceDirect

Materials Letters j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / m a t l e t

Carbon nanotubes growth on silicon nitride substrates Isadora Berlanga a, Rubén Mas-Ballesté a,⁎, Félix Zamora a, Jesús González-Julián b, Manuel Belmonte b,⁎⁎ a b

Departamento de Química Inorgánica, Universidad Autónoma de Madrid, Cantoblanco 28049, Madrid, Spain Institute of Ceramics and Glass (CSIC), Kelsen 5, 28049 Madrid, Spain

a r t i c l e

i n f o

Article history: Received 21 October 2010 Accepted 13 February 2011 Available online 18 February 2011 Keywords: Carbon nanotubes Chemical vapour deposition Ceramics Porous substrates Silicon nitride

a b s t r a c t Carbon nanotubes were synthesized on silicon nitride substrates by thermal chemical vapour deposition using an iron precursor catalyst. The nanotubes were characterized by AFM, FESEM, TEM and micro-Raman spectroscopy. The surface topography of the substrate, dense and flat or porous and rough, controlled the catalyst distribution and carbon nanotubes growth. Flat surfaces led to the synthesis of single-walled carbon nanotubes, whereas the porous ones promoted the growth of multi-walled carbon nanotubes of 60 nm diameter. These nanotubes preferentially grew on the porous sites, exhibiting a good substrate–nanotube interface. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Carbon nanotubes (CNTs) have attracted much attention since their discovery 20 years ago [1] due to their unique physical properties [2]. Based on these, uncountable promising applications in a wide variety of nanoscience fields, such as electronic, photonic, energy, environmental protection and biotechnology can be found. Many of these properties have been exploited by the preparation of composite materials. In this sense, extensive research has been conducted in polymer composites containing CNTs [3] and to a lesser extent in ceramic based ones [4]. Among the different CNT growth processes, thermal chemical vapour deposition (T-CVD) is probably the most suitable to mass production of long CNTs with low density of defects [5]. T-CVD uses nanoparticles of transition metals as a catalyst to promote the CNTs growth mechanism, the control of the size and the distribution of the catalytic nanoparticles being critical to determine the diameter and location of CNTs [6]. Thus, single-walled CNTs (SWCNTs) were produced by using catalyst particles with diameters of a few nanometers; whereas multi-walled CNTs (MWCNTs) were grown with catalyst particle sizes over 10 nm [7]. The aligned long CNTs growth on precise positions would allow developing, for instance, field emission flat panel displays [8] or self-lubricating components for tribological applications [9]. A different approach to control the diameter and location of CNTs consisted on restricting the CNTs growth inside channels of materials such as porous silicon, zeolites or certain types of alumina. By using templates of these materials, ordered arrays with well-defined orienta⁎ Corresponding author. Tel.: +34 914972947; fax: +34 914974833. ⁎⁎ Corresponding author. Tel.: +34 917355863; fax: +34 917355843. E-mail addresses: [email protected] (R. Mas-Ballesté), [email protected] (M. Belmonte). 0167-577X/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.matlet.2011.02.046

tions of SWCNTs and MWCNTs were obtained. These CNTs networks are essential towards practical electronic applications [2]. In the present work, the diameter and distribution of CNTs grown by T-CVD were investigated by varying the substrate topography, which was given by its porosity. Silicon nitride (Si3N4) ceramics were selected as substrate because these present good tribomechanical properties [10], enabling their use in technological applications as engineering parts, such as valves in diesel engines, ball bearings or metal cutting and shaping tools [11]. The CNTs growth on specific sites of the Si3N4 surface could act as lubricating microreservoirs, enhancing the tribological response of the ceramic component. Despite CNTs previously grown on amorphous hydrogenated Si3N4 thin films deposited on silicon substrates [12], to our knowledge, the successful CNTs growth on bulk Si3N4 substrates has not been reported. 2. Experimental Si3N4 substrates were obtained from homogenous powder mixtures containing Si3N4 plus 2 wt.% of aluminium oxide and 5 wt.% of yttrium oxide, both used as sintering additives. The powder composition was spark plasma sintered (Dr. Sinter, SPS-510CE) at two different temperatures, 1550 and 1600 °C, to get porous and dense specimens, respectively. The complete processing of the materials is described elsewhere [13]. Porous (3.2 vol.%) and fully dense Si3N4 discs of 20 mm diameter and 3 mm thickness were cut and ground to plates of 10× 10× 2 mm3. The surface roughness (Ra) of the porous specimens was 0.17 μm, while the dense plates were polished to Ra below 0.01 μm. In order to prepare the catalyst precursor for the CVD growth, the substrates were previously cleaned with 2-propanol, dried with Ar and dipped twice at a speed of 0.5 mm·s–1 in a 0.6 mM 2-propanol solution of Fe(NO3)3 9H2O. Thereafter, substrates were introduced in a quartz

