AL-FARABI KAZAKH NATIONAL UNIVERSITY

Z. A. Mansurov

SOOT FORMATION

Almaty
«Kazakh University»
2015
1
 UDK 541.183; 543.54; 541.124; 541.126
LBC 24.23
M 24

The Academic Council of Department Chemistry and Chemical Technology

and RISO of al-Farabi Kazakh National University
was recommended for publication

Reviewers:
Doctor of Physics and Chemisty Sciences P.A. Vlasov
N.N. Semenov Institute of Chemical Physics of Russian Academy of Sciences (Moscow)
Ph.D. in Chemistry M. Nazhipkyzy
al-Farabi Kazakh National University (Almaty)

Author expresses thanks to Ph.D-student Moldir Auyelkhankyzy for her assistance in

preparation of this monograph for publication.

Mansurov Z.A.
М 24 Soot formation: monograph / Z.A. Mansurov. – Almaty: Kazakh University, 2015.
– P. 167.
ISBN 978-601-04-0730-5
The monograph is devoted to the problems of soot formation during combustion of hydrocarbons:
the mechanism of soot formation, the influence of pressure and temperature on the processes in flame are
considered. Own electric field of the flame as well as the effect of external electric field on soot
formation, on the yield and structural characteristics of combustion products were also considered.
Modeling aspects of soot formation processes are stated and special attention was paid to the phenomenon
of cool-flame soot formation.
The results on synthesis in flame of unique nanostructures, such as carbon nanotubes, fullerenes and
graphenes at atmospheric and low pressure, with and without exposure to electric field are presented here.
Various methods for preparation of fullerenes, carbon nanotubes and graphenes their properties,
morphology, application fields and identification methods are considered. The existing models describing
synthesis of fullerenes, nanotubes and graphenes in flames are described.
This monograph can be useful to a wide range of professionals involved in nanotechnology as well
as bachelors, masters and Ph.D students and doctors.

UDK 541.183; 543.54; 541.124; 541.126

LBC 24.23

© Mansurov Z.A., 2015

ISBN 978-601-04-0730-5 © Al-Farabi KazNU, 2015

2
 C ONTENTS

PREFACE 6
REFEREE COMMENTS 8
INTRODUCTION 10

Chapter 1. Phenomenology of soot formation 12

1.1 Threshold of soot formation 12
1.2 Influence of pressure on the process of soot formation 14
1.3 Soot formation in diesel engines 15
Control questions 19
References 19

Chapter 2. Cool-sooting combustion of hydrocarbons and the

structure of the flame front 21
2.1 Procedure of experiment 23
2.2 Temperature profiles and physico-chemical characteristics 24
2.3 Formation of PAHs in low-temperature soot flames 26
2.4 Soot formation in electric field 28
2.5 Paramagnetism of soot particles 29
2.6 Mesophase pitches 32
Control questions 36
References 36

Chapter 3. Mechanism of soot formation 37

3.1 Polyyne model of soot formation 38
3.2 Polyaromatic model of soot formation 40
3.2.1 Hydrogen-abstraction – carbon-addition mechanism 40
3.2.2 Formation of the first aromatic ring 42
3.3 Ion model of soot formation 44
3.4 Detailed simulation of formation of PAHs and soot 49
3.5 Mechanism of soot formation in rich flames 51
Control questions 52
References 53

Chapter 4. Control of the formation of PCAHs and soot in a flame 56

4.1 Influence of chemical additives on the formation of polycyclic aromatic
hydrocarbons 56
4.2 Influence of an electric field 57
Control questions 60
References 60

Chapter 5. Synthesis of fullerenes in flame 62

5.1 The history of fullerenes 62
5.2 Methods of obtaining fullerenes 63
5.2.1 Arc vaporization of graphite 64
5.2.2 Laser ablation 65
5.3 Methods of investigation of fullerenes 66
5.3.1 Mass spectroscopy of fullerenes 66
5.3.2 Transmission electron microscopy of fullerenes 68
3
 5.4 Synthesis of fullerenes in flame 69
5.4.1 Synthesis of fullerenes under the conditions of combustion of a benzene- 69
oxygen-argon mixture at a low pressure in an electric field
5.5 Properties of fullerenes 75
5.5.1 Solubility of fullerenes 75
5.5.2 Physical properties 77
5.6 Application of fullerenes 78
Control questions 79
References 79

Chapter 6. Synthesis of carbon nanotubes in flame 82

6.1 The history of carbon nanotubes 82
6.2 Methods of obtaining carbon nanotubes 83
6.2.1 Arc discharge 83
6.2.2 Pulsed laser vaporization 84
6.2.3 Chemical vapor deposition (CVD) 85
6.3 Methods of investigation of carbon nanotubes 86
6.4 Synthesis of carbon nanotubes in flame 87
6.4.1 Controlling MWNT growth in flames 91
6.4.2 Synthesis of single-wall carbon nanotubes in flame 95
6.5 Properties of carbon nanotubes 103
6.6 Application of carbon nanotubes 104
Control questions 106
References 106

Chapter 7. Synthesis of hydrophobic soot in flame 110

7.1 The history of hydrophobic surface 110
7.2 Methods of obtaining hydrophobic surfaces 112
7.2.1 Roughening of hydrophilic materials 112
7.2.2 Fabrication of rough surfaces from hydrophobic materials 113
7.3 Superamphiphobic surfaces 115
7.4 Synthesis of hydrophobic soot in flame 116
7.5 Theoretical background of hydrophobic surface 120
7.5.1 Surface roughness 120
7.5.2 Modification of roughness of low surface energy materials 123
7.6 Application of superhydrohobic surfaces 124
Control questions 125
References 125

Chapter 8. Synthesis of graphene layers in flame 128

8.1 The history of graphene 128
8.2 Methods of obtaining graphene 129
8.2.1 Micromechanical exfoliation of graphite 129
8.2.2 Liquid phase exfoliation of graphite 129
8.2.3 Oxidation of graphite 130
8.2.4 Chemical vapor deposition synthesis of graphene 132
8.2.5 Obtaining of graphene in an electric arc 133
8.2.6 Thermal decomposition of silicon carbide 134
8.2.7 Epitaxial growth of graphene on the metallic surface 135
8.3 Methods of investigation of graphene 136
8.3.1 Raman spectroscopy of graphene 136
8.3.2 Visual observation of graphene 137
8.3.3 Electron microscopic study of graphene 139
8.4 Synthesis of graphene layers in flame 140
8.4.1 Synthesis of graphene layers in a hydrocarbon flame 140

4
8.4.2 Synthesis of graphene layers in the dual flame 141
8.4.3 Synthesis of graphene layers in the diffusion back flame 142
8.4.4 Synthesis of graphene layers in premixed flames 145
8.4.5 Synthesis of graphene at low pressure in the flame 156
8.5 Properties of graphene 158
8.5.1 Electronic properties 158
8.5.2 Mechanical strengh 159
8.5.3 Optical properties 160
8.5.4 Superhydrophobicity of graphene 160
8.6 Application of graphene 163
Control questions 163
References 164

