MPA_Review of Carbon Nanotube Applications, Synthesis Methods and Processes for Mass Production

Fachhochschule Köln
Cologne University of Applied Sciences
SHEONGWEI NG
REVIEW OF CARBON NANOTUBE APPLICATIONS, SYNTHESIS
METHODS AND PROCESSES FOR MASS PRODUCTION

Referent: Thomas Rieckmann, Prof. Dr.-Ing.
Korreferent:
Fakultät für Anlagen-, Energie- und Maschinensysteme
Institut für Anlagen und Verfahrenstechnik
MASTERPROJEKT 2
Review of Carbon Nanotube Applications,
Synthesis Methods and Processes for Mass
Production
von SheongWei NG
Köln, 30.09.2015
Mat.-Nr.: 11107729

I
Summary
Numerous of works were carried out to have better understanding of the new carbon allotrope,
carbon nanotubes (CNTs) since their discovery in 1991. As the result of the works, CNTs have been
proven to possess high thermal and electrical conductivity, high mechanical strength, low density
and good chemical and environmental stabilities. These remarkable properties make them a
cutting edge material, which play an importance role in the society.
CNTs can already be found in our daily life products such as sport equipment, conductive
production, automotive parts and plastic reinforcement despite their production as bulk
unorganized architecture nanomaterials. So it can be said that CNTs have not yet fulfilled their
potential for the commercial products, which was tested and proven in the lab. Efforts are still
being committed to produce CNTs with desired properties and organized architecture at industrial
scale. Apart from that, research is still ongoing to explode the application of CNTs in other
technological areas such as catalyst and catalyst support, energy production and storage, and
medicine. And those works have showed promising results for the replacement of conventional
materials with CNTs.
Different methods of CNT synthesis was introduced for the past two decades, but only three are
being widely used, specifically arc discharge, laser ablation and chemical vapor deposition (CVD).
Modification was done to further improve their yield, quality, energy efficiency as well as
construction and operation cost. Both the arc discharge and laser ablation generally require high
energy input for the production, but they produce high quality of CNTs, which are crucial for the
research. The major drawbacks for their application in industrial are the low energy efficiency and
batch type nature. On the other hand, CVD methods can be operated continuously at relative low
temperature and moderate pressure. Beside that the diameter, length and alignment of the CNTs
produced by CVD methods can be easily controlled through variation of synthesis conditions.
Those advantages make CVD methods best-suited method for large-scale production of CNTs.
Scaling up the lab scale equipment to industrial scale apparatus for the production of CNTs had
been a problem. But the problem was addressed by many researchers and several processes were
purposed and developed for large-scale production. Notably, high-pressure carbon monoxide
(HiPco) process, cobalt-molybdenum catalytic (CoMoCAT) process, Endo’s process, multiwall
nanotube process and Baytube process are those CVD processes currently used in industrial for
commercial purpose. Reaction temperature, pressure, catalyst size, morphology and composition,
reactant concentration and flow rate, type of carbon feedstock along with type of reactor used
are the essential considerations, which have to be made for the implication of those processes for
sustainable manner of high yield, selective and low cost production of CNTs in industrial.

II
Content
Content
1 Introduction..................................................................................................................................... 1
2 Type and growth mechanism of carbon nanotubes ....................................................................... 3
2.1 Single-walled carbon nanotubes (SWNTs) ............................................................................... 3
2.2 Multi-walled carbon nanotubes (MWNTs)............................................................................... 4
2.3 Growth mechanism of carbon nanotubes................................................................................ 5
3 Applications of carbon nanotubes .................................................................................................. 6
3.1 Composite materials ................................................................................................................ 6
3.2 Catalyst and catalyst support................................................................................................... 7
3.3 Energy storage.......................................................................................................................... 8
3.4 Environment............................................................................................................................. 9
4 Synthesis methods ........................................................................................................................ 10
4.1 Physical process...................................................................................................................... 10
4.1.1 Arc discharge................................................................................................................... 10
4.1.2 Laser ablation .................................................................................................................. 11
4.2 Chemical process.................................................................................................................... 12
4.2.1 Chemical vapor deposition (CVD).................................................................................... 12
4.2.2 Modification of CVD ........................................................................................................ 13
4.3 Comparison between the physical and chemical process ..................................................... 14
5 Process for mass production......................................................................................................... 15
5.1 High-pressure carbon monoxide (HiPco) process .................................................................. 15
5.2 Cobalt-molybdenum catalytic (CoMoCAT) process................................................................ 18
5.3 Endo’s catalytic chemical vapor decomposition (CCVD)........................................................ 20
6 Conclusion ..................................................................................................................................... 22
Literature and Reference ................................................................................................................. 23

III
Symbols and indices
Al2O3 Aluminum oxide
Ar Argon
C2H2 acetylene
C2H4 Ethylene
C2H6 Ethane
C4H10 Butane
C4H6 Butadiene
CH4 Methane
CNTs Carbon nanotubes
CO Carbo monoxide
Co Cobalt
CO2 Carbon dioxide
CoMoCAT Cobalt-molybdenum catalyst
Cu Copper
Fe Iron
Fe(CO)5 Iron pentacarbonyl
H2 Hydrogen
H2S Hydrogen sulfide
HiPco High-pressure carbon monoxide
Li Lithium
MgO Magnesium oxide
Mn Manganese
Mo Molybdenum
MWNTs Multi-walled carbon nanotubes
N2 Nitrogen
NaOH Sodium Hydroxide
NH3 Ammonia
Ni Nickel
NiS2 Nickel disulfide
NO Nitric oxide
O2 Oxygen
ODH Oxidative dehydrogenation
ORR Oxygen reduction reaction
Pd Palladium
PECVD Plasma enhanced chemical vapor deposition
Pt Platinum
S Sulfur
SiC Silicon carbide
SiO2 Silicon dioxide
SWNTs Single-walled carbon nanotubes
Ti Titanium
WACVD Water assisted chemical vapor deposition
wt% Weight percent
Zn Zinc

1
1 Introduction
It was well known that charcoal, graphite and diamond are the 3 main allotropes of carbon. But
not until the discovery of Buckminsterfullerene in 1985 (Kroto, et al., 1985), carbon nanotubes
(CNTs) in 1991 (Iijima, 1991) and graphene in 2004 (Novoselow, et al., 2004), which brought
carbon allotropes to nanoscale. They all have the same structural unit: a single-layer graphene
sheet, but exist in different forms. This unique individual morphology provides them special
characteristics and properties, which are distinct from each other. These discoveries and their
possible applications have great impacts on material science and our daily life. Figure 1-1 shows
the 3 newly found carbon allotropes and how they are interconnected.
Figure 1-1: Illustration of Buckminsterfullerene, CNTs and graphene. The white arrows show how
the carbon allotropes are linked (Tessonnier, et al., 2011).
Particularly, CNTs have been receiving a lot of attentions due to their extraordinary properties
such as high thermal and electrical conductivity, high mechanical strength, low density and good
chemical and environmental stability. For example, CNTs are capable to carry an electrical current
density of 4 x 10-9
A cm-2
(Hong, et al., 2007) and their thermal conductivity is 3500 W m-1
K-1
at
room temperature (Pop, et al., 2006). These magnitudes are 1000 and 10 times higher than
copper electrical and thermal conductivity respectively. These interesting properties of CNTs
caused the journal publications and issued patents related to CNTs rise annually and their
production capacity increase exponentially since 2004, as shown in Figure 1-2.
CNTs are currently mass produced as unorganized bulk nanotubes, which have limited properties
compare to those lab-scale synthesized CNTs. Despite that CNTs have been commercialized and
applied in diverse commercial products such as sport equipment, conductive production,
automotive parts, plastic reinforcement, scaffolding for bone growth and so forth. But there is still
ongoing research to discover and develop new application area of CNTs, for instance catalyst and
catalyst support, medicine, energy production and storage, and so on. Mass production of CNTs
with desired and organized structure is one of the focuses of the research.