1480

I. Berlanga et al. / Materials Letters 65 (2011) 1479–1481

Fig. 1. AFM topography image of SWCNTs (with height in the range of 1.5–2.5 nm) grown on flat Si3N4 substrate.

tube furnace and subjected to a flow of mixture comprising 200 sccm of H2 and 800 sccm of Ar for 20 min at 850 °C. Subsequently, the conditions were readjusted by keeping constant the flows of H2 and Ar, and adding 6 sccm of ethylene for 12 min. After this procedure, the sample was allowed to slowly cool down during a period of 4 h keeping a constant flux of Ar of 800 sccm. The grown CNT samples were characterized by field emission scanning electron microscopy (FESEM, Hitachi S-4700), atomic force microscopy (AFM) in dynamic mode, using a Nanotec Electronica system. Olympus cantilevers were used with a nominal force constant of 0.75 N m−1. The images were processed using WSxM software [14], and micro-Raman spectroscopy (Renishaw inVia) with an excitation energy of 2.54 eV and a 514 nm wavelength laser. Transmission electron microscopy (TEM, JEOL JEM 3000 F) with an accelerating voltage of 300 kV having an energy dispersive X-ray spectrometer (EDS) attached was used to analyze the CNTs growth on porous substrates, which were scratched with a diamond tip to get the CNTs. The collected dark powder was dispersed in distilled water with the aid of ultrasonication (24 kHz) and sodium dodecyl sulphate (SDS, 0.2% v.v.) and, then, 10 times diluted and deposited by drop-casting on the TEM grids (lacey carbon-coated copper grids, 200 mesh).

3. Results and discussion In a previous work [15], the factors that contributed to the length and density of CNTs grown on flat silicon oxide (SiO2) substrates were reported, confirming that the nature of the catalyst precursor, the temperature of the CVD process and the density of metallic nanoparticles that served as catalyst were the key parameters. Based on this study, similar experimental conditions were used for polished and flat Si3N4 substrates. The results obtained are illustrated in Fig. 1, where AFM analysis shows one dimensional structures of heights ranging from 1.5 to 2.5 nm with lengths of several microns. These observations evidenced the growth of SWCNTs on the ceramic surface, confirming that the chemical nature of the substrate (SiO2 versus Si3N4) had a negligible influence when comparing similar surface finishing. However, the T-CVD process on porous Si3N4 substrates led to significant changes on the CNTs growth (Fig. 2). At a first glance, a larger density of CNTs (Fig. 2a,b) was grown compared to that obtained on flat substrates. Besides, CNTs seem to be mainly located along the grinding directions. FESEM observations at higher magnifications (Fig. 2c,d) depicted that the nanotube lengths ranged from 500 nm to 3 μm with homogeneous diameters and notably thicker (~60 nm), than those

Fig. 2. FESEM micrographs of the CNTs grown on Si3N4 porous substrates taken at different magnifications: a) CNTs growth along the grinding direction (vertical lines), b) high CNTs density on the substrate, c) and d) CNTs anchored to the substrate through the porous cavities. MWCNTs grown on porous sites are pointed out by arrows.

Intensity (a.u.)

I. Berlanga et al. / Materials Letters 65 (2011) 1479–1481

D-band

1481

G-band

1300 1350 1400 1450 1500 1550 1600 1650 1700

Raman shift (cm-1) Fig. 4. Micro-Raman spectrum of the MWCNTs grown on porous Si3N4 substrate.

