11 Synthesis of aligned Carbon Nanotubes Using Spray Pyrolysis and Arc Discharge Techniques

Dr. Anchal Srivastava

Contents of this Unit

1.  Introduction

2. Synthesis of Carbon Nanotubes Using Spray Pyrolysis Technique

3. Structural and Morphological Study

4.  Carbon Nanotube Formation Mechanism

5. Factors that Affect CNTs Growth

5.1. Effect of Temperature on CNTs Formation

5.2. Effect of Carrier Gas Flow

5.3. Effect of Ferrocene Presence on Mist

5.4. Effect of Hydrocarbon/Ferrocene Ratio

5.5. Effect of Quartz Tube Length

5.6. Effect of External Electric Field on the Growth of Nanotubes

6.  Summary

Learning Outcomes

• After studying this module, you shall be able to
• Learn the synthesis of carbon nanotubes using spray pyrolysis technique. Learn the growth mechanism of carbon nanotube.
• Know, factors that affect CNTs growth such as precursor concentration, injection rate and duration, carrier gas flow rate, substrate surface, growth temperature, and the composition of gases inside the reactor

1.   Introduction

Spray pyrolysis is one of the most widely used methods for carbon nanotube synthesis. This is mainly because of its scalability and low- cost of the carbon nanotube produced. Spray pyrolysis is a promising method to create carbon nanotubes (CNTs) on various surfaces. By injecting metallocene-hydrocarbon solutions into a heated quartz reactor, catalyst particles and carbon nanotubes can be formed simultaneously. Factors that affect CNTs growth include the precursor concentration, injection rate and duration, carrier gas flow rate, substrate surface, growth temperature, and the composition of gases inside the reactor. In this module, the effect of these parameters on the morphology of CNTs synthesized by spray pyrolysis method is presented in detail.

The spray-pyrolysis methods for the carbon nanotubes (CNTs) synthesis, Fe clusters are formed on substrate walls and these clusters act as carbon trap. In the beginning of CNTs formation, the Fe clusters are surrounded by carbon. For the growing of the CNTs, the carbon is incorporated to the end of the growing tube and Fe remains at this end, which is moving away of surface. These iron nanoparticles diffuse the new carbon species that are incorporating to the CNT to radial and axial directions, conforming thus the multi-walled carbon nanotubes (MWCNTs).

2.  Synthesis of Carbon Nanotubes Using Spray pyrolysis Technique

The spraying system is illustrated in Fig. 1. A quartz tube was attached to a medical nebulizer, which was used to obtain a mist of small drops of hydrocarbon/catalyst. Argon was used as carrier gas to generate the ferrocene/benzene mist in the nebulizer. The quartz tube was heated by a cylindrical furnace, equipped with a very precise temperature control.Aguilar-Elguézabal, A., et al. reported that for all experiments the precursor solution was sprayed into quartz tube during 10 min. Nebulizer container was filled with 25 mL of the mixture of ferrocene and benzene. Argon was used as carrier gas and flow was regulated by mass flow controller (MFC). When temperature of furnace was stable, the argon was feed to nebulizer and the mixture argon/ferrocene/benzene entered to quartz tubing without any previous heating. Quartz tube gas exit was open to atmosphere and non-reacted benzene was measured by gas chromatographic analyses of samples taken from gas exit.

Fig. 1. Schematic diagram of the spray pyrolysis setup for the synthesis of MWCNTs.

Once quartz tubing was exposed to spray pyrolysis, furnace is cooled under argon flow to ambient temperature. Thereafter, black film made of CNTs formed at inner surface of quartz tube is removed mechanically with a brush or by immersion in acetone with ultrasound treatment.

3.      Structural and Morphological Study

As Aguilar-Elguézabal, A., et al. reported, the CNTs were formed on the inner surface of quartz with the appearance of uniform black film. Near to the gas outlet a reddish film is observed which denotes the oxidation of iron precursor, probably as a consequence of oxygen diffusion from the atmosphere during the furnace cooling. A set of experiments was selected in order to determine the influence of the main variables on the characteristics of the synthesized CNTs.

The Table 1 shows the results of the study at temperatures between 700 °C and 1000 °C and as can be seen, only at 900 °C CNTs were formed.