2
Figure 1-2: Annual number of journal publications, issued patents and production capacity of CNTs
(De Volder , et al., 2013)
Their outstanding properties and wide range of applications on commercialized products draw the
interests of the scientists and researchers to develop various synthesis methods and investigate
the effects of the synthesis parameters for better production or modification of CNTs with special
properties. Arc discharge, laser ablation and chemical vapor deposition (CVD) are the three main
developed CNT syntheses. CNTs synthesized with arc discharge and laser ablation require high
energy input, which limit them from being used for continuous industrial production. On the
other hand, CVD-based methods can be operated under mild conditions and the ease of
controlling the CNT properties through variation of parameters, make them the focus of research
for large-scale production. The biggest challenge for the mass production of CNTs is to synthesize
them at industrial scale with high yield, low cost and in sustainable manner.
Among the 3 newly discovered carbon allotropes, only CNTs have reached large-scale industrial
production in order to meet their high market demand (Thayer, 2007). Numerous of scalable CVD-
based processes have been developed, modified and adopted into industrial production, including
HiPco process from Rice University, CoMoCAT process from University of Oklahoma, Endo process
from Shinshu University, Nano agglomerate fluidized process from Tsinghua University, Multiwall
nanotube process from Hyperion Company and Baytube process from Bayer Material Science
(Zhang, et al., 2011). The market value for CNTs in 2010 was estimated to be 90.5 million dollar
and global revenues are projected to exceed 1 billion dollars by 2015 (Apul, et al., 2015).
This review will highlight three different aspects of CNTs, namely applications, synthesis methods
and mass production processes of CNTs. First, examples for promising present and future
applications of CNTs, which are related to chemical and process engineering, will be extracted
from literatures and elaborated. In the next section, the three main synthesis methods of CNTs
are illustrated and explained. And their modification progresses will be addressed as well. Beside
that advantages and disadvantages for large-scale production of each method will also be
reviewed. Finally, the three mass production processes, which are currently used in industrial for
large-scale production, will be explained and the considerations in the design of industrial scale
production for each process will be introduced. And further discussion of the processes will be
supported by the parametric study results, which are extracted from the literatures.

3
2 Type and growth mechanism of
carbon nanotubes
Theoretically, CNTs are cylindrical structure made by rolling one-atom-thick graphite, called
graphene, into a seamless cylinder. If the tube is made out of one cylindrical graphene, it is called
single-walled carbon nanotubes (SWNTs). But if the tube system is made out of more than one
layer of cylindrical graphene, it will be named as multi-walled carbon nanotubes (MWNTs). The
synthesis of SWNTs and MWNTs can be differentiated through the reactor used, reaction
parameters, types of catalyst and also the size and morphology of the catalyst. For instant, small
catalyst particles (0.5-5 nm) are mostly used to synthesize SWNTs with CVD method, whereas big
catalyst particles (8-100 nm) are favored for the synthesis of MWNTs (Zhang, et al., 2011). The
different between the 2 types of CNTs are illustrated in Figure 2-1.
Figure 2-1: Schematic representation of SWNT and MWNT (Martins, et al., 2013).
Apart from being a straight SWNTs or MWNTs, many attempts have been conducted to vary the
tubule morphology of CNTs since its discovery in 1991, for example: waved, coiled, regularly bent,
branched CNTs and CNTs with nanobud. The reason behind the attempts is the special properties
and potential applications that come along with the special designed tubule morphologies.
2.1 Single-walled carbon nanotubes (SWNTs)
As was mentioned in the previous section, SWNTs are cylindrical tube made by wrapping a one-
atom-thick layer of graphite. It was first discovered by Iijima and Ichihashi in 1993, 2 years after
the discovery of CNTs, by arc discharge method with iron (Fe) as catalyst (Iijima, et al., 1993).
SWNTs can be categorized into 3 different forms such as zig-zag, armchair and chiral, depending
on the way the graphene is wrapped (see Figure 2.1-1). And the properties of the SWNTs,
especially their electrical properties, are strongly depending on the existed form. The armchair
form of SWNTs is considered as metallic or highly conducting nanotubes, whereas other forms can
make the SWNTs as semiconductor (Eatemadi, et al., 2014). Owning to these properties, SWNTs
are mostly used in electronic devices and sensors, which require highly structured CNTs. But the
price of SWNTs remains higher than MWNTs due to their complex mass production process,
which limits them from widespread applications.

4
Figure 2.1-1: The three different forms of SWNTs (Eatemadi, et al., 2014)
2.2 Multi-walled carbon nanotubes (MWNTs)
MWNTs consist of more than one layer of cylindrical shape of graphene and were first obtained
by Iijimal in 1991 with arc discharged method (Iijima, 1991). They have diameter ranging from few
nanometers up to several hundred nanometers depending on the number of layers the system
possesses. There are 2 commonly used models for the description of MWNTs, namely Russian Doll
model and Parchment model. In the Russian Doll model, MWNTs are a set of SWNTs with the
largest diameter at the outer most layers and the diameter of the SWNT decreases when moving
to next inner layer until the inner most layers, which has the inner diameter of the MWNT. In the
Parchment model, a single graphene sheet is rolled in around itself for manifold times, resembling
a scroll of parchment or a rolled paper. The Russian Doll model is more commonly observed.
Double-walled carbon nanotubes (DWNTs) are a special form of MWNTs, which contain only 2
layers of SWNTs. They were first synthesized in 2003 using catalytic chemical vapor deposition
(CCVD) method (Flahaut, et al., 2003). DWNTs have identical morphology and properties like
SWNTs. But their chemical and environmental stabilities are significantly better than SWNTs due
to the protection from the outer layer. The outer layer of DWNTs provides extra CNT layer for
functionalization, which add new properties to the CNTs without destroying the tubule
morphology of the inner SWNT and their properties.
Following table shows some differences between SWNTs and MWNTs.
SWNTs MWNTs
Single layer of graphene Multiple layers of graphene
Catalyst is requires for synthesis Can be produced without catalyst
Bulk synthesis is difficult as it requires proper
control over growth and atmospheric condition
Bulk synthesis is easy
Purity is poor Purity is high
A chance of defect is more during
functionalization
A chance of defect is less but once occurred it
is difficult to improve
Characterization and evaluation is easy It was very complex structure
It can be easily twisted and is more pliable It cannot be easily twisted
Table 2.2-1: Comparison between SWNTs and MWNTs (Eatemadi, et al., 2014).

5
2.3 Growth mechanism of carbon nanotubes
Despite many studies have been done for better understanding of the CNT growth mechanism
and several possibilities have been proposed, the actual CNT growth mechanism is yet to be
discovered and established. Generally, the widely-accepted mechanism can be considered as a
three-step process. Firstly, the carbon feedstock will decompose into elementary carbon atoms
on the surface of the catalyst particles. Secondly, the carbon atoms will either diffuse through or
diffuse on the side of bulk catalyst particles. The latter is mostly accepted because it explains the
hollow core of the CNTs (Tessonnier, et al., 2011). The main driving force of the diffusion was
suggested to be the temperature and concentration gradient (Harris, 2009). Finally, the carbon
atoms will precipitate out and form the cylindrical network of CNTs on the surface of the catalyst
particles.
During the final step of the process, there are 2 possible growth models, which can take place
depending on the strength of the interaction between the catalyst particles and the substrate,
namely tip-growth and base-growth models. If the catalyst-substrate interaction is weak, the
carbon atoms are capable to push the catalyst particle off the substrate and it will stay on top of
the CNTs while the growth process continues on the bottom part of the particle. This
phenomenon is called tip-growth model. On the other hand, if the catalyst-substrate interaction is
too strong to be detached by the carbon atoms, the catalyst particle will remain anchored on the
surface of the substrate while the CNT growth process continues on the top part of the particle.
This phenomenon is knows as base-growth models. In both models, the growth process will be
terminated once the catalyst particle is fully covered with excess carbon and subsequently their
catalytic activity will cease. The widely-accepted growth mechanisms of CNTs are illustrated in
Figure 2.3-1.
Figure 2.3-1: Illustration of the general CNT growth mechanism: (a) tip-growth model, (b) base-
growth model (Kumar, 2011).
The mechanism discussed above is the basic and widely-accepted CNT growth mechanism. The
uncertainness of the physical and chemical state of the catalyst particles during the growth and
the mode of diffusion during the second step of the mechanism are still troubling the researcher
before the establishment of correct and accurate CNT growth mechanism can be conclusively
done (Kumar, 2011).