Fig. 3. TEM micrograph of an individual MWCNT grown on porous Si3N4 substrate. The inset corresponds to the electron diffraction pattern of an individual nanotube.

previously obtained on flat Si3N4 surfaces. This nanotube diameter suggests the growth of MWCNTs instead SWCNTs. In fact, TEM image (Fig. 3) showed the typical structure of MWCNTs formed by multiple rolled graphene layers, with an outer and inner diameter of 60 and 50 nm, respectively. The inset shows the selected area electron diffraction (SAD) pattern of a MWCNT, the tube axis being vertical. The 00l maxima confirm the orientation of the graphene layers parallel to the tube axis [16]. In addition, micro-Raman spectroscopy was also employed to analyze the quality of the MWCNTs grown. Fig. 4 shows the characteristic peaks of CNTs corresponding to D- (1350 cm−1) and G bands (1580 cm−1). While D-band is associated to defects in MWCNTs, G-band gives an idea of MWCNTs crystallinity. Therefore, the intensity ratio between D and G bands (ID/IG) will allow to estimate the crystallinity of CNTs and, in this particular case, ID/IG was ~0.90, confirming a high degree of crystallinity of the MWCNTs. It is interesting to remark that MWCNTs mainly grew from the porous sites of the substrate (Fig. 2c,d), where the catalyst concentration was supposedly higher, implying a strong substrate–nanotube mechanical interlock, which is important for tribomechanical applications. Therefore, it seems that the morphology of the nanotubes can be controlled, not only modifying the catalyst particle size [7], but also varying the surface topography of the substrate. Consequently, SWCNTs or MWCNTs with different diameters could be synthesized simply controlling the size of the cavities where the catalyst is introduced. Besides, the patterned distribution of those cavities would promote the development of CNTs microreservoirs. 4. Conclusions The CNTs growth by T-CVD on Si3N4 substrates depended on the surface topography. Dense and flat substrates led to SWCNTs growth of 1.5–2.5 nm diameter and a few microns length, while the porous and rough substrates promoted the synthesis of MWCNTs with 60 nm

diameter and maximum lengths of 3 μm. These nanotubes were preferentially located at the porous sites following the grinding grooves. Controlling the size and distribution of the porosity, different types of CNTs can be grown, which constitute a promising approach to develop new hybrid devices with possible applications on the design of components with surface lubricating reservoirs. Acknowledgements The financial support of the Spanish Ministry of Science and Innovation (MICINN) through projects MAT2006-7118, MAT200909600 and MAT2010-20843-C02-01, CAM (S2009_MAT-1467) and CC10-UAM/MAT-5881 are recognized. J. González-Julián acknowledges the financial support of the JAE (CSIC) fellowship Program. Dr. MasBallesté thanks the Spanish M.E.C. for funding through the “Ramón y Cajal” program. References [1] Iijima S. Nature 1991;354:56–8. [2] Harris PJF. Carbon nanotube science. Synthesis, properties and applications. Cambridge: Cambridge University Press; 2009. [3] Byrne MT, Gun´ko YK. Adv Mater 2010;22:1672–88. [4] Cho J, Boccaccini AR, Shaffer MSP. J Mater Sci 2009;44:1934–51. [5] Maruyama S, Kojima R, Miyauchi Y, Chiashi S, Kohno M. Chem Phys Lett 2002;360: 229. [6] Hongjie D. Acc Chem Res 2002;35:1035–44. [7] Öncel C, Yürüm Y. Carbon nanotube synthesis via the catalytic CVD method: a review on the effect of reaction parameters. In Fullerenes, Nanotubes, and Carbon NanostructuresTaylor & Francis Group; 2006. p. 17–37. [8] Choi WB, Chung DS, Kang JH, Kim HY, Jin YW, Han IT, et al. Appl Phys Lett 1999;75: 3129–31. [9] Miyake K, Kusunoki M, Usami H, Umehara N, Sasaki S. Nano Lett 2007;7:3285–9. [10] Petzow G, Herrmann M. Struct Bond 2002;102:47–167. [11] Riley FL. J Am Ceram Soc 2000;83:245–65. [12] Handuja S, Singh SP, Srivastava P, Vankar VD. Mater Lett 2009;63:1249–51. [13] Miranzo P, González-Julián J, Osendi MI, Belmonte M, Ceram Int 2011;37:159–66. [14] Horcas I, Fernandez R, Gomez-Rodriguez JM, Colchero J, Gomez-Herrero J, Baro AM. Rev Sci Instrum 2007;78:13705. [15] López V, Welte L, Fernández MA, Moreno-Moreno M, Gómez-Herrero J, de Pablo PJ, et al. J Nanosci Nanotechnol 2009;9:2830–5. [16] Koziol K, Shaffer M, Windle A. Adv Mater 2005;17:760–3.

Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.