5
 PREFACE

S
oot formation in combustion processes – is a multi-aspect, multi-disciplinary
phenomenon, the thematic justification is very important both as for ecology, the
producing of technical carbon and also for producing of different nanomaterials:
fullerenes, carbon nanotubes, and graphene.
Despite many researches, there is no full understanding of conversion mechanism of
initial fuel into end products – soot, fullerenes, carbon nanotube (CNT), and graphene.
Varying the experimental conditions (pressure, ratio of fuel /oxidant, application of an
electric field, introduction of various mixtures) it can be obtain the carbon materials with
various properties.
Researches on soot formation and related phenomena – is a subject of regularly
realized symposiums on combustion, carbon materials, where the different special aspects
in the field of soot formation are discussed, round-table conferences the results of which
are multi-authorized monographs (D. Siegla and G. Smith, J. Lahaye and G. Prado, H.
Jander, H. Wagner and H. Bockhorn).
This tutorial is written on the base of special course is readed to Ph.D.- students of
Department of Chemical Physics and Materials.
It necessary to note the major contribution to the investigation of soot formation
processes of such researches as H. Calcote, I. Glassman, P. Tesner, H. Wagner, К.
Homann, J. Warnatz, J. Howard, J. Lahaye, G. Prado, H. Jander, I. Kennedy, K. Saito, M.
Frenklach, H. Richter, В. Knorre, А. Krestinin, P. Vlasov, B. Wang and others.
In the first chapter of this monograph is considered the phenomenology of soot
formation including threshold of soot formation, influence of pressure on soot
formation process, soot formation in diesel engines.
In the second chapter of this monograph is considered cool-sooting combustion of
hydrocarbons and the structure of the flame front including temperature profiles and
physico-chemical characteristics, formation of PAHs at a low-temperature in soot
flames, soot formation in electric field.
In the third chapter is considered the mechanism of soot formation including
polyyne model of soot formation, polyaromatic model of soot formation, hydrogen-
abstraction – carbon-addition mechanism, formation of the first aromatic ring, ion
model of soot formation, detailed simulation of formation of PAHs and soot.
The fourth and fifth chapters are dedicated to the history, method of obtaining, its
properties and applications of fullerenes and carbon nanotubes. For the first time the
fullerenes were synthesized by H. Kroto and others in 1985. A fullerene is
a molecule of carbon in the form of a hollow sphere, ellipsoid, tube, and many other
shapes. During obtaining fullerenes the following methods such as arc evaporation of
graphite, laser ablation and synthesis in flame are used. Special attention was given to
synthesis of fullerenes in combustion of benzene-oxygen –argon mixture at low
6
pressure in an electric field. Carbon nanotubes are the nanoobject like nanowires
which act as “nano” only in two dimensions, but in third dimension they appeared as
“micro” object. For the first time the carbon nanotubes were obtained by S. Iijima in
1991. Among the various methods for investigation of carbon nanotubes, transmission
electron microscopy (TEM) is one of best observing methods are produced in an
electric arc between graphite electrons.
The sixth chapter shows the superhydrophobic materials, super amphiphobic
surfaces, theoretical background of hydrophobic surface and the synthesis of
hydrophobic soot in flame. Superhydrophobic materials have recently attracted
attention from both academic and industrial circles because of their importance in
fundamental research and potential industrial applications such as bio-surfaces, anti-
biofouling, transparent and antireflective superhydrophobic coatings, structural color,
fluidic drag reduction, enhancing water supporting force, controlled transportation of
fluids, superhydrophobic valves and coating films for electronic devices.
The seventh chapter is dedicated to graphenes as we know graphene is one of the
latest representative of carbon nanomaterials, is obtained by micromechanical method
by A. Geim and K. Novoselov. Graphene is a single layer two-dimensional carbon
structure, the surface of which is laid out in regular hexagons with sides of 0.142 nm
and carbon atoms at the vertices. In its properties, graphene is a semimetal with a
small overlap of the conduction and valence bands.

Professor of Department of Chemical Physics and Materials,

General Director of the Institute of Combustion Problems
Z.A. Mansurov

7
 REFEREE COMMENTS

The textbook Mansurov Z.A. "Soot formation", Almaty: Kazakh University, 2014,
210 P., represents a highly professional review of the up-to-date state of the theoretical
and experimental surveys on the formation of various carbon allotropes, which is
based on the monograph Z.A. Mansurov, N.G. Prikhod'ko, A.V. Savel'ev.
Obrazovanie PAH, fullerenov, uglerodnih nanotrubok I sazhi v processax goreniya
(Formation of PAHs, Fullerens, Carbon Nanotubes and Soot in Combustion
Processes), Almaty: Kazakh University, 2012, 383 P., and includes the data appeared
during the last two years.
This book can be useful for a wide range of the readers interested in the physical
chemistry of combustion and for scientific researchers and engineers involved in
nanotechnology of carbon nanostructures. There are some questions at the end of each
chapter that makes it possible to use this book as a tutorial for senior students,
bachelors, masters and Ph.D-students.
This book concerns with an urgent aspect of physical chemistry of combustion:
nucleation and phase transition phenomena, in particular, carbon particle formation of
different structure depending of particular experimental conditions. Theoretical and
experimental aspects of formation of soot pmticles, fullerenes, carbon nanotubes and
graphenes under various experimental conditions, including external electric fields, are
considered.
The book involves in details such important aspects as the influence of charged
palticles (electrons and positive and negative ions) and external electric field on the
phase transition processes in carbon-containing systems. Such exotic phenomena as
cool-flame soot formation and super hydrophobic soot surface formation in flames are
also discussed.
Areas of fullerene applications, the absence of their toxicity and carcinogenic
activity and prospects for application of various fullerene derivatives in biology and
medicine are discussed in details.
Possible areas of application of carbon nanotubes are also considered. Carbon
nanotubes possess the unique chemical and physical properties that make them
promising objects for practical application. Small sizes and their unique structure
cause their unusual mechanical and electronic properties, at that, their electronic
properties can be purposefully changed within a wide range ofparameters.
Special attention is paid for single-wall nanotubes, which can be a basis for
creation of a new generation of electronic chips. The unique electronic properties of
single-wall carbon nanotubes already in the nearest future can be used in design of
various electronic devices: microchips for the new generation of electronic, logic,
mechanical and electromechanical gadgets.

8
 The methods of carbon nanotube synthesis are described in great details. Their
synthesis in flames is more flexible as compared with the other traditional methods
because changing the combustion conditions, composition and the type of catalyst one
can obtain carbon nanotubes with the desired properties.
The reviewer recommends Z.A. Mansurov's textbook "Soot formation" to be
published.
Reviewer:

Vlasov Pavel Aleksandrovich, Doctor of Science (Physics and Mathematics), the

Head of the Laboratory of High-Temperature Kinetics and Gas Dynamics of the
Semenov Institute of Chemical Physics, Russian Academy of Sciences.