6
3 Applications of carbon nanotubes
The number of CNT applications related articles is still increasing despite the decrease of the
number of articles related to CNT synthesis since 2009 (Zhang, et al., 2011). This phenomenon
shows that the focus point of the studies about CNTs is shifting from synthesis to application,
especially for the application in catalyst, energy and environmental area. Despite the existing
wide range of applications, research about the new application and usage of this new type of high
performance carbon nanomaterial is still ongoing due to the market demand.
The applications of the nanotubes can be categorized into 2 groups, namely large and limited
volume applications. Large amounts of CNTs of good quality are needed for the large volume
applications e.g. as components in conductive, electromagnetic, high strength composites,
supercapacitors, fuel cell catalyst and transparent conducting films. While high structure and
reproducibility standards are required for limited volume applications, such as drug delivery
system, electronic devices and sensors.
Despite the annual substantial production of CNTs, the applications of CNTs in commercialized
products have mostly been limited to the use of bulk CNT powders, which is a mass of
unorganized fragment of nanotubes. Specific structures and agglomeration states are required for
many CNT applications. Bulk nanotubes might not be showing the similar properties as the
organized CNT architectures, which were tested in lab. Nonetheless, the bulk CNT powders still
yield promising performance for commercial applications compare to conventional materials
used.
In the following section, the promising present and future applications of CNTs related to
chemical and process engineering will be focused on four different application areas, namely
composite materials, catalyst and catalyst support, energy storage, and environment.
3.1 Composite materials
Since the first report regarding the preparation of CNTs and polymer composite materials in 1994
(Ajayan, et al., 1994), many efforts have been made to composite the CNTs with others polymer in
order to produce desired functional composite materials. The difficulty in structure control, poor
process ability and existence of impurities remain as the main challenges for the application of
individual bulk CNTs. Compositing CNTs with other polymers seems to be one of the solutions to
the problems by further enhancing the properties of the bulk CNT materials.
CNTs can be composited with polymer through functionalization. Figure 3.1-1 shows some
possible functionalization mechanism for SWNTs.
A) Functionalization of the defect group at the end of the tubes and side walls;
B) Functionalization of the covalent side wall through addition reactions and subsequent
nucleophilic substitution;
C) Functionalization of noncovalent exohedral with surfactants;
D) Functionalization of noncovalent exohedral with polymers;
E) Functionalization of endohedral with C60.

7
Figure 3.1-1: Some possible ways of functionalization for SWNTs (Hirsch, 2002).
CNTs are composited with others polymers to form conductive CNT polymer, for example
composition of CNTs with poly (p-phenylenevinlene-co-2,5-dicotoxy-m-phenylenevinylene)
(PmPV) showed a very high conductivity due to the conducting path provided by the CNTs to the
polymer (Coleman, et al., 1998). The conductive CNT polymers have been used in automobile
industries for electrostatic-assisted painting of mirror housing, as well as fuel lines and filters that
dissipate electrostatic charge (De Volder , et al., 2013). Furthermore, attempts are in progress to
composite CNTs with aluminum for the advanced lightweight of automobile parts (Peng, et al.,
2015) and with carbon fibers for lightweight wind turbine blades (De Volder , et al., 2013).
3.2 Catalyst and catalyst support
Currently there is no concrete industrial application of CNTs as catalyst and catalyst support, but
research has proven that CNTs have the potential to replace metals as catalyst and catalyst
support for many organic and inorganic reactions. Better environmental acceptability, favorable
management of energy with good thermal conductivity and inexhaustible resources make carbon-
based materials such as CNTs an interesting alternative to some current industrialized chemical
process (Su, et al., 2010).
Many studies have been conducted to replace conventional metal and metal oxide catalyst with
CNTs for the oxidative dehydrogenation (ODH) of unsaturated hydrogen carbons and alkane
activation, oxygen reduction reaction (ORR) and the transesterification of triglycerides. For
instance, surface functionalized CNTs showed an increase of selectivity for the ODH of ethane
(C2H6) due to the suppression of electrophilic oxygen (O2) intermediates on the carbon surface
(Frank, et al., 2010). In another study, Butadiene (C4H6) was efficiently catalyzed by surface
modified CNTs during the ODH of n-butane (C4H10) due to the capability of CNTs to keep O2/ C4H10

8
at low ratio during the reaction (Zhang, et al., 2008). Apart from ODH, studies showed that CNTs
are capable of reducing (Matsumoto, et al., 2004) and even replacing (Gong, et al., 2009) the high
price and limited supply platinum (Pt), which is conventionally used as catalyst for ORR. Alkaline
earth oxides, calcined hydrotalcites and nano-magnesium oxide (MgO) are the classical
heterogeneous basic catalyst used for biomass conversion. Small surface area and partial
dissolution into reaction media make them a drawback for the application. But CNTs seem to be
an alternative solution for the problem, as study proved that amino group grafted MWNTs show
high activity and stability in transesterification of triglycerides (Villa, et al., 2009).
Besides being used as catalyst, CNTs can also be applied as catalyst support and their
performances are promising and even better than the traditional catalyst supports such as metal,
metal oxide and active carbon. The high mechanical strength of CNTs makes them suitable to be
used in mechanically taxing stirred batch reactor. Their high surface area and inherently
microporous provide a better place for dispersion and impregnation of catalyst. And they have a
longer life span than conventional catalyst supports due to their high chemical and environmental
stability. MWNTs supported catalytic nickel disulfide (NiS2) nanoparticles showed better
desulfurization activity and better resistance to the solid sulfur (S) deposition of selective
oxidation of hydrogen sulfide (H2S) into elemental S compare to silicon carbide (SiC) supported
NiS2 nanoparticles (Nhut, et al., 2004). Another study showed that catalytic palladium (Pd)
nanoparticles supported on CNTs showed higher selectivity for the oxidation of benzylic alcohol to
benzaldehyde in comparison to activated carbon (Villa, et al., 2010). Furthermore, CNTs
supported catalytic Pd also showed promising result for the hydrogenation of alkene, alkyne and
nitric oxide (NO) as well as conversion of nitro to amino group (Oosthuizen, et al., 2011).
3.3 Energy storage
Owning to their high chemical stability, electrical conductivity, surface area, electrolyte
accessibility and low resistance of charge transport, CNTs have been used as anode materials for
lithium (Li) ion batteries, which can be found in notebook computers and mobile phones. When
MWNTs are assembled in bundles, the interlayer space of MWNTs provides the room for the
storage of large amount of Li+
ion (Leroux, et al., 1999). A study reported that the defection on the
CNT wall by nitrogen atom (N) doping further increased the storage capacity of MWNTs, as larger
portion of interwall space is available for the storage of Li+
ions (Shin, et al., 2012).
Pt has been used as catalyst to improve the performance of fuel cell. The major drawback factor
for large-scale practical applications of fuel cell is the high price and limited supply of Pt. Reducing
the amount of Pt used in fuel cells is an essential step for commercialization of fuel cells as energy
source. A study showed that the usage of Pt can be reduced as much as 60 %, if CNTs are used as
catalyst support instead of carbon black. The reasons behind the improved performance are the
formation of triple phase boundaries of the electrode and the high conductivity of CNTs
(Matsumoto, et al., 2004). A recent study even reported that fuel cell can have a better
performance, when doped CNTs were used as electrode without the present of Pt as catalyst
(Gong, et al., 2009).
Apart from storing electrical energy, CNTs were reported to be capable for hydrogen (H2) storage.
The capillary effects of the small size CNTs provide space for high density of condensation of H2
gas inside SWNTs (Jones, et al., 1997). Thus, the H2 can be stored as gas phase instead of liquid

9
phase. It was believed that current storage methods, which store H2 in liquid phase, can be
possibly replaced by CNT-based method. Because the major problem with available methods is
the potential energy lost during the cooling and condensation of H2 gas. But it was reported that
CNTs have a maximum hydrogen uptake capacity of only 0.2 wt% (Barghi, et al., 2014), which is
significantly lower than commercially available hydrogen storage need to be. Arguably, it is
because of the impurities present in the CNTs. But even the very pure sample of CNTs, which
were purified with the help of Microwave digestion method, showed only a maximum capacity of
3.7 wt% (Yuca, et al., 2011). Yet the value is still lower than the targeted value of 5.5 wt% required
by U.S. Department of Energy (DOE) for automotive application (Froundakis, 2011). Researches
are still ongoing to overcome the problems caused by limited H2 uptake capacity of CNTs.
3.4 Environment
Adsorbent in water purification is an upcoming application for the CNTs. For example,
commercialized portable filters now contain CNT meshes to purify contaminated drinking water
(De Volder , et al., 2013).
Beside commercial applications, CNTs are potential adsorbent for wastewater treatment in
industrial due to their hollow and layered structures, high specific surface area, and
hydrophobicity. Many reports have showed that CNTs are capable of removing natural organic
matter and synthesis organic contaminants through adsorption, for example polycyclic aromatic
hydrocarbon, benzene derivatives, phenolic compounds, pharmaceuticals, polychlorinated
biphenyls, dialkyl phthalate esters, protein, organic dyes and dioxin (Apul, et al., 2015).
Furthermore, current adsorbents for metal ions have low adsorption capability and removal
efficiencies. With the unique properties of CNTs, they might be an alternative solution for the
problems (Rao, et al., 2007).
The drawback of the instant industrial application of CNTs in this field can be considered in two
different perspectives, namely nature system and engineering. From nature system perspective,
CNTs might enter the environment through the wastewater treatment either intentional or
unintentional and their toxicity can be enhanced by the adsorbed organic contaminant (Apul, et
al., 2015). This raises the concept about the risk of human health, as reports showed that CNTs
are capable of causing cell death due to the their accumulation after entering human cell (Porter,
et al., 2007) and causing side effects to human lungs (Lam, et al., 2006). Whereas from the
perspectives of engineering application, study showed that with the present of natural organic
matter, microporous activated carbon fiber and granular activated carbon have better adsorption
capacity of synthesis organic contaminants such as phenanthrene and 2-phenyl-phenol in
comparison to CNTs (Zhang, et al., 2011).