Table 1 Effect of control variables on the synthesis of CNTs

a Concentration of ferrocene on benzene.

b  Initially ferrocene/benzene mixture is sprayed during two minutes, the remaining 8 min only benzene is

sprayed into quartz tubing.

c  Ferrocene/benzene is sprayed during one minute, thereafter only benzene is sprayed during four minutes

and again this cycle is repeated to complete 10 min of treatment.

4 Carbon nanotube formation mechanism

During the spray pyrolysis, small droplets of ferrocene/ benzene enter to the quartz chamber, where the molecules of ferrocene and benzene are thermally cracked, and several reactions occur like dehydrogenation, ring condensation, benzene and cyclopentadiene ring opening, and iron atoms agglomeration among others. The CTNs formation is produced when Fe+2 is reduced to metallic Fe, which catalyze the hydrogen loss of benzene. Thus dehydrogenated benzene begin to bond to other dehydrogenated benzene rings to form the graphite wall of CNTs. At proper conditions, the precursor of CNTs are formed in the gas/vapor phase and consists of Fe particles surrounded by graphite layers as is shown in Fig. 2, then these precursors can reach the quartz surface and start the growing of a CNT or to incorporate to a CNT that is growing. The incorporation of graphite precursor (without Fe) to a growing CNT is also possible since according to TEM analysis, the core of CNTs is filled intermittently with Fe. Probably, the temperature of CNTs synthesis favors the aligning of graphite than the Fe atoms contained in a primary sphere of 50 nm, like the sphere observed in Fig. 3(a).

Fig. 2. Mechanism of CNTs growing from ferrocene/benzene drops in the mist fed to reaction chamber. After the typical reaction of pyrolysis on catalyst/ hydrocarbon mixture drop, discrete particles of iron/amorphous graphite are formed and thereafter these nanoparticles reach the surface of quartz or incorporate to a CNT that is growing. The irregular CNTs walls become aligned during the synthesis due to temperature.

So, some ferrocene and benzene located in the outer part of the drop leave the system, or these atoms form another species that are not incorporated to CNTs. As mentioned later, the diameter of CNTs is reduced when the gas flow of argon is increased, and this behavior supports the proposed mechanism, since at higher carrier gas flow there is a reduction on the drop size formed in nebulizer.

5.      Factors that affect CNTs growth

5.1 Effect of temperature on CNTs formation

Kamalakaran et al. and Zhang et al reported that a film made of amorphous carbon and iron is formed on quartz surface with a thickness lower than1 μm and an appearance of clouds for the synthesis of CNTs at 1000 °C. At this temperature, the hydrocarbon/catalyst molecules probably decompose by pyrolysis in the core of quartz tube and the rings of benzene molecules break, so the surface is reached by the cracked molecules in such a way, that it is not possible the reorganization of carbon in a graphitic structure and synthesis at 700 °C only some iron islands were found attached to quartz surface (substrate) and carbon also was detected in small amounts on the substrate surface. The absence of high amounts of carbon at substrate surface can be correlated with the high amount of hydrocarbon detected by gas chromatography in the gas stream that leaves the reaction chamber under 700°C. At 800°C the surface was covered by higher amount of carbon/iron spheres with diameter around ~0.4 μm. Kamalakaran et al. who used benzene and xylene, respectively as

 

Fig. 3. Typical TEM micrographs of CNTs. These micrographs are representative of the conditions where aligned and dense MWCNT were obtained at several conditions.

carbon source, nevertheless CNTs formation by pyrolysis is reported at 600 °C when instead of vapor, acetylene gas is used as carbon source. The effect of temperature on the CNTs diameter cannot be separated from chamber geometry and type of raw materials. In some experiments we found a constant diameter for several temperatures meanwhile others report that diameter is highly dependent on temperature using the acetylene as carbon source.

Andrews et al. reported the formation of CNTs at 650 °C using xylene/ferrocene mixture, and that results can be associated with their experimental set. For their work, they used a different quartz tube diameter and a different way to fed hydrocarbon/catalyst to reaction chamber, since they injected vapor in the chamber, and in our system a mist with small droplets is fed. Under these conditions, Andrews and col. obtained CNTs at that low temperature probably under a synthesis mechanism similar to the observed when acetylene is used as carbon source. One disadvantage of the use of vapor instead of mist can be a lower yield of CNTs produced from hydrocarbon. According to As Aguilar-Elguézabal, A., et al., the best temperature for the formation of CNTs was 900 °C. At this temperature, the formation of CNTs on quartz surface was as a layer with a homogeneous thickness, being the length of these CNTs from 120 to 140 μm. Fig. 4 shows the micrograph of the CNTs on quartz substrate at two different magnification scales.