9
 INTRODUCTION

T he physicochemical properties of soot and the effects arising in a flame

in the process of soot formation have been objects of numerous
investigations for more than 100 years [1, 2], and such investigations did
not loose their significance to the present day. One of the basic books about the
electrical properties of flame is the book "The electrical properties of flames and of
incandescent solids", which was published by Wilson H.A. in 1912. The front page of
this book is shown below. The electrical properties of flames and of incandescent
solids. This is explained, first of all, by the fact that soot is an industrial product
produced on a world scale in the amount of 107 tons a year. The black
(technological) carbon is used as a filler of elastometers (90 % of the
technological carbon is used for this purpose, and 2 ⁄ 3 of it — in the production
of tires) and has a wide range of applications in printers. However, soot is a
carcinogenic pollutant of the environment, formed as a result of combustion of
hydrocarbon fuels in power plants and engines. For example, diesel engines with
direct fuel injection initially transform approximately 10-20 % of the introduced
fuel into the soot [2, 3]. Simultaneously with soot formation, fullerenes and
nanotubes are formed by the mechanism competing with the mechanism of soot
formation [4, 5]. K nowledge of the conditions and mechanisms of formation of
soot, fullerenes, and nanotubes in a flame allows one to change the combustion
such that soot particles, fullerenes, or nanotubes are predominately formed.
The nature of soot remains an object of intensive investigations aimed at
determining the physical and chemical properties of solid carbons and explaining a
number of fundamental phenomena taking place in condensed media with different
structural orderings. Investigation of the structure of sooting flames provides
valuable information on the chemical and physical processes occurring prior to the
soot formation and in its process.
At present, a large number of experimental data on the processes of soot
formation have been accumulated and different phenomenological models have
been proposed [6-11]. However, the mechanism of soot formation is not perfectly
understood yet. This is explained by the fact that even in simple cases, such as
homogeneous pyrolysis of hydrocarbons, this process includes a large number of
rapid simultaneous reactions leading to formation of a new solid phase - soot
particles (e.g., the time of transformation of methane with the molecular mass of
16 into soot with the molecular mass more than 106 a.m.u is equal to 10-4-10-2s).

10
The front page of the book «The electrical properties of flames and of incandescent solids» [2]

11
 Chapter 1

PHENOMENOLOGY OF SOOT FORMATION

1.1. Threshold of soot formation

Formation of soot is mainly determined by the type of fuel, its combustion

temperature, the pressure in the system, and the ratio between the combustible and the
oxidizer. In accordance with the laws of thermodynamics, solid carbon should appear
in a premixed fuel-air flame at a normal temperature in the case where the ratio
between the combustible and the oxidizer (the ratio of carbon atoms to oxygen) is
equal to C/O ≈ 1.0 [7], which is not always the case. Soot begins to form, with the
exception of the detonation process, at a ratio C/O smaller than unity; the value of this
ratio is determined by the type of fuel and the temperature [7, 8]. It has been stated
that the majority of flames in a Bunsen burner, with the exception of the C2H2–air
flames, emit soot at the vertex of their cone. The thresholds of the soot formation
(C/O)cr in Bunsen and flat-flame burners are different because of the differences
between them in the polycellular-flame formation and the temperature (Table 1). The
results obtained show that the formation of soot in a flame is determined by the flame
kinetics and takes place due to oxidation of the fuel.
Table 1

Soot-formation threshold (c/o)cr for premixed flames (P = 1 bar, T = 293 K) [7, 8]

Fuel-air mixture (C/O)cr

Bunsen burner Flat flame
С2Н6 – air 0.48 0.47
С3Н8 – air 0.47 0.53*
С3Н8 – О2 – 0.67*
С2Н4 – air 0.6 0.62
С2Н4 – О2 – 0.7
С4Н8 – air 0.48 –
С6Н6 – air 0.57 0.62*
Methylnaphthalene – air 0.42 –
С2Н2 – air – 0.83
*Ppolycellular flames

In this case, the afterflame gases released as a result of combustion of mixture with
a C/O ratio somewhat smaller than the soot-formation threshold and the mixture with
C/O > (C/O)cr have identical compositions. Consequently, the basic processes of soot
formation proceed in the main reaction zone of a flame, where oxidization of fuel and
formation of higher hydrocarbons compete with each other.
It has been stated that the threshold of soot formation depends on the temperature
and pressure of the flame [12-14] and that two soot-formation thresholds exist for the
C/O ratio at two different temperatures. As the pressure increases, the minimum of the
(C/O)cr ratio shifts towards the stoichiometric mixture (C/O = 0.33 for a C2H2–air
12
mixture). The (C/O)cr ratio in the lower-temperature region increases with the decrease
in temperature. At temperatures lower that 1350–1400 K, a substance with a large
molecular mass is present in the burned gas; however, it cannot transform into soot at
such a low temperature during the contact time t < 0.1 s. For this substance, the ratio
H/C is approximately equal to unity; however, its components were not identified. It
should be noted that this low "threshold temperature" is close to the temperature
detected at the central line of a
laminar diffusion flame at the
beginning of soot formation.
When the excess carbon exceeds
the soot-formation threshold, the
final soot volume increases sub-
stantially and the efficiency of
transformation of carbon into soot
increases for different fuels.
When passing from the soot-
formation threshold, where the
concentration of soot is about 10-9
g/cm3, to the larger values of C/O,
the final soot volume f V∞
increases substantially relative to
the volume of the gas in the flame.
Fig. 1 shows, at a qualitative
level, the efficiency of tran- Fig. 1. Final volume fraction of the soot depending
sformation of the excess carbon on the excess carbon for the fuel-air flames obtained
in flat burners at a normal pressure, f V∞,
exceeding the sootformation cm3/cm3 [12, 13].
threshold into soot for different
fuels [12, 13].
A detailed chemical reaction
model for the growth of po-
lycyclic aromatic hydrocarbons
and soot particle nucleation and
growth was proposed Frenklach
[15] and Wang [15, 16]. Illus-
trative information on transforma-
tion of hydrocarbon fuels into po-
lycyclic aromatic hydrocarbons
(PAH) and soot in the process of
their combustion [17] is presented
in Fig. 2. PAHs are usually for-
med as a result of combustion of
rich mixtures and can possess car-
cinogenic properties (as, e.g., ben-
zpyrene). At present it is com-
monly supposed that these hydro-
Fig. 2. Schematic representation of soot formation in
carbons are the main precursors homogeneous systems or
(nuclei) of soot particles with in flames of a premixed mixture [15]
molecular mass of 500–2000
a.m.u. Then these particles grow due to the increase in their surface in the reactions
with many acetylene molecules and as a result of the coagulation.
13
 1.2. Influence of pressure on the process of soot formation

