Towards the low temperature growth of uniform diameter multi walled carbon nanotubes by catalytic chemical vapour deposition technique

1Department of Chemistry, Anna University Chennai, Chennai-25, India 2Department of Chemical Engineering, University of Massachusetts, Amherst, USA *Corresponding author. E-mail address: soma_nano@yahoo.co.in (T. Somanathan) Towards the low temperature growth of uniform diameter multi walled carbon nanotubes by catalytic chemical vapour deposition technique Thirunavukkarasu Somanathan1,* and Arumugam Pandurangan1,2

the electrical properties of multiwalled CNT bundles grown by chemical vapor deposition (CVD) have been intensively investigated for interconnect applications [1][2][3][4]. Lowtemperature growth of CNTs is one of the most fundamental issues to be resolved for development of CNT based on interconnects technology compatible with LSI manufacturing processes.
Generally, carbon nanotubes have been synthesized by three different techniques: arc discharge between two graphite electrodes [5], laser evaporation of carbon target [6] and chemical vapour deposition (CVD) [7]. The major drawback of arc discharge and laser evaporation methods is that they are extremely uncontrolled in terms of process parameters, resulting in CNTs that contains significant fractions of unwanted material and that are difficult to manipulate and assemble in specific designs. CVD method is based on the thermal decomposition of hydrocarbon compounds over transition metal catalyst particles.
It appears to be a simple and economic technique to synthesize this kind of material at a low temperature, ambient pressure and it represents the best hope for large scale production.
To date, all the metal-catalyzed CNT growth studies have mainly focused on transition metals. Curiously, f-block transition elements also referred to as inner transition elements have still not been well explored. One reason could be that inner transition elements are not known to be highly catalytic as the transition elements. It has also been suggested that lanthanoids such as gadolinium and europium exhibit insufficient carbon solubility, slow carbon diffusion, and limited carbide formation to catalyze CNT growth [8]. Recently, gadolinium and europium were used as catalyzed to synthesize single walled carbon nanotubes [9].
Gd (III) ion exchanged NaY zeolite has been reported as a gastrointestinal contrast agent [10]. However, Y zeolite has a pore size of around 1

Characterization of Catalysts and Carbon nanotubes
The powder XRD diffraction patterns for the calcined

Physico-chemical characteristics of Gd incorporated MCM-41 materials
The physico-chemical characteristics of the Gd incoporated MCM-41 materials that were used for the synthesis of carbon nanotubes were investigated by means of XRD, N 2 sorption studies, TGA, SEM, TEM and EPR analysis.

XRD of the catalysts
In order to aware about the periodic and porous nature of the materials, the XRD analysis was performed. The XRD patterns of Gd-MCM-41 (Si/Gd = 50, 75 and 100) are shown in

N 2 sorption studies
The N 2 sorption analysis was employed for Gd-MCM-41 materials and their data are presented in Table 1. The N 2 sorption studies were typical for type IV, with a hysteresis loop characteristic of mesoporous materials (see Fig. 2). The isotherms exhibited three stages. The first stage is due to monolayer adsorption of nitrogen to the walls of the mesopores at a low relative pressure (P/P 0 < 0.25). The second stage is characterized by a steep increase in adsorption (P/P 0 > 0.25). As the relative pressure increases, the isotherm exhibits a sharp inflection characteristic of capillary condensation within the uniform mesopores. The P/P 0 at the inflection is related to the diameter of the mesopore [15]    Gd-MCM-41 (100) catalysts are presented in Fig. 3. The initial weight loss up to 120C is due to desorption of physically adsorbed water and the weight loss from 120C to 350C are attributed to the organic template. The oxidative desorption of the organic template takes place at 180C and the minute quantity of weight loss above 350C to 550C is related to water loss from the condensation of adjacent Si-OH groups to form siloxane bond [11,17].

SEM and TEM analysis for the catalyst
The particle size and morphology of Gd-MCM-41 (100) sample was determined by SEM analysis and shown in Fig. 4.