Fig. 4. CNTs formed by spray pyrolysis at 900 °C under argon flow of 4 lpm, ferrocene concentration on benzene of 18.7 g/L during 10 min of solution spraying.

5.2 Effect of carrier gas flow

As Aguilar-Elguézabal, A., et al. the effect of argon gas flow in the CNTs growth was studied in the range of 3 to 6 liter per minute (lpm), maintaining constant the temperature (900 °C). It was observed that CNTs are formed within an optimal flow range. As can be seen in Table 1, the best results were obtained for argon flow of 4 and 5 lpm and under these conditions, the CNTs diameters were between 70 and 110 nm, meanwhile the CNT length was between 120 to 160 μm. For argon flow of 3 lpm, only carbon/iron microspheres of diameter below 0.5 μm are formed. When flow was above 5 lpm, the diameter of CNTs was smaller and ranged from 20 to 50 nm, and the length decreased about the half (i.e., 60 to 70 μm). As can be seen in Fig. 5, the top surface of CNTs layer was rough for argon flow of 5 and 6 lpm, whereas under an argon flow of 4 lpm the top surface was smooth. For an argon flow of 6 lpm, the precursor’s residence time begins to be short, and a higher amount of non-reacted benzene is obtained at gas exit.

Fig.5. Effect of argon gas flow during the synthesis of CNTs. (a) 3 lpm; (b) 4 lpm; (c) 6 lpm.

5.3 Effect of ferrocene presence on mist

Earlier works on CNTs synthesis reported that prior to CNTs synthesis, metallic or ceramic substrates were covered with metal thin films, and the feed of carbon source molecule under reduced atmosphere at temperatures from 600 to 1100 °C led to CNTs formation. According to Aguilar-Elguézabal, A., et al. results, there is not necessity to maintain the feeding of metal source during all the time of CNTs synthesis. To determine whether in our system is also possible to suspend the feeding of the metal source, two experiments were made in which ferrocene/benzene mixture was fed during two minutes of the experiment and only benzene during the remaining 8 min. In the first case, ferrocene/benzene mixture was introduced during two minutes and during the other 8 min only benzene solution was introduced. For the other experiment, the mixture was sprayed during one minute and thereafter the benzene solution during four minutes and then, again, during one minute the mixture of ferrocene/benzene was sprayed and during the remaining four minutes only benzene solution was sprayed. CNTs with length near to 20 μm were obtained when the spray of ferrocene/benzene mixture was maintained during two continuous minutes at the beginning of the experiment.

Fig. 6. SEM image of CNTs obtained by spray of ferrocene /benzene during two minutes and benzene for eight minutes. (a) Sample obtained by ferrocene/ benzene sprayed continuously during two minutes and followed by the spraying of only benzene for eight minutes; (b) CNTs obtained by the spraying of ferrocene/benzene during one minute, followed by benzene for four minutes, and again mixture during one minute and finally benzene four minutes.

For the experiment in which ferrocene/benzene mixture was sprayed intermittently during one minute at the beginning and another minute at the middle of the experiment, the CNTs length was of 4 μm, this lower length of CNTs indicate that apparently the second dosage of ferrocene/benzene mixture did not contribute to the formation of CNTs. The results obtained under the above mentioned conditions can be interpreted as if the CNTs were only formed when the mixture of ferrocene and benzene is being sprayed into quartz tube. This probably is due to amorphous carbon deposition on the top of the original CNTs layer during the intermediate stage of benzene flow. Fig.6 shows the CNTs formed uniformly for the two experiments, but with different length, the differences on diameter were determined with higher amplifications, and also by TEM analysis. According to these results, the mechanisms as the tip-growth and extrusion, S.B. Sinnott, et al. and Y. Tian, et al have been proposed the formation of nanotubes, (fig.2). Another conclusion is that the length of CNTs is directly related to the feeding time of ferrocene/benzene to reaction chamber at proper conditions.