10
4 Synthesis methods
The research on the fascinating science and technology of CNTs is mainly promoted by the
development of controllable synthesis methods, which provide more desired samples for
investigation purpose. As many applications require CNTs to have specific structures and
agglomeration states, studies are still ongoing to discover new methods and further modify the
existing methods in order to meet the needs. Numerous methods were purposed for the
production of CNTs since their discovery and the methods can be classified into following 2 major
groups:
1. Physical process
2. Chemical process
In the following section, three widely used methods, namely arc discharge, laser ablation and
chemical vapor deposition, will be presented individually and discussion about their modification
will be made.
4.1 Physical process
4.1.1 Arc discharge
The first ever used method for the synthesis of CNTs is the arc discharge method (Iijima, 1991).
Figure 4.1.1-1 shows a typical illustration of arc discharge method. Normally, graphite rods will be
used as the electrode for anode and cathode in an enclosed chamber, which will be pressurized
with inert gas like helium (He) and argon (Ar) at a given pressure. The electrodes are connected to
a voltage stabilized direct current (DC) power supply. The adjustable anode will be moved closer
to the cathode until an arc appears. The arcing gap between the electrodes should be constantly
kept at approximately 1 mm or less during the synthesis. The current will discharge the carbon
from the anode and the evaporated carbon atoms will recondensed as CNTs on the cathodic rod.
The synthesis will only last for few minutes and the anodic rod has to be replaced as the rod will
be consumed during the process. This prohibits the continuous production of CNTs. The arc
discharge method typically generates deposits on cathodic rod at the rate of 20-100 mg min-1
(Kingston, et al., 2003).
Figure 4.1.1-1: Arc discharge apparatus for the synthesis of CNTs (Saito, et al., 1996).

11
Efforts have been committed to produce good yield of high quality CNTs through arc discharge
method by using catalyst and changing the synthesis parameters and conditions. Studies were
conducted using pure metal as catalyst for the synthesis. Cobalt (Bethune, et al., 1993) and iron
(Iijima, et al., 1993) was impregnated on holed anodic graphite electrode and as a result SWNTs
were produced. Synthesis parameters such as pressure of the chamber can alternate the quality
of the CNTs produced. A study showed that the number of layer of CNTs can be increased as the
pressure of He in the pressurized chamber increased. But after 66.66 kPa (500 torr) there was no
change in sample quality but decrease in total yield (Ebbesen, et al., 1992). Apart from the
reaction pressure, current is another factor that might change the quality of the CNTs. The current
should be kept as low as possible and the stable plasma state should be maintained because the
low current prevents the formation of hard and sintered materials, which lead to low yield of
CNTs (Ebbesen, et al., 1993). Another possibility is to replace the inert gas pressurized chamber
with liquid. Studies showed that CNTs produced by arc discharge under liquid nitrogen and water
have higher quality than those produced under gas (Antisari, et al., 2003).
4.1.2 Laser ablation
Another physical process for the synthesis of CNTs is laser ablation process, which was introduced
in 1995 (Guo, et al., 1995). It was claimed that CNTs produced with this method have higher yield
and purity, and this method has better control over growth conditions compare to arc discharge
method. The apparatus set up is illustrated in Figure 4.1.2- 1. The graphite target, which contains
small amount of cobalt (Co) and nickel (Ni), will be struck by laser beam in a high temperature
reactor. The CNTs will form on the lower temperature region of the reactor as the vaporized
carbon atoms condense. The tubular reactor will be filled with continuously flow of inert gas such
as He and Ar to create inert atmosphere and carry the grown CNTs to the water-cooled copper
collector. This method was further refined by using double pulsed laser for even vaporization of
the graphite and minimization of formation of soot on the collector (Thess, et al., 1996).
Figure 4.1.2- 1: Schematic of laser ablation apparatus for synthesis of CNTs (Yakobson, et al.,
1997).

12
The variation of average diameter, length, structure and yield of the CNTs can be done by
changing the process parameter such as temperature, laser used and catalyst composition.
SWNTs was successfully produced at room temperature by using 1 kW of carbon dioxide (CO2)
laser beam for the vaporization of graphite target, which contain small amount of Co and Ni
particles (Kokai, et al., 1999). The results showed that the yield of SWNTs was increased as the
temperature increased and the highest yield was recorded at 1200 °C. Production rates of SWNTs
as high as 1.5 g h-1
was reported by using 1 kW of free electron laser for the vaporization of metal
particles loaded graphite. Integration of free electron laser with this method is capable of
producing SWNTs at the rate of 45 g h-1
after process optimization (Eklund, et al., 2002).
4.2 Chemical process
4.2.1 Chemical vapor deposition (CVD)
Use of CVD method to synthesize MWNTs was first reported in 1993 by catalytic decomposition of
acetylene (C2H2) at temperature of 700 °C and graphite supported Fe was used as catalyst (Jose-
Yacaman, et al., 1993). Three years later, SWNTs were successfully synthesized through Mo
catalyzed disproportionation of CO at 1200 °C (Dai, et al., 1996). A typical set up for CVD method
is illustrated in Figure 4.2.1-1. The mixture gas of hydrocarbon gas, which acts as the carbon
feedstock such as carbon monoxide (CO), methane (CH4), ethane (C2H6), ethylene (C2H4) and
acetylene (C2H2), and process gas, which acts as carrier gas such as ammonia (NH3), nitrogen (N2)
and hydrogen (H2), will be fed into the reaction chamber. The decomposition of hydrocarbon gas
takes place in the reaction chamber and carbon atoms deposit and growth on the catalyst loaded
substrate at the temperature ranging from 400-1200 °C. Because of its higher yield and simpler
equipment compared to arc discharge and laser ablation, CVD is the most promising method for
large-scale production of CNTs.
Figure 4.2.1-1: Illustration for typical CVD set up (Mubarak, et al., 2014).
As production of SWNTs by CVD usually required high temperature (900-1200 °C), CO and CH4 are
used for carbon feedstock due to their high thermal stability. Apart from temperature and
feedstock, there are another two key factors that affect the nature and types of CNTs produced
by CVD method, namely the catalyst used and the preparation of the substrate. Silicon and glass
are normally used as the substrate material. And study showed that Ni, Fe and Co-based catalyst

13
are the most active catalyst for the decomposition of hydrocarbon in comparison to others
transition metals such as manganese (Mn), copper (Cu), zinc (Zn) and titanium (Ti) (Deck, et al.,
2006). Solution deposition, electron beam evaporation and physical sputtering are commonly
used to deposit the catalyst particles on the substrate material. These deposition methods have
to be chosen specifically for the production of desired CNTs, as the methods have influences on
CNT properties.
The carbon atoms are deposited on the surface on the catalyst particles and CNTs grow on them.
The particles will be encapsulated inside the CNTs after the termination of the growth and these
transition metals are proven to have significant influences on the CNT properties (Brukh, et al.,
2008). Silicon dioxide (SiO2) seems to be the alternative catalyst for the metal free CVD synthesis.
A study showed that SiO2 nanoparticles are capable of catalyzing the synthesis of SWNTs and their
size distribution is much narrower (Huang , et al., 2009).
4.2.2 Modification of CVD
Owning to the ability for continuous operation, simplicity for scaled up to large industrial process,
availability for abundant of raw materials and simplicity of reactor design, CVD is a promising
synthesis method for CNTs to meet the high market demand. And CVD is the only known method
for producing aligned CNTs (Rafique, et al., 2011). Numerous of researches have been done to
enhance the method for higher CNT production yield and better architecture CNTs.
CNTs produced by CVD are randomly entangled. And this limits their applications as electrodes or
electrodes filler in energy conversion and energy storage, as structurally aligned CNTs are critical
for the applications (Zhang, et al., 2011). Plasma enhanced chemical vapor deposition (PECVD)
was first successfully used to synthesize aligned CNTs at 666 °C (Ren, et al., 1998). In this work,
C2H2 was used as carbon feedstock and the CNTs were grown on Ni deposited glass. NH3 gas was
introduced into the reaction as dilution gas and it showed catalytic activity on the CNT growth.
The reason behind the conformal alignment is believed to be the electrical self-bias imposed on
the substrate surface from the plasma environment (Bower, et al., 2000). A further study showed
that higher plasma power will slow down the growth rate of CNTs due to the rapid decomposition
of carbon feedstock C2H2 at high plasma power (Bell, et al., 2006). The C2H2 has to be slowly
decomposed to prevent formation of amorphous carbon. Besides that the ratio between NH3 and
C2H2 is another crucial factor for high quality and quantity of CNTs. Higher ratio of NH3 to C2H2 is
favored as NH3 generates atomic hydrogen species to remove excess carbon and suppresses the
decomposition of C2H2 due to its weaker molecular chemical bonds (Hussain, et al., 2015).
Another modified CVD is called water assisted chemical vapor deposition (WACVD), which was
introduced in 2004 to synthesize SWNTs (Hata, et al., 2004). In this work, C2H2 was used along
with H2 and Ar or He, which contained small and controlled amount of water vapor. In the normal
CVD synthesis, the amorphous carbon formed will coat on the catalyst particles and cause the
reduction of their catalytic activity and lifetime. The results of the study showed that water can
promote and preserve the catalytic activity for a longer period of time. The reason behind it is the
ability of water to produce large amount of hydroxide groups on carbon, which convert the
deposited carbon to CO and H2 by gasification and subsequently inhibit the catalyst from ripening
(Xie, et al., 2013).