The volume fraction of the soot fV formed in a system at the pressure varying
within the range 1–10 bar is approximately proportional to its square [12-14]. At the
pressure higher than 10 bar, fV becomes proportional to the first power of the pressure.
The increase in soot formation with increasing pressure can be explained by the
increase in the density of carbon atoms. This assumption was verified in the
experiments on the combustion of ethylene mixed with oxygen and other oxidizers (P
= 1 bar) [18]. The reaction zone in these experiments was spatially resolved. The
density of the carbon atoms in the soot particles was three times higher than that in the
corresponding air flames. The coagulation coefficients in these flames were constant.
An interesting result was obtained for the change in the yield of C2H2 with increasing
pressure. As is known [8, 19], C2H2 is the main component of the residual combustion
products of hydrocarbon fuels at atmospheric pressure. However, its fraction in these
products decreases rapidly with the increase in the pressure. At P = 70 atm, the
concentration of C2H2 is lower than its detection limit. After the burning of C2H4 in air
at the pressure of 70 atm at different C/O ratios, the residual combustion products
contain mainly water vapor, methane and soot.
The data of investigations of coagulation in flames with a high density of carbon
atoms at a normal pressure agree with the results obtained in [8] for C2H4–air flames at
a high pressure and support the assumption that soot formation increases with the
increase in the density of carbon atoms.
The volume fraction of soot
increases with the increase in the
pressure and in the C/O ratio, and
the dependence of the soot for-
mation on the temperature has a
bell-like shape [8, 12] (Fig. 3).
The latter is due to two cir-
cumstances. Radicals-precursors
are required for soot formation;
therefore, soot is not formed at a
low temperature. Moreover, the
soot precursors pyrolyze and oxi-
dize at a high temperature, so that
the soot-formation process is
restricted by the interval T =
1000-2000 K.
Fig. 3. Dependence of the soot amount (the fraction of C Investigation of the depen-
atoms in the soot particles)
on the pressure and temperature [12]. P, bar; T, K. dence of the carbon density on C/O
ratio for soot, PAHs, and C2H2 at P
= 1, 10, and 70 bar [20] has shown
that the increase in pressure increases the rate of formation of PAHs and soot. The
C/O ratio also influences the soot volume fraction: this fraction can increase with
increase in the C/O ratio at a high pressure too.
Once the threshold of the soot-particle formation is exceeded, the carbon density
begins to increase. At large values of C/O, the rate of increase in the carbon density
decreases. The dependence of the carbon density ρ(C/O) is stronger at P = 10 and 70
bar than at P = 1 bar.

14
 The density of carbon in a PAH is two times lower than that in soot; however, the
dependences of the carbon density of soot and PAH on the ratio C/O are very similar
for different pressures. On the basis of the data obtained, the following expression can
be written:

ρsoot, PAH≈ [C/O-C/Ocr ]2.5–3 at T 1700 K , 1 ≤P ≤10 bar .

In [21], the mechanisms of soot formation in diffusion methane-oxygen flames at a

pressure of 100 atm were experimentally investigated in accordance with the prospects
of using methane and liquid oxygen as liquid jet fuels in reusable rocket engines. This
is explained by the fact that the indicated fuel makes it possible to decrease the
production expenses and the dimensions of the reservoir as compared to those in the
case where liquid hydrogen is used as a fuel. The experiments were carried out in a
coflow burner comprising a fuel-injection nozzle of stainless steel with an outer
diameter of 3.06 mm and an air-injector diameter of 25 mm.
A three-dimensional image of the volume soot fraction as a function of pressure is
presented in Fig. 4. The maximum soot concentration was increased from 7 ppm at 10
atm to 55 ppm at 40 atm. However, a further increase in pressure caused the decrease
in the volume soot content to 4 ppm at 90 atm. The maximum soot volume fraction
increased as fV,max = P1.5 for pressures falling within the range 10–40 atm. The pressure
power is 2.3 for pressures from 50 to 70 atm and about 7.6 for pressures in the range
70–90 atm.
It has been stated that the dependence of
soot formation on pressure can be repre-
sented as a function of the percent content
of the fuel carbon converted into soot at the
flame height. The specific mass flow rate of
carbon in the form of soot can be
determined as a function of the flame
height:

Ms zvz zps/2nrfV r, zdr,

here, vz is the velocity of the gas flow along

the axis of the flame; ps = 1.8 g/cm3 is the
soot density; z is the height along the axis Fig. 4. Dependence of the soot volume
of the flame; vz(z) = 2az, where a is a fraction on the pressure and the spatial
constant acceleration that is usually taken position of a flame. fV, ppm; R, δ, mm.
to be equal to 25 m/s2. The percent content
of carbon (in relation to soot) in the converted fuel is approximately equal to ns =
ms/mC, where mC is the specific mass flow rate of carbon at the output of the nozzle.

1.3. Soot formation in diesel engines

A carbon (soot) microparticle is associated for the most part with the exhaust gases
of a diesel engine or pyrolitic equipment. The soot formed in an engine is an
undesirable product that decreases the serviceability of the engine, contaminates the
environment, and affects the health of people because soot particles contain molecules