5.4 Effect of hydrocarbon/ferrocene ratio

Three ferrocene concentrations were used for the CNTs synthesis at 900 °C as can be seen on Table 1. At the lower ferrocene concentration, the CNTs were scarcely formed, the growth was not perpendicular to surface and the lengths of CNTs were lower than 1 μm (Fig.7). The largest CNTs were obtained at the highest ferrocene concentration as can be seen in Fig. 6, (Aguilar-Elguézabal, A., et al.) and as in the case of high argon flow, top surface of CNTs was rough. For these conditions the diameter distribution was wide, and it ranged from 70 to 110 nm.

Fig.7. SEM analysis of the CNTs synthesized at ferrocene concentration of: (a) 9.3 g/L and (b) 36.2 g/L.

These results of Aguilar-Elguézabal, A., et al. confirm that iron from ferrocene is the key to induce the formation of the graphitic structure and thus, the CNTs formation. Bai et al. studied the influence of ferrocene/ benzene ratio and found that higher ratios produced CNT with lower diameter, but experimental set was quite different from the used by Aguilar-Elguézabal, A., et al. and results cannot be compared. The main difference is that hydrocarbon/catalyst is fed to reaction chamber as vapor instead of mist.

5.5 Effect of quartz tube length

An additional experiment was made by Aguilar-Elguézabal, A., et al. with the objective of improving the yield of ferrocene/benzene. For this experiment the length of quartz tubing was increased to expose the precursor mist to a larger quartz surface at 900°C, the expectations were to reduce to minimum the emission of non-reacted precursors through the exit of the gas. The study of the thin layer which was formed on extra quartz surface showed that most of surface was covered by clouds of carbon/iron mixture, and nanotubes were only formed at the beginning of the tubing. These results also indicate that the formation of CNTs is influenced by the degree of thermal cracking at which the precursors reach the surface where CNTs are growing. As drops travel along the reaction chamber and precursors reach the chamber temperature, dehydrogenation, cracking and condensation among other reactions take place with ferrocene and benzene. Only at the beginning of the chamber they formed species are able to reach the quartz surface in proper conditions to form the CNTs structures. Once the vapors have displaced more than the reaction chamber, the cracking stage of precursor molecules does not favor the formation of CNTs structure.

5.6 Effect of external electric field on the growth of nanotubes

Anchal Srivastava, et al. investigated the effect of external electric field on the growth of carbon nanotubules. Different electric fields corresponding to 3, 6, 9, 15, and 21 V have been applied(fig.8) On the application of 3 V the formation of bundle has been observed; however, the alignments of tubules in the bundle are not parallel. On further increasing the field corresponding to 6 V, the parallel alignment of the tubules have been observed.

Fig. 8. (a) Electron micrograph of bundle of carbon nanotubules produced as a result of application of electric field corresponding to 3 V (b) Electron micrograph showing the presence of parallel carbon nanotubule bundles produced by the application of electric field corresponding to 6 V (c) Representative electron micrograph of bundle of carbon nanotubules formed due to application of electric field corresponding to 15 V (d) Typical electron micrograph of the bundle showing random orientation of the tubules in the bundle synthesized by application of electric field corresponding to 21 V.

The alignment of the tubules parallel to each, other is thought to originate due to orientation of the tubule axis along the direction of the electrical field for an optimum value of the applied voltage (field). Higher impressed voltages, e.g., 15 and 21 V, as already described, do not lead to the formation of parallel tubules in the form of bundles.

Summary

• Spray pyrolysis of ferrocene/benzene mixture was successfully used to produce aligned MWCNTs inside of quartz tubing.
• It was found that for spray pyrolysis the size of drops (size of catalyst particle) from nebulizer determine the diameter.
• Length is mainly determined by the amount of ferrocene/benzene fed to reaction chamber at proper conditions.
• A novel mechanism is proposed for the production of CNTs by spray pyrolysis.
• Nanoparticles of graphitic carbon with iron core are formed in gas phase at furnace from the ferrocene/hydrocarbon droplets fed by nebulizer. These nanoparticles reach the surface and attach to quartz surface (substrate) or to the top of a nanotube increasing thus its length. The iron in the center of disordered nanotube have the tendency to migrate along the nanotube, and during the migration, iron induces the straightening of the nanotube.