14
4.3 Comparison between the physical and chemical process
Physical processes like arc discharge and laser ablation generally have to be conducted in
advanced and costly apparatus at very high temperature. And due to the fast process time only
production of short and low yield of CNTs is possible as well as constantly replacement of graphite
target prohibits them from being used as a continuous process. But they produce high quality
CNTs especially for the production of SWNTs. These high quality samples of CNTs are critical for
the nanotube research to achieve important results. Physical processes are ideal for production at
laboratory scale for research purpose, but following disadvantages limit their use as large-scale
industrial process for commercialized applications:
1. Large amount of energy is needed for vaporization of carbon atoms from target material,
which makes them energy extensive methods. It is impossible and uneconomical for this
huge amount energy to be generated for industrial use.
2. Large graphite is needed to be targeted for vaporization of carbon atoms.
3. Highly tanged CNTs are produced and mixed with unwanted form of carbons. Thus
purification is needed to purify the CNTs and assemble them into desired form. The
designing of such refining process is expensive and difficult.
On the other hand, chemical processes require only cost effective and convenient equipment for
controllable growth of CNTs. The chemical reaction takes place at relative low temperature and
ambient pressure. Besides that CVD, PECVD and WACVD can be operated continuously without
the need of replacing carbon feedstock, which makes them promising methods for continuous
industrial scale production. Chemical processes offer following advantages for the use as large-
scale production:
1. Simple reaction process and reactor design, controllable and manipulatable reaction.
2. Easy availability of raw materials as carbon feedstock.
3. Cheap production as little amount of energy is needed and cheap raw materials are
abundant.
4. Unique process for the synthesis of vertically aligned CNTs.
5. Similar operation to chemical unit operations makes them to be easily scaled up to large
industrial process.
Following table (Table 4.3-1) summarizes the main differences between the three methods.
Property/Process Arc discharge Laser ablation CVD
Raw materials availability Difficult Difficult Easy, abundantly available
Energy requirement High High Moderate
Process control Difficult Difficult Easy, can be automated
Reactor design Difficult Difficult Easy
Production rate Low Low High
Purity of product High High High
Yield of process Moderate High High
Post Treatment Refining Refining No extensive refining
Process nature Batch Batch Continuous
Per unit cost High High Low
Table 4.3-1: Comparison of CNT production methods (Rafique, et al., 2011)

15
5 Process for mass production
5.1 High-pressure carbon monoxide (HiPco) process
HiPco process is a type of CVD method for large-scale production of SWNTs and was introduced in
1999 (Nikolaev, et al., 1999). There are at least two large-scale reactors, which are currently
operated for industrial purpose: one at Rice University and another at a spin-off company, Carbon
Nanotechnologies Inc. (Harris, 2009), which has the production capacity of 65 g/h (Eklund, et al.,
2007). As was mentioned in the previous section, in conventional CVD method catalysts are
deposited or embedded on the substrate before the decomposition of hydrocarbon and growth
of CNTs on substrate begin. Instead, in HiPco process volatile organometallics are introduced into
the feed flow along with carbon feedstock. The organometallics will react, decompose and
condense in situ to form sized clusters, upon which CNTs nucleate and growth. With this method
the CNTs produced are free from catalytic supports and the product yield and purification yield
are as high as 97 % and 90 % respectively (Isaacs, et al., 2010).
Figure 5.1-1 shows the fundamental reactor and Figure 5.1-2 illustrates the block flow process
diagram for the HiPco process. In the initial lab work for parametric study (Bronikowski, et al.,
2001), 8.4 L/min of pure CO gas were rapidly mixed with 1.4 L/min gas mixture of CO and iron
pentacarbonyl (Fe(CO)5), which contained about 33.33 Pa (0.25 Torr) of Fe(CO)5 vapor. The
standard running conditions were 30 atm of CO pressure, 1323 K (1050 °C) of reaction
temperature and 24-72 hours of reaction time. The production rate of SWNTs under the reaction
conditions is 450 mg/h or 10.8 g/day. The Fe(CO)5 thermally decomposed and reacted to produce
Fe particles for the production of Fe clusters, which act as nuclei for the CNT growth. The solid
CNTs are produced catalytically through exothermic CO disproportionation on the surface of Fe
particles according to the Boudouard reaction:
𝐶𝑂(𝑔) + 𝐶𝑂(𝑔) ⇌ 𝐶(𝑠) + 𝐶𝑂2(𝑔)
Figure 5.1-1: Schematic of CO flow-tube reactor for HiPco process (Nikolaev, et al., 1999).

16
Figure 5.1- 2: The block flow process diagram for the production of CNTs with HiPco process.
The forward reaction of Boudouard reaction is an exothermic reaction and has more gas molecule
on the left side of the reaction. According to the Le Chatelier’s principle, increasing the
temperature and pressure will theoretically favor the forward reaction and produce high yield of
CNTs. This assumption is proven with the result from a parametric study of the HiPco process
(Bronikowski, et al., 2001) as shown in Figure 5.1-3 and Figure 5.1-4. Carbon dioxide (CO2) is the
by product for the Boudouard reaction, therefore the production of CNTs can be monitored by
measuring the corresponded maximum amount of CO2 produced assuming that all carbon
products are nanotubes.
Temperature plays a crucial role in the HiPco process. The effects from temperature on the
process have to be addressed during the considerations for the design of commercial scale
process and reactor. The gas phase catalyst, Fe(CO)5, will decompose rapidly at 250 °C and the
Boudouard reaction takes place at a significant rate only at temperature above 500 °C (Nikolaev,
et al., 1999). Thus, the heating rate of the gas mixture between the temperature ranges of 250-
500 °C will determine the result of the process. If the heating rate is too slow, larger Fe clusters
will form, which make them too big for nucleation of nanotubes and will overcoat with
amorphous carbon (Hafner, et al., 1998). On the other hand, high heating rate causes smaller Fe
clusters to form, which evaporate quickly at the temperature where formation of SWNTs occurs.
This leads to low yield of SWNTs. From Figure 5.1-3, it can be seen that the production of CO2 is
very low when the temperature is lower than 800 °C. But the production increases after 800 °C
and shows its maximum production at 1050 °C before fall off at higher temperature. The reason
behind the fall off was believed to be the higher rate of evaporation of active catalytic Fe clusters
at high temperature compare to the growth rate of SWNTs and decomposition rate of the catalyst
(Bronikowski, et al., 2001).
As the reaction pressure increased, the maximum production of CO2 increased simultaneously as
shown in Figure 5.1-4, because left side of equilibrium Boudouard reaction contains less gas
molecules. Higher pressure leads to higher disproportionation rate of CO, subsequently higher
Reactor
CO2 (g)
Condesation of
CNTs
P
CO (g)
CO (g) + Fe(CO)5 (g)
CO (g)CO (g)
CNTs (s)
Adsorption of CO2
CO (g) + CO2 (g)
CNTs (s) + CO (g) + CO2 (g)
Heating of CO gas

17
growth rate of SWNTs on the catalytic clusters. This allows the production of longer SWNTs, as the
carbon atoms have longer period of time to nucleate and growth on the clusters before they are
deactivated. Beside that higher growth rate of SWNTs leads to narrow diameter distribution of
SWNTs, because more small Fe clusters will be used for the nucleation and growth of SWNTs
before they grow into larger clusters through accretion.
The reactor of the process can be further modified by recycling the unconverted CO and mixing
with the CO feed. After the reaction, mixture of unconverted CO, SWNTs and CO2 will pass
through a series filters and cooled surfaces to collect the SWNTs by condensation and adsorption
beds containing sodium hydroxide (NaOH) to remove CO2 according to following chemical
equation: 2𝑁𝑎𝑂𝐻(𝑎𝑞) + 𝐶𝑂2(𝑔) → 𝑁𝑎2 𝐶𝑂3(𝑎𝑞) + 𝐻2 𝑂(𝑙). The unconverted CO will be
recirculated back to the reactor, thus forming a closed loop. This recycle process is assumed to be
capable of reducing the amount of CO needed from 162.5 g/h to 0.045 g/h (Isaacs, et al., 2010).
Figure 5.1-3: The CO2 yield against the reactor temperature, while the reactor pressure was
maintained at 30 atm (Bronikowski, et al., 2001).
Figure 5.1-4: The maximum CO2 yield and the concentration of Fe(CO)5 that produces maximum
CO2 against the CO pressure, while temperature was maintained at 1050 °C (Bronikowski, et al.,
2001).