15
of polyaromatic hydrocarbons. For example, American regulatory bodies restrict the
emission of soot particles by medium-power trucks to 0.05 g/(kW·h) [22].
In a diesel engine, soot particles are formed in the region between the fuel-rich
side of the reaction zone of a diffusion flame and the fuel jet. Soot-formation can be
considered as transformation of the gas phase into the solid one [7], in the process of
which conversion of hydrocarbon-fuel molecules containing a small amount of carbon
atoms into carbon particles containing several millions of carbon atoms are an
extremely complex process. At first, the smallest particles are formed due to
coagulation of PAHs. Then the soot concentration increases as a result of adsorption
of PAHs and, especially, due to the reactions on the particle surface cause increasing
it. This process is described by the mechanism of hydrogen splitting and addition of
carbon [23]. The primary particles coagulate into large aggregates, that cannot always
be detected in the exhaust gases of diesel engines. Detection of these aggregates
depends substantially on their temperature and the time of transformation of the
particles in the oxidation medium. Therefore, only several thousandth of the initial
soot mass are not oxidized completely and enter the exhaust gases of a diesel engine.
The soot emitted by diesel engines forms a large part of the atmospheric aerosol
even in the case where good conditions of their work are provided, and its amount
depends on the working parameters of an engine. The soot content in the aerosols in
some regions of Germany reaches 20%. The emission of soot by a four-cylinder diesel
engine of volume 2.4 liter [7] is illustrated in Fig. 5. It is seen that in the case where
the speed of an engine is 2500 rpm, soot emission is relatively small; however, it
increases with the increase in the engine speed.
The amount of soot emitted
by a diesel engine is determined
by the atomization and configu-
ation of the fuel jet, the method
of air supply, the degree of
turbulence of the fuel flow, the
pressure in it, the time of its
injection, and the delay time of
ignition [24].
Investigations of the soot
particles taken from the exhaust
parts of diesel engines have
shown that the size distribution
of these particles depends
insignificantly on the working
conditions of an engine and its
Fig. 5. Parameters of the soot emission for a four-cylinder type. The average diameter of
diesel engine of volume 2.4 liter [7]. such particles is about 200–300
Y, g/h; η, rpm; Pg, bar.
Å. The soot samples taken from
the exhaust systems of diesel
engines and from the low-temperature exhaust gases contain heavy hydrocarbons
condensed into agglomerates transformed into gummy substances; in this case, the
H/C ratio is as large as unity.
As pointed out above, the injection time is a fairly significant factor. An early
injection [25] makes it possible to inject a larger amount of a fuel and to mix it with
air prior to ignition. Moreover, a high injection rate serves to decrease soot formation.
The increase in the temperature of the supplied air leads to the decrease in the ignition
16
delay and, consequently, promotes soot formation [24]. As was noted in [9], the real
values of the C/O ratio should be close to critical because the volume fraction of the
soot and the sizes of the soot particles increase with the increase in the ratio C/O.
Fine and ultra-fine particles emitted by modern Diesel engines with sizes below
100 nm are of significant importance. Therefore, the emission behavior of modern
diesel engines is of particular interest in regard to particle concentration and
agglomerate sizes and number densities. Particle sizes and particle size distribution
together with the particle surface play a significant role in regard to physiological
effects on human health. Former investigations done by SMPS measurements
observed similar particle size distributions inside the combustion chamber in the late
expansion phase and in the tailpipe of a modern common rail diesel engine. Even
when performing these experiments with different fuels, the maximum in the
aggregate particle size distribution was found to be between 90 nm and 110 nm [26].
The information on soot concentration, soot particle sizes of primaries and
agglomerates and spray distribution are investigated by means of laser diagnostic
techniques, e.g. laser induced incandescence (LII), laser induced fluorescence (LIF)
and scattering methods [27].
In [28], the influence of
unstationary engine operating
conditions on soot formation and
emission, especially the in-
fluence of load changes and
acceleration, was studied by
RAYLIX method. A special
point of interest is the inves-
tigation of the formation of smal-
ler primary particles (<100 nm)
which bear the risk of being
respirable and consequently ha-
rmful to human health. Therefore
these investigations have been
carried out to get a better under-
standing of those processes.
Operating points with changing
loads at constant engine speed
are of a special interest for off-
road applications like work Fig. 6. Soot concentration and particle properties for the
machines (diggers, fork lift 900 rpm load changes [25]
trucks, etc.). Load changes were
performed at different engine
speeds, grading and times. It was started from a stationary point, after the load change
no steady operating point is reached during the detection time. Fig. 6 shows the results
of a measuring series with constant engine speed of 900 rpm and an IMEP step from
0.4 MPa to 0.8 MPa (0.3-0.7 MPa BMEP) (BMEP=Brake Mean Effective Pressure)
within 2 s and 5 s [27].
The upper part of Fig. 6 presents soot concentrations in the combustion chamber
(upper left diagram) and inside the optical flange (upper right). When starting the
camera system, the gate pulse enables the start of the 2CM. The measurement period
of the 2CM is limited to 140 combustion cycles corresponding to 18.6 s. Afterwards
the data have to be stored. So it is not possible to record the complete load change
17
using the 2CM. Therefore, the in-cylinder soot concentration is recorded in two
sections as shown. In the lower part the particle properties measured in the optical
flange like the mean particle radii (lower left) and the number densities (lower right)
for both the 2 s and the 5 s load change are shown. The results of the RAYLIX
measurements give detailed information on the behavior of particle properties as
shown in Fig. 5. It is not possible to detect every single combustion cycle with that
technique because of a relatively high digitalization and read-out time of the CCD
cameras (up to 500 ms depending on recording conditions). Additionally to the
measured load change, the descent back to the starting point has been detected. The
overshoot in soot concentration as described above can also be seen in the soot
concentration in the optical flange as well as in the in-cylinder soot concentration and
is followed by a constant value. The 2CM measurements detected a local maximum at
about combustion cycle 60. During the load change the decrease of mean particle radii
can be seen while number densities increase disproportionately. This effect lasts until
the end of the high level operating point. It is remarkable that the 2 s change results in
higher soot concentration and number densities whereas the primary particle sizes are
smaller.
Further investigations at
different engine speeds were
carried out to compare the
behavior of soot particle pro-
perties. They were performed
at 900 rpm and 1200 rpm
with IMEP steps from 0.2 to
0.8 MPa. In these investi-
gations the duration for all
load changes was 3 s. As
before, the soot concentra-
tions show a steep increase
with the beginning of the load
change and a constant value
afterwards (Fig. 7). A noti-
ceable overshoot can be seen
in the 2CM-1200-curve
which ends in a much lower
constant value than the 900
Fig. 7. Soot concentrations and particle properties at 900 rpm rpm curve. The second
and 1200 rpm [26]
measurement periods were
made in the high level
operating point. The progress of the RAYLIX curves is similar to those of the first
investigations. A slight overshoot in the soot concentration in the optical flange can be
seen in both cases with a following lower level constant value. This effect can also be
seen in the mean particle radii and the number densities. Particle sizes decrease with
rising load while number densities increase.
From independent REM recordings a size classification of soot particles deposited
on a filter paper used for estimation of the FSN was carried out. A lognormal particle
size distribution with mean primary particle radii of rm = 11.9 nm and a standard
deviation of σ = 0.38 fit the data obtained from these recordings very well.
Additionally, this value for the mean primary particle radius is in good agreement with
the values found by RAYLIX in context to Fig. 6 and 7. Generally, it can be said that
18
during load changes and at the high level operating points much more small particles
are emitted. When descending back to low level operating points soot concentrations
decrease as well as number densities whereas the mean particle sizes increase again.
So it can be stated that load changes starting from middle speed and middle load to
higher speed and higher load do not have such a significant influence on soot
formation and oxidation and thus on soot emissions than load changes beginning at
lower speeds and loads [28]. Explanations can be found generally in better combustion
conditions at higher engine speeds. Besides higher fuel injection pressures resulting in
a better spray distribution especially the turbo charger’s efficiency, increase with the
increase of exhaust gas mass leading to a higher charge air pressure, more swirl and as
a consequence lower soot emissions.

CONTROL QUESTIONS

1. How would you explain phenomenology of soot formation?

2. Can you give the description of threshold of soot formation for premixed flames?
3. What is the influence of pressure on soot formation process?
4. What do you know about the threshold of soot formation?
5. Can you give the explanation of soot formation in diesel engines?

