On the kinetics of carbon nanotube growth by thermal CVD method
Introduction
Carbon nanotubes (CNTs) have great potential for various nanotechnology applications. Since their first successful synthesis by arc-discharge technique and observation by the transmission electron microscope (TEM) analysis [1], and the subsequent methods with the metal catalyst in an hydrocarbon/inert mixing gas atmosphere [2], [3], extensive investigations on CNTs have been carried out due to their unique physical properties [4], [5] and potential applications [6], [7], [8]. Compared to the earlier gas phase synthesis techniques, the catalytic chemical vapor deposition (CVD) methods [9], [10], [11], [12], [13], [14], [15], either thermal pyrolysis or plasma-enhanced, have the advantageous ability in in-situ selective growth on patterned area for large area array fabrication in the applications such as field emission display (FED), chemical sensors, etc. Thus far, catalytic metals deposited on substrates, such as nickel (Ni) film on silicon or glass substrates, have been widely used for this purpose. Methane (CH4), acetylene (C2H2) or ethylene (C2H4) was often used as carbon source and ammonia (NH3) was sometimes used as reactive gas for the growth of aligned CNTs in thermal CVD method [12], [13], [16], [17]. Even thought it was generally agreed that NH3 helps to maintain catalyst metal surface active [18], whether NH3 is needed in every stage of CNTs synthesis is controversial and the detailed mechanism of how CNTs grow vertically aligned is still unclear [12], [18], [19]. In our previous work [20], we found that the effects of NH3/CH4 ratios on the CNTs growth in atmospheric thermal CVD process at a fixed temperature of 900 °C showed distinct trends with and without diluting gas argon (Ar). Our experimental results suggested that there exists a critical supply of CH4 or other carbon source when CNTs growth is limited by carbon atom diffusion that might be determined by the growth temperature or/and effective active area of catalyst particles. In the case of low carbon supply and high carbon diffusion, so called carbon supply on catalyst surface limited (CSL), either graphite structure formation/CNTs growth is hampered, or CNTs grow in spaghetti-like morphology; whereas in the case of high carbon supply and low carbon diffusion, so called carbon diffusion rate limited (CDL), carbon atoms will saturate on nanoparticle surface and excess carbon atoms form amorphous carbon and CNTs growth is hampered due to catalyst passivation. It is therefore believed that the optimal CNTs growth situated on the condition that carbon atom supply and diffusion reach an identical rate; that is, CNTs grow with “best quality” at this optimal growth condition. In this article, systematic experiments were performed furthermore to examine the role of NH3 on the growth of CNTs in atmospheric thermal CVD under various temperatures and carbon source/NH3 ratios. In addition, since there have been numerous papers attempted to investigate the growth mechanisms by examining the bamboo-like structure [21], [22], [23], [24], [25], this study has also carried out a statistical analysis through an intensive TEM observations on the tube diameters, bamboo spacing (or called inter-bamboo-layer distance), and the formation rate of each diaphragm under various temperatures and carbon source/NH3 ratios. It is hoped that, in combining the kinetics-based model and the examined role of NH3 on CNTs growth as well as bamboo-like morphology, a better understanding and better control of CNTs catalytic growth can be achieved.
Section snippets
Experimental
Patterned (10×10 μm2) Ni catalyst films of thickness 10 nm were deposited on SiO2/p-Si(100) wafer by optical lithography and e-gun evaporation followed by lift-off process. The pieces of catalyst deposited substrates were then placed on a quartz plate and leaded into a 46-mm inner-diameter resistance heated quartz tube furnace at atmospheric pressure. The quartz tube was pumped down to ∼10−3 mbar using a mechanical pump, before purging Ar gas back to atmospheric pressure. The substrates were
Results and discussion
There were two sets of experiments, one with fixed C2H4 flow rate of 30 sccm (7.5% of total flow rate) and the other with fixed NH3 flow rate of 30 sccm. The SEM micrographs of the grown CNTs were shown in Fig. 1a–f and d–i for each set respectively. Between the two sets, Fig. 1d–f was a reference group in common with the gas mixing ratio of C2H4/NH3 (Rm) equal to one at various temperatures. Although not evidently shown in Fig. 1, examinations under higher magnification showed that all CNTs
Conclusion
It is believed that the optimal CNTs growth situated on the condition that carbon atom supply and diffusion reach an identical rate; that is, CNTs grow with “best quality” at this optimal growth condition. In this article, systematic experiments were performed furthermore to examine the role of NH3 on the growth of CNTs in atmospheric thermal CVD under various temperatures and carbon source/NH3 ratios. Besides, a temperature fixed pretreatment process and rapid pumping method at the end of
Acknowledgements
Financial support for this work from the National Science Council under Contract No. NSC-92-2622-E-007-016 and some TEM analysis supported by Chung-Shan Institute of Science and Technology are gratefully acknowledged.
References (26)
- et al.
Carbon
(1995) - et al.
Chem. Phys. Lett
(1999) - et al.
Chem. Phys. Lett
(1999) - et al.
Diamond Relat. Mater
(2001) - et al.
J. Eur. Ceram. Soc
(2001) - et al.
Thin Solid Films
(2001) - et al.
Diamond Relat. Mater
(2004) - et al.
Chem. Phys. Lett
(2000) Carbon
(1995)- et al.
Chem. Phys. Lett
(2001)