18
5.2 Cobalt-molybdenum catalytic (CoMoCAT) process
CoMoCAT is another notable CVD method for large-scale production of SWNTs, which was
introduced in 2000 (Kitiyanan, et al., 2000). Currently, South West Nanotechnologie Inc. is using
CoMoCAT fluidized bed reactor for commercial production of CNTs (Agboola, et al., 2007).
CoMoCAT is a specified designed catalyst with synergistic effect of Co and Mo. Even though study
showed that Mo is capable to catalyze the CO disproportionation reaction for the production of
SWNTs, the reaction was carried out at very high temperature, to be specific at 1200 °C (Dai, et
al., 1996). Another report regarding the CoMoCAT showed that Mo is inactive for the production
SWNTs in the temperature range of 600-800 °C. And Co only has 7 % of selectivity toward SWNT.
But with the bimetallic CoMoCAT, the selectivity toward SWNTs was increased to more than 80 %
(Alvarez, et al., 2001). Beside that SWNTs produced through CoMoCAT process showed
significantly narrower distribution of diameters compare to SWNTs obtained from HiPco process
(Resasco, et al., 2002). And the SWNTs obtained from CoMoCAT can have purity higher than 90 %
(Isaacs, et al., 2010)
Figure 5.2-1 shows the basic illustration of a fluidized bed reactor and Figure 5.2-2 illustrates the
block flow process diagram of the CoMoCAT process used by South West Nanotechnologies Inc.
for the production of SWNTs. For the preparation of the bimetallic catalyst, aqueous solution of
cobalt nitrate (Co(NO3)2) and ammonium heptamolybdate ((NH4)6Mo7O24) were impregnated on
SiO2. After that the Co-Mo/SiO2 catalyst is dried in an oven at 80 °C and calcined in flowing air at
500 °C. The calcined catalyst will be placed inside the reactor, which was heated by H2 to 500 °C
and He to 700 °C. After that CO will be introduced into the reactor and undergo
disproportionation reaction (Boudouard reaction) to form SWNTs. The typical reaction
temperature and pressure ranges from 700 to 900 °C and from 1 to 10 atm respectively. The
production rate is about 0.25 g SWNT/g catalyst in a couple of hours (Rafique, et al., 2011).
Figure 5.2-1: A schematic of a fluidized bed reactor for the production of SWNTs using CoMoCAT
process (Jansen, et al., 2009).

19
Figure 5.2-2 : The block flow process diagram for the production of CNTs with CoMoCAT process
used by South West Nanotechnologies Inc.
The uniqueness of this process is the special designed bimetallic catalyst, which shows the
synergistic effect between Co and Mo. The molar ratio of Co:Mo has influence on the total carbon
yield and selectivity to SWNTs, as shown in Table 5.2-1. The production of SWNTs by CO
disproportionation is strongly affected by the of Co2+
species, which are stabilized by Mo oxide
species (Alvarez, et al., 2001). High molar ratio of Co:Mo promotes the production of carbon but
depresses the selectivity to SWNTs. On the other hand, the reaction is more selective to SWNTs
with deceasing molar ratio of Co:Mo but has lower total yield of carbon. Co is selective toward
SWNTs by interacting with Mo in a superficial Co molybdate like structure and maintain as well-
dispersed Co2+
ions at low Co:Mo ratio. But at high ratio of Co:Mo, Co forms a non-interacting
phase, which will be reduced to metallic Co. The metallic Co forms large clusters through sintering
at high temperature, at which CNTs are normally synthesized. The large Co clusters favor the
production of less desirable forms of carbon such as fibers and graphite (Resasco, et al., 2002).
Similar to HiPco process, the CO feedstock undergoes equilibrium disproportionation reaction,
which forward reaction will be theoretically favored, when the temperature is increased. As a
result, high yield of carbon products should be obtained. Surprisingly, results from a parametric
study (Alvarez, et al., 2001) showed that the total carbon yield decreased as the temperature was
increased from 600 –800 °C, as shown in Table 5.2-1. But the selectivity to SWNT was dramatically
increased with increasing reaction temperature as higher temperature suppresses the production
of SWNTs and supports the formation of MWNTs and amorphous carbon on the catalyst. Identical
results and trends were obtained when the temperature was further increased to 850 °C and
950 °C (Resasco, et al., 2004). Higher rate of catalyst deactivation compare to the decomposition
rate of CO is the reason behind the low carbon yield at high temperature. Beside that high
reaction temperature supports the formation of large diameter of SWNTs. Because sintering of Co
clusters accelerates with temperature to form larger clusters, upon which the SWNTs nucleate
and grow (Resasco, et al., 2004).
(NH4)6Mo7O24 (aq)
Impregnation on
SiO2 (s)
Drying in an oven
at 80 °C
Reactor
Calcination in
flowing air at
500 °C
CO2 (g)
Separation of CO2
throught
Adsorption
P
CO (g) + CO2 (g)
CO (g)
Co-Mo/SiO2 (s)
CO (g)
Co(NO3)2 (aq)
CO (g)
CNTs (s)

20
Catalyst Operating
temperature
Reaction
conditions
Total carbon
yield (%)
Selectivity to
SWNT (%)
Co:Mo (1:2) 700 °C 1 h, 50 % CO 1.5 88
Co:Mo (1:4) 700 °C 1 h, 50 % CO 1.6 96
Co:Mo (2:1) 700 °C 1 h, 50 % CO 2.2 57
Co:Mo (1:1) 600 °C 1 h, 50 % CO 2.7 25.8
Co:Mo (1:1) 700 °C 1 h, 50 % CO 1.7 62.5
Co:Mo (1:1) 800 °C 1 h, 50 % CO 1.0 86.6
Table 5.2-1: Total carbon yield and selectivity to SWNT obtained by CO disproportionation on
CoMoCAT with different Co:Mo ratio and reaction temperatur (Alvarez, et al., 2001)
Apart from temperature, CO concentration in the gas phase and reaction time are other
considerations in the design of commercial scale process for the high yield and selective
production of SWNTs. Study found out that the dominant product at low CO concentration was
amorphous carbon. But the yield of SWNTs grew with increasing concentration. As for the
reaction time, when the reaction was conducted in a short period of time, the amount of SWNTs
produced was very small and the product was mainly amorphous carbon. However, the growth of
SWNTs became dominant as the reaction time gets longer (Alvarez, et al., 2001).
5.3 Endo’s catalytic chemical vapor decomposition (CCVD)
The Endo’s CCVD was introduced in 1988 for the production of carbon fibers (Endo, 1988) and
was adopted for the production of MWNTs in 1993 (Endo, et al., 1993), which is a continuous
process known as floating reactant method. The process was scaled-up by Showa Denko KK Japan
for industrial production and the capacity was reported to be 16 kg/h (Eklund, et al., 2007). Figure
5.3-1 illustrates the setup for the floating reactant method and its corresponding block flow
process diagram is showed in Figure 5.3-2. In the process, hydrocarbon vapor, metal catalyst and
carrier gas such as Ar and H2 are fed into the reactor from the top of the reactor. The metal
catalyst particles are floating in the furnace zone of the reactor and gradually falling to the
bottom of the reactor due to gravitational force. The hydrocarbon vapor reacts and followed by
deposition of carbon atoms and growth of CNTs on the catalyst particles. Thus, CNTs can be
collected at the bottom of the reactor. Mixture of benzene vapor and H2 gas was used to produce
CNTs at the temperature of 1000 °C, as reported in the study (Endo, et al., 1993).
Figure 5.3-1: Schematic setup of floating reactant method for MWNT production (Endo, et al.,
2006)