REFERENCES

1. J.J. Thomson. On the masses of the ions in gases at low pressures, Phil. Mag. Ser. 5, 48, 547 (1899).
2. H.A. Wilson. The Electrical Properties of Flames and Incandescent Solids, University Press, London
(1912).
3. J. Warnatz, U. Maas, R. Dibble. Combustion. Physical and Chemical Fundamentals, Modelling and
Simulation, Experiments, Pollutant Formation // Springer-Verlag, Berlin, Heidelberg, 2006, P. 378
4. Z.A. Mansurov. Soot formation in combustion processes // Combust. Explosion Shock Waves, 2005,
41, P. 137–156
5. H.W. Kroto, J.R. Heath, S.C. O’Brien, et al. C60: Buckminsterfullerene, Nature, 318, 162–163
(1985).
6. P. Gerhardt, S. Loffler, K.H. Homann. The formation of polyhedral carbon ions in fuel-rich
acetylene and benzene flames, 22nd Int. Symp. Combust., The Combustion Inst., Pittsburgh (1988),
pp. 395–401.
7. B.S. Haynes, H.G. Wagner. Soot formation, Progr. Energy Comb. Sci., 7, 229–273 (1981).
8. K.H. Homann, H.G. Wagner. Some aspects of soot formation, in: J. Ray Bawen (Ed.), Dynamics of
Exothermicity (Combust. Sc. Technol. Book Series, Vol. 2), Carbon and Breach Publishers (1996),
pp. 151–184.
9. P.A. Tesner. Soot formation in combustion, Fiz. Goreniya Vzryva, 15, No. 2, 3–14 (1979).
10. H.G. Wagner. Soot formation in combustion, 17th Int. Symp. Combust., The Combustion Inst.,
Pittsburgh (1979), pp. 3–19.
11. I. Glassman. Soot formation in combustion processes, 22nd Int. Symp. Combust., The Combustion
Inst., Pittsburgh (1988), pp. 295–311.
12. H. Böhm, D. Hesse, H. Jander, et al. The influence of pressure and temperature on soot formation in
premixed flames // 22th Symp. (Intern.) jn Combustion. – Pittsburgh: The Combustion Inst., 1988. –
P. 403-411.
13. M. Bönig, C. Feldermann, H. Jander, et al. Soot formation in premixed C2H4 flat flames at elevated
pressure // 23nd Symp. (Intern.) on Combustion. – Pittsburgh: The Combust. Inst., 1990. – P. 1581-
1587.
14. M. Bönig, C. Feldermann, T. Heidermann, et al. Soot formation in premixed C2H4-air flames for
pressure up to 100 atm // 24th Symp. (Intern.) jn Combustion. – Pittsburgh: The Combustion Inst.,
1992. – P. 991-998.
15. M. Frenklach, H. Wang. Detailed mechanism and modeling of soot particle formation // Ed.
Heidelberg Springer Verlag, 1994.
16. H. Wang. Formation of nascent soot and other condensed-phase materials in flames // Proceedings of
the Combustion Institute, 2011. – Vol. 33. – P. 41–67.

19
17. H. Bockhorn. Soot Formation in Combustion: Round Table Discussion / Ed. Heidelberg Springer
Verlag, 1994.
18. N.A. Henein. Analysis of pollutant formation and control and fuel economy in diesel engines // Prog.
Energy Combust. Sci. – 1976. – Vol. 1. – P. 165-207.
19. H. Böhm, M. Bönig, C. Feldermann, et al. Pressure depehdence of formation of soot and PAH in
premixed flames: Soot Formation in Combustion / Ed. H. Bockhorn // Springer Series in Chemical
Physics. – Berlin: Springer Verlag, 1994. – Vol. 59. – Pp. 145–164.
20. H. Bohm, M. Bonig, C. Felderman, et al. Pressure dependence of formation of soot and PAH in
premixed flames, in: H. Bockhorn (Ed.), Soot Formation in Combustion, Springer Series in
Chemical Physics, Springer Verlag, Berlin (1994), Vol. 59, pp. 145–164.
21. J. Hyun, O.L. Gulder. Soot formation and temperature structure in small methane–oxygen diffusion
flames at subcritical and supercritical pressures, Combust. Flame, 157, 1194–1201 (2010).
22. T. Flesch, C. McCarthy, A. Basn, C. Udovich, et al. A new clean diesel technology: Demonstration
of ULEV emissions on a navistar diesel engine fueled with dimethyl ether, SAE paper 950061: Proc.
Int. Congress and Exposition, Detroit, Michigan.
23. M. Frenklach, D.W. Clary, W.C. Gardiner, S.E. Stein. Effect of fuel structure on pathways to soot,
21st Int. Symp. Combust., The Combustion Inst., Pittsburgh (1986), P. 67.
24. I.M. Khan. Formation and combustion of carbon in a diesel engine, Inst. Mech. Eng. Pros., 184, No.
35, 36–43 (1969).
25. N.A. Henein. Analysis of pollutant formation and control and fuel economy in diesel engines, Prog.
Energy Combust. Sci., 1, 165–207 (1976).
26. A. Pungs, S. Pischinger, H. Bäcker, G. Lepperhoff. Analysis of the Particle Size Distribution in the
Cylinder of a Common Rail Diesel Engine during Combustion and Expansion, SAE-Paper No.:
2000-01- 1999, Paris (2000).
27. J.E. Dec, A.O. zur Loye, D.L. Siebers. Soot Distribution in a D. I. Diesel Engine using 2-D Laser-
Induced Incandescence, SAE-Paper No.: 910224 (1991).
28. M. Stumpf, A. Velji, U. Spicher, et. al. Investigations on soot emission behavior of a common-rail
diesel engine during steady and non-steady operating conditions by means of several measuring
techniques. SAE-Paper No.: 01-2154 (2005).

20
 Chapter 2

COOL-SOOTING COMBUSTION OF HYDROCARBONS

AND THE STRUCTURE OF THE FLAME FRONT

The study of the structure of soot formation flames gives valuable information on
chemical and physical processes occurring up to and in time of soot formation.
Investigation at hydrocarbon combustion of kinetics is of great scientific and practical
interest, in particular in relation to the problems of protection of the environment from
the products of incomplete combustion in petrol and diesel engines [1].
Despite a great number of investigations of hydrocarbon combustion now, the
field of low-temperature soot formation (900-1200 К) has not been studied well. The
obtaining of soot, technical carbon (TC) can serve as one of the ways of natural gas
processing, degree of which utilization now insufficiently full. The process how the
preliminary heating of a combustible mixture influences upon the soot formation is not
sufficiently investigated but apparently, the change in the value of a preliminary
heating is one of the parameters of smooth control by completeness of combustion
both conversion in soot and characteristics of formed soot.
As is known [3] the field of cool flames covers a temperature region from 500 to
800 K for long-chain hydrocarbons and 800-1200 K for methane. Soot formation is
frequently observed on the upper temperature boundary of cool flames, but owing to
the difficulty of studying of such flames it is possible to leave the soot formation
region by changing experimental conditions. At the same time the soot formation
process has never been considered as the transition phenomenon between cool and
hot flames, and low temperature soot formation itself has not been investigated in
details.
The way of olefine obtaining at cool flame butane oxidation [4] is described and it
is marked, that on the upper temperature border of cool flames the process is
complicated by soot formation.
The oscillation oxidation regime with soot formation is observed [5] when
studying cool flame oxidation of propane-oxygen mixture in the ratio 1:1 from 720 K
to 800 K. From comparison of oxidation products (Table 1) is possible to assume
certain dependence between soot formation and increase of ethylene concentration
and carbon oxide with possible change of turning mechanism at insignificant increase
of reactor temperature.
Wagner H. et al. for the first time introduced the term "cool sooting flames", when
studying the influence of pressure and temperature on soot formation in previously
stirred ethylene and benzene-air flames [6]. It was shown that the low temperature
threshold of soot formation exists with a high-temperature threshold of soot formation.
In [5] is noted, that the properties of "cool sooting flames" differ from those of
hot flames and have common features with pyrolysis of aromatic compounds. The
necessity of thorough study of "cool flame soot particles" for understanding the nature
and mechanism of low temperature soot formation is indicated.
Investigations on the kinetics of soot formation during combustion of hydrocarbon
is of significant scientific and practical interest [1-4]. One problem is that soot from
the low-temperature combustion of hydrocarbons has polycyclic aromatic
hydrocarbons (PAHs) absorbed on its surface and consequently has carcinogenic
activity.
21
 Table 1