EPR analysis
Gadolinium (III) ions are often used in paramagnetic complexes due to their good paramagnetism. The EPR spectrum with three and more absorption signals is usually assigned to isolated Gd 3+ ions, while a single broad absorption signal encompassing g = 2 is assigned to the group of Gd 3+ ions [19].
After crystallization the EPR spectra mainly consist of relative large line with g = 2.0 suggesting that the surrounding of Gd 3+ ions are experiencing weak crystal fields resulting from structural relaxation [20] which is shown in Fig. 6.  When the deposition temperature is higher than 550°C, the yield of CNTs is decreased up to 68.45 %. At higher temperature, the decomposition of C 2 H 2 promotes and consequently increases the concentration of carbon atoms, which can result in a dissolution rate that is higher than the rates for diffusion and precipitation. In that case carbon atoms will accumulate on the surface of the catalyst to form a carbon shell [22]. The catalyst particles then lose their activity and it retards the growth of CNTs.
Moreover, when the temperature is too high, a chemical reaction may take place between the carbon and the metal, leading to the formation of metal carbide, which is passivity the catalyst and then started dropping at approximately 600°C. At too high temperatures, the catalyst become highly mobile and quickly agglomerates into metal particles that are too large particles to initiate CNTs nucleation. These sintered large particles are rapidly encapsulated by carbon deposits and become inactive.
Hence 550°C was found to be optimum low growth temperature to achieve the very high yield and hollow structure of good respectively which may be ascribed to its greater ability to decompose unsaturated hydrocarbons. If the metal catalyst is exposed to acetylene, metal particles can easily diffuse on the surface to form larger particles, leading to MWCNT growth. The percentage of carbon yield for above mention catalysts are presented in Table 2, it can be clearly understood that support plays a prominent role in determining the dispersion and hence the catalytic activity of the metals in the production of CNTs.
Though Gd 3+ ion is toxic and when it is sequestered by chelation or encapsulation with another material, it reduced is toxicity therefore we utilized nanosized mesoporous siliceous material as a chelating agent in order to reduce the toxicity of gadolinium ion. Since, in our present study, we have used porous silica material to reduce the toxicity of metal and it would be an excellent carrier for the metal because of its porous nature.  bonding), since carbon nanotubes are known to be good transporters of water [23] and protons [24], whereas the active centers in gadofullerenes do not have this access. From a practical point of view, the rate of proton exchange is especially important since it contributes to the proton relativity [25]. These gadolinium carbon nanotubes are the contrast agent materials where superparamagnetic metal centers have access to many coordinated or exchanging water molecules per Gd 3+ ion. These Gd 3+ nanotubes species are linear superparamagnetic molecular magnets with MRI efficacies are more than any Gd 3+ based contrast agent in future clinical use. Figure 8 shows the XRD pattern of Gd containing

Characterization of carbon nanotubes XRD
MWCNTs synthesized at temperature of 550°C. The highest sharp peak at 2θ = 25.8° can be assigned to graphite (002), indicating that graphite layers are regularly stacked. While the other peaks at 44.3° (101) correspond to the remaining graphitic particles, a similar trend was observed by previous reporters [26]. Figure 9 shows the synthesized MWCNTs characterized by TGA in air. The weight loss is due to the combustion of carbon with oxygen and therefore, corresponds to the carbon content in the sample. The major mass loss observed in the temperature range of 375585°C due to the oxidation of carbon nanotubes, which is consistent with previous report [27].

SEM and TEM analysis
SEM image of CNTs prepared by Gd-MCM-41 (100) catalyst with an optimized condition of flow rate of acetylene and temperature were 40 ml/min and 550°C respectively, is shown in Fig. 10. The image clearly shows thinner nanotubes with metal particles at the tip of the tubes, and it also in agreement with TEM observations.
The TEM image (see Fig. 11a) indicates that CNTs grown are multiwalled and bamboo shaped. The structure shows that the CNTs grow with layers of graphite deposited on the surface of Gd catalyst were separated from the catalyst surface resulting in the growth of CNT with closed end (see Fig. 11b). However, if the deposition on Gd catalyst is too fast due to higher

Raman spectroscopy
Raman Spectroscopy is an important tool for investigating CNTs, which provides information about the crystalline nature of the sample. Figure 13 shows Furthermore, pyrolytic carbon particles deposited on nanotubes also contributed to the D-band intensity [33,34]. In the present system, G band is at higher intensity, indicating that the CNTs are graphitized, which is in agreement with the result of TEM observation.