21
Figure 5.3-2: The block flow process diagram for the production of CNTs with Endo’s process.
The size of the catalyst particle is curial for the controlling of the diameter and number of layers
of the CNTs. Small size catalyst favors the formation of SWNTs and small diameter of CNTs,
whereas large size catalyst promotes the production of MWNTs and large diameter of CNTs.
There are two way of controlling the catalyst particle size. The widely used method is deposition
of catalyst on the quartz substrate and subsequently limits the aggregation of the catalyst
particles. Another way is mixing with ceramic particles such as aluminum oxide (Al2O3), MgO and
zeolites to form ceramic-supported catalysts. The major advantage of using ceramic particles is
large surface area provided for the supporting of catalyst. The metal concentration and
temperature for the preparation of catalyst can be varied to obtain desired catalyst size (Endo, et
al., 2006).
Like other CVD methods, the carbon source and its flow rate is important for controlling number
of layers of CNTs. Carbon feedstock with low carbon content such as CH4 and C2H6 favors the
production of SWNTs due to their high thermal stability. With high carbon flow rate, it is difficult
for the efficient production of SWNTs and the products obtained are mainly MWNTs and
amorphous carbon (Endo, et al., 2006).
Reactor
Carrier gas
Ar and H2
CNTs (s) + Ar (g) + H2 (g)
Metal catalystHydrocarbon vapor

22
6 Conclusion
Being a cutting edge material, CNTs successfully showed the world their unique properties, which
are essential for the improvement of society daily life. Despite their production as bulk material,
with the ongoing research focusing on the mass production, the problem can possibly be solved in
the near future. Until then CNTs will be capable to show their truth potential to the world, which
were only seen by the researcher in the laboratory.
As was mentioned before, the focus of the up-to-date research is on CNT applications and mass
production instead of their synthesis methods in laboratory. This scenario indicates that the three
widely used synthesis methods are generally accepted as the most efficient and useful methods
to produce CNTs for various purposes. Arc discharge and laser ablation method provide high
quality CNTs to sustain the research of CNTs. Without them the exploration and development of
knowledge about CNTs might not be so highly promoted. Owning to the advantages such as low
production cost, controllable synthesis and so on, scalable CVD-based synthesis methods have
been developed and adapted into industrial for large scale production of CNTs.
Scaling up the laboratory apparatus to industrial scale plant used to be the major problem for CNT
mass production since their discovery, but the problem had been addressed. And currently, CNTs
have been successfully mass produced on ton scale by several companies such as, Hyperion
Company, Carbon Nanotechnologies Inc., South West Nanotechnologie Inc., Showa Denko KK
Japan, Arkema and others. As the demand of CNTs is growing fast in global market, it can be
expected that the companies will come out with improved technique and process for the mass
production of high quality CNTs in order to maintain their position in CNT industry. Process
intensification for example catalyst route innovation, feedstock saving and coupled process
(Zhang, et al., 2011) might be a new direction for the industry to further enhance their process for
the production of high quality CNTs with low cost.
There is no doubt that CNTs offer the society a lot of possibility to improve and catch up the fast
growing population and life. But the other side of the coin has to been taken into the account in
order to maintain the sustainable development of the CNTs. There are many studies focus on the
mass production and applications of CNTs, but their toxicity and negative effects on human and
environment cannot be ignored and should be addressed in order to prevent long-term harm to
human being and the nature.
As a newly discovered material, CNTs still have a long way to go and much work have to be done
in order to catch up the footstep of those traditional bulk chemicals. Following problems are some
of the current stumbling block for the mass production of CNTs, but they can be the stepping
stone for the sustainable mass production of CNTs, if the solutions to the problems are found:
1. Lack of an understanding of the CNT growth mechanism.
2. Difficulty to couple between CNT structure, mass production process and properties, as
well as between the main synthesis process and the post treatment such as dispersion,
forming of composites and others (Zhang, et al., 2011).
3. Lack of integrated overview on all the steps throughout the CNT mass production process.
4. Lack of understanding of the CNT negative effects on human being and environment.

23
Literature and Reference
Agboola, A.E., et al. 2007. Conceptual design of carbon nanotube processes. Clean Technologies
and Environmental Policy, 9. 2007, 289-311.
Ajayan, P.M., et al. 1994. Aligned carbon nanotubearrays formed by cutting a polymer resin -
nanotube composite. Science, 265. 1994, 1212-1214.
Alvarez, W.E., et al. 2001. Synergism of Co and Mo in catalytic production of single-walled carbon
nanotubes by decomposition of CO. Carbon, 39. 2001, 547-558.
Antisari, M.V., et al. 2003. Synthesis of multiwall carbon nanotubes by electric arc discharge in
liquid environments. Carbon, 41. 2003, 2393-2401.
Apul, O.G., et al. 2015. Adsorption of synthetic organic contaminated by carbon nanotubes: A
critical review. Water Research, 68. 2015, 34-55.
Barghi, S.H., et al. 2014. Chemisorption, physisorption and hysteresis during hydrogen storage in
carbon nanotubes. International of Hydrogen Energy, 39. 2014, 1390-1397.
Bell, M.S., et al. 2006. Carbon nanotubes by plasma-enchanced chemical vapor deposition. Pure
Chemistry and Applied Chemistry, 78. 2006, 1117-1125.
Bethune, D.S., et al. 1993. Cobalt-catalysed growth of carbon nanotubes with single-atomic-layer
walls. Nature, 363 . 1993, 605-607.
Bower, C., et al. 2000. Plasma-induced alignment of carbon nanotubes. Applied Physics Letters,
77. 2000, 830-832.
Bronikowski, M.J., et al. 2001. Gas-phase production of carbon single-walled nanotubes from
carbon monoxide via the HiPco process: a parametric study. Journal of Vacuum Science &
Technology A, 19,. 2001, 1800-1805.
Brukh, R., et al. 2008. Stabilizing single-walled carbon nanotubes by removal of residual metal
ctalysts. Chemical Physics Letters, 459. 2008, 149-154.
Coleman, J.N., et al. 1998. Precolation dominated conductivity in a conjugated polyer carbon
nanotube compostie . Physical Review B, 58. 1998, R7492.
Dai, H., et al. 1996. Single-walled nanotubes produced by metal-catalyzed disproportionation of
carbon monoxide. Chemical Physics Letters, 260. 1996, 471-475.
De Volder , M.F. , et al. 2013. Carbon nanotubes: present and future commercial applications.
Science, 339. 2013, 535-539.
Deck, C.P., et al. 2006. Prediction of carbon nanotubes growth success by the analysis of carbon-
catalyst binary phase diagrams. Carbon, 44. 2006, 267-275.
Eatemadi, A., et al. 2014. Carbon nanotubes: properties, synthesis, purification and medical
applications. Nanoscale Reasearch Letters, 9. 2014, 1-13.
Ebbesen, T.W., et al. 1992. Large-scale synthesis of carbon nanotubes. Nature, 358. 1992, 220-
222.
Ebbesen, T.W., et al. 1993. Patterns in the bulk growth of carbon nanotubes . Chemical Physics
Letters, 209. 1993, 83-90.
Eklund, P., et al. 2007. International assessment of research and development of carbon nanotube
manufacturing and applications. Maryland : World Technology Evaluation, Inc., 2007.
Eklund, P.C., et al. 2002. Large scale production of single-walled carbon nanotubes using ultrafast
pulses from a free electron laser. Nano Letters, 2. 2002, 561-566.
Endo, M. 1988. Grow carbon fibers in the vapor phase. Chemtech, 18. 1988, 568-576.

24
Endo, M., et al. 2006. Development and application of carbon nanotubes. Japanese Journal of
Applied Physics, 45. 2006, 4883-1892.
Endo, M., et al. 1993. The production and structure of pyrolytic carbon nanotubes (PCNTs).
Journal of Physics and Chemistry of Solids, 54. 1993, 1841-1848.
Flahaut, E., et al. 2003. Gram scale CCVD synthesis of double walled carbon nanotubes. Chemical
Communication, 12. 2003, 1442-1443.
Frank, B., et al. 2010. Oxidative dehydrogenation of ethane over multiwalled carbon nanotubes.
ChemCatChem, 2. 2010, 644-648.
Froundakis, G.E. 2011. Hydrogen storage in nanotubes & nanostructures. Materials Today, 14.
2011, 324-328.
Gong, K., et al. 2009. Nitrogen doped carbon nanotube arrays with high electrocatalytic activity
for oxygen reduction. Science, 323 . 2009, 760-764.
Guo, T., et al. 1995. Catalytic growth of single-walled nanotubes by laser vaporization. Chemical
Physics Letters, 243. 1995, 49-54.
Hafner, J.H., et al. 1998. Catalytic growth of single-walled carbon nanotubes from metal particles.
Chemical Physics Letters, 296. 1998, 195-202.
Harris, P.J.F. 2009. Carbon nanotube science: synthesis, properties and applications. Cambridge :
Cambridge University Press, 2009.
Hata, K., et al. 2004. Water-assisted highly efficient synthesis of impurity-free- single-walled
carbon nanotubes. Science, 306. 2004, 1362-1364.
Hirsch, A. 2002. Functionalization of single walled carbon nanotubes. Angewandte Chemie
International Edition, 41. 2002, 1853-2859.
Hong, S., et al. 2007. Nanotube electronics: a flexible approach to mobility. Nature
Nanotechnology, 2. 2007, 207-208.
Huang , S., et al. 2009. Metal-catalyst-free growth of single-walled carbon nanotubes on
substrates . Journal of the American Chemical Society, 131. 2009, 2094-2095.
Hussain, S., et al. 2015. Growth and plasma functionalization of carbon nanotubes. Journal of
Cluster Science, 26. 2015, 315-336.
Iijima, S. 1991. Helical microtubules of graphitic carbon. Nature, 354. 1991, 56-58.
Iijima, S., et al. 1993. Single shell carbon nanotubes of 1 nm diamter. Nature. 1993, 603-605.
Isaacs, J.A., et al. 2010. Economic assessment of single-walled carbon nanotube processes.
Journal of Nanoparticle Research, 12. 2010, 551-562.
Jansen, R., et al. 2009. Manufacturing characterization and use of single-walled carbon
nanotubes. Material Matters, 4. 2009, 23-27.
Jones, A.D.K., et al. 1997. Storage of hydrogen in single walled carbon nanotubes. Nature, 386.
1997, 377.
Jose-Yacaman, M., et al. 1993. Catalytic growth of carbon microtubes with fullerence structure.
Applied Physics Letters, 62. 1993, 202-204.
Kingston, C.T., et al. 2003. Fabrication of carbon nanotubes. Analytical Letters, 36. 2003, 3119-
3145.
Kitiyanan, B., et al. 2000. Controlled production of single-walled carbon nanotubes by catalytic
decomposition of CO on bimetallic Co-Mo catalysts. Chemical Physics Letters, 317. 2000,
497-503.
Kokai, F., et al. 1999. Growth dynamics of single-walled carbon nanotubes synthesized by CO2
laser vaporization. The Journal of Physical Chemistry B, 103. 1999, 4346-4351.
Kroto, H.W., et al. 1985. C 60: Buckminsterfullerene. Nature, 318. 1985, S. 162-163.