Dependence of concentration of oxidation products on the temperature of the reactor

second section (Р=40 kPа, Т1=300 K, t=10 sec)

Т, К С3Н8 С3Н6О С2Н4 С2Н4О СН3О СН3СНО СО СО2 СН2О Soot

663 1.02 0.08 0.1 - 0.05 0.04 0.2 - 0.04 -
684 0.97 0.05 0.08 - - - 0.16 - - -
696 1.28 0.11 0.16 - - - 0.38 - - -
723 2.25 0.36 0.61 0.05 0.1 0.06 1.22 0.05 0.09 +
740 2.97 0.57 1.22 0.11 0.19 0.09 2.03 0.08 0.16 +
768 2.61 0.4 2.6 0.2 0.3 0.2 2.93 0.15 0.25 +

The influence of electrical field on the processes of hydrocarbon combustion is of

great importance for the directed chemical synthesis and soot formation control
concerning ecology. It is stated that even electric field of low intensity has significant
influence on the combustion kinetics [5]. The process of soot formation in
hydrocarbon flame is determined largely by the fact that soot particle gets positive
charge in the flame [6-9]. It is possible to control such a process by means of external
electric fields. It is worth noting that such investigations are usually carried out using
laboratory plane burners, where the matrix serves as cathode and anode is situated
above the burner; it is a plate with holes for gas withdrawal.
We investigated the soot forming methane combustion in the electric field in a
flowing quartz reactor, the design of which is close to that of industrial reactor. The
aim of work is optimization of combustion conditions by choosing of the electrode
shape and the distance between electrodes, and study on the value of applied potential
and its polarity influence on soot particle sizes and yield.
There is a lack of quantitative data on the paramagnetism of soot particles because
of the experimental difficulties of sampling. And high reactivity of the active centres
on soot particles creates ambiguity in their identification.
It is important [10] to study the kinetics of the initial stages of soot formations; and
it is clear that 'young' soot particles are more active than the 'old' ones (the amplitude
of an electron spin resonance (ESR) signal has been taken as an indicator of
reactivity). The paramagnetism of soot particles sampled from different points of the
flame front when burning propane and oxygen, was investigated, too.
Production of high module carbon fibers on the base of mesophase pitches is now
quite respective. The main raw material for obtaining mesophase pitches are heavy
resins of oil pyrolysis of coal resins. They differ in their fractional composition,
structure and rate of molecule association. Differences in chemical composition and
feedstock structure have an influence on the behavior of these materials during the
therm treatment, on structural features and properties of mesophase pitches. The very
high demands are made to the pitches on the content of aromatic hydrocarbons and
product purity.
We have stated that formation of the whole range of polycyclic compounds is
observed under the conditions of low temperature soot formation in rich hydrocarbons
flames. The resins extracted from these soots have been shown to contain as much as
70 % of polyaromatic hydrocarbons including those highly condensed. In this
connection, the possibility of obtaining mesophase pitches based on the products of
low temperature propane combustion under the conditions of soot formation was
investigated for the first time.

22
 2.1. Procedure of the experiment

The description of the experimental procedure and the analytical technique was
reported in [11-12]. Soot formation in low-temperature methane and propane
combustion was studied in the heated two-section reactors [11]. The experiments with
methane-oxygen mixtures were performed at the atmospheric pressure, varying the
temperature of the first section (T1) from 673 K to 873 K, the temperature of the
second section (T2) from 873 K to 1223 K, the residence time from 1.7 s to 10.2 s, the
ratio of the components being CH4:02=1.63:1.90 and 2.1:1. In comparison with
propane-oxygen flames the regime of soot formation in methane-oxygen flames is
usually observed at low reactor temperatures (T1=423 K, T2=773-973 K), t=2s and the
ratio of the components C3H8 : 02 =1.53:1. Flame temperature was measured by a
chromel-alumel thermocouple (d = 50 m ) in a thin quartz coating preventing radical
recombination and heating.
The soot extract solutions were studied using spectroscopic methods. The UV-
spectra of absorption was registered by a spectrophotometer Specord M-40, the
fluorescence spectra - by a spectrophotometer Hitachi EP-3P. The fluorescence was
exited by a narrow radiation from the xenon lamp from the diffractional
monochromator.
The scheme of the quartz reactor with electrodes is shown in Fig. 1. Potential (100
Å) to the electrodes was supplied by a high-voltage stabilizer of HS type. The shape of
electrodes and the distance between them were chosen in accordance with the greatest
coverage of combustion zone by electric field and the least disturbance of gas flow
electrodes in the reactor. In the first series of experiments the lower electrode was
made as a spiral and was close to a disk by shape. The upper electrode was a metal
dowel in quartz cover.
The distance between electrodes was 90 mm. It is
stated that in this case the flame gets stabilized on
the lower electrode and a great portion of soot
deposits on the cathode; it causes the change in the
combustion process parameters. Therefore, in the
next series of experiments the lower electrode was
placed in the crosspiece between section of reactor,
and the upper one, made as a ring with a nickel net
put on it, was situated over the outgoing tube (the
distance between electrodes was 205 mm). Soot
practically did not deposit on the electrodes.
In the first series of experiments it was applied
to the upper electrode both positive and negative
relatively to the lower electrode potential the value
of which was varied from 0 to 2.2 kV. In the second
series it was applied to the upper electrode negative 1 – two-section reactor. 2, 3 are the
(relatively to the lower electrode) potential the value electrodes of the first and second
of which was varied from 0 to 0.5 kV. series of experiments, 4 is power
The experiments were carried out at tem- supply, 5 is heated soot collector
peratures of the first (Т1) and the second (T2) Fig. 1. Diagram of quartz reactor
sections of the reactor equal to 773 and 1173 K for investigation of methane
respectively. Fuel rate Qcн4 and oxidant Qo2 was combustion in electric field with
equal to 2000 and 1075 cm3/min, respectively. different location of electrodes