25
Kumar, M. 2011. Carbon nannotube synthesis and growth mechanism. [Buchverf.] S. Yellampalli.
Carbon nanotube - Synthesis, characterozation, application. Rijeka, Croatia : InTech, 2011.
Lam, C.W., et al. 2006. A review of carbon nanotube toxicity and assessment of pontential
occupational and environmental healt risks. Critical Review in toxicology, 36. 2006, 189-217.
Leroux, F., et al. 1999. Electrochemical insertion of lithium in catalytic multi walled carbon
nanotubes. Journal of Power Sources, 81. 1999, 317-322.
Martins, P.A., et al. 2013. Carbon nanotube directions and prespectives in oral regenerative
medicine. Journal of Dental Research. 2013.
Matsumoto, T., et al. 2004. Reduction of Pt usage in fuel cell electrocatalysts with carbon
nanotube electrodes. Chemical Communications. 2004, 840-841.
Mubarak, N.M., et al. 2014. An overview on methods for the production of carbon nanotubes.
Journal of Industrial and Engineering Chemistry, 20. 2014, 1186-1197.
Nhut, J.M., et al. 2004. Carbon nanotubes as nanosized reactor for the selective oxidation of H2S
intp elementar sulfur. Catalysis Today, 91. 2004, 91-97.
Nikolaev, P., et al. 1999. Gas-phase catalytic growth of single-walled carbon nanotubes from
carbon monoxide. Chemical Physics Letters, 313. 1999, 91-97.
Novoselow, K.S., et al. 2004. Electric field effect in automically thin carbon films. Science, 306.
2004, S. 666-669.
Oosthuizen, R.S., et al. 2011. Carbon nanotubes as supports for palladium and bimetallic catalyst
for use in hydrogenation reactions . Platinum Metal Reviews, 55. 2011, 154-169.
Peng, T., et al. 2015. Uniformly dispersion of carbon nanotube in aluminum powders by wet
shake-mixing approach. Power technology, 284. 2015, 32-39.
Pop, E., et al. 2006. Thermal conductance of an individual single-wall carbon nanotube above
room temperature. Nano Letters, 6. 2006, 96-100.
Porter, A.E., et al. 2007. Direct imaging of single walled carbon nanotubes in cells. Nature
nanotechnology, 2. 2007, 713-717.
Rafique, M.M.A., et al. 2011. Production of carbon nanotubes by different routes-a review.
Journal of Encapsulation and Adsorption Sciences, 1. 2011, 29-34.
Rao, G.P., et al. 2007. Sorption of divalent metal ions from aqueous solution by carbon
nanotubes: A review . Seperation and Purification Technology, 58. 2007, 224-231.
Ren, Z.F., et al. 1998. Synthesis of large arrays of well-aligned carbon nanotubes on glass. Science,
282. 1998, 1105-1107.
Resasco, D.E., et al. 2002. A scalable process for production of single-walled carbon nanotubes
(SWNTs) by catalytic disproportionation of CO on a solid catalyst. Journal of Nanoparticle
Research, 4. 2002, 131-136.
Resasco, D.E., et al. 2004. Decomposition of carbon-containing compounds on solid catalyst for
single-walled nanotube production. Journal of nanoscience and technology. 4, 2004, 398-
407.
Saito, Y., et al. 1996. Carbon nanocapsules and single layered nanotubes produced with platinum
group metal (Ru, Pd, Os, Ir, Pt) by arc discharge. Journal of Applied Physics, 80. 1996, 3062-
3067.
Shin, W.H., et al. 2012. Nitrogen doped multiwall carbon nanotubes for lithium storage with
extremely high capacity. Nano Letters, 12. 2012, 2283-2288.
Su, D.S., et al. 2010. Metal free heterogeneous catalysis for sustainable chemistry. ChemSusChem,
3. 2010, 169-180.

26
Tessonnier, J.P., et al. 2011. Recent progress on the growth mechanism of carbon nanotubes: a
review. ChemSusChem, 4. 2011, S. 824-847.
Thayer, A.M. 2007. Carbon nanotubes by the metric ton. Chemical & Engineering News, 85. 2007,
29-35.
Thess, A., et al. 1996. Crystalline ropes of metallic carbon nanotubes. Science-AAAS-Weekly Paper
Edition, 273. 1996, 483-487.
Villa, A., et al. 2009. Amino functionalized carbon nanotubes as solid basic catalysts for the
transesterification of tryiglycerides. Chemical Communications. 2009, 4405-4407.
Villa, A., et al. 2010. Pd on carbon nanotubes for liquid phase alcohol oxidation. Catalysis Today,
150. 2010, 8-15.
Xie, K., Muhler, et al. 2013. Influence of water on the initial growth rate of carbon ananotubes
from ethylene over a caobalt-based catalyst. Industrial & Engineering Chemistry Research,
52. 2013, 14081-14088.
Yakobson, B.I., et al. 1997. Fullerence nanotubes: C 1000000 and beyond. American Scientist.
1997, 324-337.
Yuca, N., et al. 2011. Hydrogen storage in single walled carbon nanotubes purified by miscrowave
digestion method. World Academy of Science, Engineering and Technology, 79 . 2011, 605-
610.
Zhang, J., et al. 2008. Surface-modified carbon nanotubes catalyze oxidative dehydrogenation of
n-butane. Science, 322. 2008, 73-77.
Zhang, Q., et al. 2011. Carbon nanotube mass production: principles and processes.
ChemSusChem, 4. 2011, 864-889.
Zhang, S., et al. 2011. The effects of dissolved matural organic matter on the adsorption of
sythetic organic chemicals by activated carbons and carbon nanotubes. Water Research, 45.
2011, 1378-1386.

27
Erklarung
SheongWei NG
Matr.-Nr.: 11107729
Münchener Straße 21
51103, Köln
Ich, SheongWei NG, erkläre hiermit, dass ich die vorliegende Masterprojektarbeit ohne fremde
Hilfe und ausschließlich unter Angabe der verwendeten Literatur und Software angefertigt habe.
_________________________________________________
Köln, 30.09.2015 – SheongWei NG

KURZBESCHREIBUNG
As the cutting edge material, CNTs offer spectacular
properties for the improvement of growing society.
Even though they found their place in many practical
applications, ongoing research is trying to show their
applicable potential in other technological areas. The
high market demand makes them the focus of the
research for mass production. And CVD methods
seem to be the most promising process for the
production at industrial scale. Many CVD-based
processes have been successfully adapted into
industrial for annually ton-scale production.
STICHWORTE: Carbon nanotubes, Application, Synthesis
Method, Mass Production, Review

MPA_Review of Carbon Nanotube Applications, Synthesis Methods and Processes for Mass Production

More Related Content

What's hot (20)

Viewers also liked (19)

Similar to MPA_Review of Carbon Nanotube Applications, Synthesis Methods and Processes for Mass Production (20)

MPA_Review of Carbon Nanotube Applications, Synthesis Methods and Processes for Mass Production