23
Combustible mixture was mixed beforehand. Soot obtained in every experiment was
weighed and investigated by electron microscope and diffractometer DRON-0.5 (Cu
K-radiation).
The experiments on the paramagnetism of soot particles were performed at
atmospheric pressure with the temperature of the first section being T1 = 423 K;
temperature, T2, in the second section varied from 923 to 1023 K. The total
consumption of fuel and oxidizer was 6800 cm3/min with [C3H8]:[O2] = 1. The
residence time in the second section was 1.35 s. Sampling was done with a quartz
probe, inserted into the second section of the reactor; the inlet diameter of the probe
was 50 m.
ESR-spectra were registered using the spectrometer JES-Me-3x with the sample in
a special cuvette cooled to the boiling point of nitrogen. Measurements of the g-factor
and splitting were made in comparison with Mn+2 ions against a background of MgO.
The number of spins was determined in comparison with a monocrystalline sample of
CuSO45H2O of the known mass.
Description of the reactor, experimental procedure and data on identification of
polycyclic aromatic hydrocarbons (PAH) at low-temperature soot formation in butane
and propane rich flames corresponding to upper temperature border of cool flames are
given in [6].
To study soot formation in rich methane-oxygen mixture flame the method of
flame stabilization in two-sectional separately heated reactor was used [4-6]. In the
first section methane-oxygen mixture was heated and in the second section the flame
was ignited and then stabilized. The flame was isolated from the pre-flame zone due to
the fact that flux rate in a harrow between sections was many times higher than flame
rate stabilized in the second section.
The formed soot and gas mixture came through exhaust tube to soot collector,
which was made of carbon tissue that allowed dividing soot from gaseous products.
The collector was heated to 423 K in order to prevent water condensation.
The experiments were carried out at the atmospheric pressure, varying the
temperature of the first section (T1) from 723 K to 873 K, the temperature of the
second section (T2) from 973 K to 1223 K, the residence time from 2.5 to 10.2 s, the
ratio of the components being CH4:02=1.5 : 1 -:- 2:1. Temperature T2 was measured by
a chromel-alumel thermocouple (d = 50 m ) in a thin quartz coating preventing
radical recombination. Two more thermocouples were placed outside of the reactor in
order to measure the temperatures of both sections. Thermocouple reading was taken
by potentiometer KSP-4 (precision 0.25).

2.2. Temperature profiles and physico-chemical characteristics

The temperature profiles in Fig. 2 are taken on length of the second section of the
reactor at combustion of methane (T1=773 K, T2=973 K) and propane without and
with addition of benzene at T1=423 K and T2=923, 973 K. For methane-oxygen
mixture, as is seen, there is a brightly expressed maximum at 1273 K.
It is stated that in the course of low-temperature soot formation during propane-
oxygen mixture combustion there is observed a threshold maximum flame
temperature, which does not depend on the temperature value of the second section,
that probably indicating the existence of the limiting stage in combustion regime
during soot formation.

24
 The obtained soot samples were analysed by
gravimetric and X-ray methods as well as with
the help of ESR-spectroscopy. The gravimetric
analysis by Pregle’s method of soot samples
obtained at low temperature showed carbon,
hydrogen and oxygen content to be 90.06; 3.17
and 6.77% respectively. In the soot obtained at
high temperature; the content of carbon is usually
96-99 %. Carbon balance was calculated
according to the results of chromatography and
obtained soot mass. Total methane conversion
makes up 93-95 % but conversion into solid
carbon product (soot) – up to 25 %.
The external surface over surface active
1 – in the inert medium; 1' – in the
substances is equal to 50 m2/g while adsorption flame; of propane flames: in the inert
surface (over nitrogen adsorption) is equal to 53 medium 2 – Т2=923 K, 3 – Т2=973 K;
m2/g, and as roughness coefficient is close to in the flame 2' – Т2=923 K, 3' –
unity, this soot can be considered to be smooth. Т2=973 K; benzene additions 4 – Т2=
In connection with a complicated phase 923 K, 5 – Т2= 973 К
composition of the obtained samples, the X-ray Fig. 2. Temperature profiles of
quantitative phase analysis based on the method methane flame Т2=973 K
offered in [7, 8] previously applied for natural
carbonaceous objects was used. Diffraction spectra analysis (DRON – 1.5 with
modified collimation, CuK – radiation) of the studied soot presents the superposition
of three X-ray amorphous phases: oxygen-containing O-phase, characterised by
position of peaks 002 at 1 =3.7 , boghead or hydrocarbon B-phase at 2 =9.4 – 10.0
 and prographitic crystalline G-phase at 3 =12 . The comparative analysis showed
that diffraction spectra of the studied soot samples differed significantly from
commercial samples of technical carbon by the presence of O- and B-phases in them.
On the other hand, the composition and quantitative ratio of X-ray phases appeared to
be close to natural carbon formations – anthracites. The quantitative analysis showed
that the studied soot had similar composition and O- phase content ranged from 12 to
14 %, B- phase composition ranged from 27 to 31 % and G-phase – from 57 to 60 %.
Diffraction coefficient, M, characterising phases is equal to 63-75 %.
Table 2

Results of radiographie analysis of soot samples

The number of plane The form of a

Т2, К t, c La, Å Lc, Å d002, Å
gratings sample
973 7 54 15.5 3.64 4
Soot
1003 7 50.7 22 3.62 6
973 7 42.7 24.3 3.55 7 Pyrolytic carbon
Note. Mixture composition СН4:О2 – 2:1; Т1 =723 К.

The results of radiographie analysis of soot samples are given in Table 2. The
values of parameters of coherent scattering region La, Lc and d002 characterising the
structure of soot package were calculated on the basis of roentgenogram data (Fig. 2).
Here La is a soot package diameter length, Lc is its height and d002 is the distance
between two next layers in a soot package.
25
 The obtained low-temperature methane soot has a lot of volatile substances. While
studied soot extraction it was found out that from 20 to 30 % of a substance
transferred into extract. From group chromatography data it follows that the extract
contains up to 70 % of polyaromatic hydrocarbons, up to 17 % of resin, 10-11 % of
asphaltenes and 1-2 % of paraffin and naphthenes hydrocarbons [9].
Table 3

Results of ESR method

Type of material Т2, К h, mm Р, mg h/Р, mm/mg

Soot 963 158 2.13 74.178
1193 430 9.0 47.77
Pyrolytic carbon 963 115 2.42 47.52
1193 11 10.13 1.067

The obtained soot was studied by ESR method. Table 3 gives peaks height, h, and
soot and pyrolysed carbon samples mass, P. Samples vacuuming influences slightly
ESR signals width. Samples volume for the analysis is the same for all of them. Soot
deposits at the walls of the reactor’s second section upper area are the pyrolysed (1-3
%) which has more orderly plane gratings structure than soot does.
With the reactor second section temperature increase free electrons quantity in
soot decreases sharply. Thus, at the change of T2 from 1193 to 963 K free electrons
concentration decreases more than 1.5 times.

2.3. Formation of PAH in low-temperature soot flames

The data on extraction of soot formed from a flame of methane and propane are
given in Table 4. It is seen that if the greatest quantity of PAH in methane soot is
equal to 22.31 %, in propane soot it is equal to 50.25 %.
Besides, it is seen in the table that in transition from a methane flame to a propane
flame, the temperature of the reactor at which maximum content of PAH is observed
decreased by 100 K with a significant decrease of T1. The considered data allow to
assume that in under the conditions of low temperature methane and propane
combustion the precursors of soot are PAH.
Table 4

Extraction of soots obtained under the conditions of low temperature methane and
propane combustion, The extracted soot mass – 1 g,
Feed rate of CH4 - 2000 cm3/min, C3H8 - 1600 cm3/min

Consumption of O2, Т1, К Т2,К The mass of The yield of dry

cm3/min residue, mg extract, %
Methane
1 1150 823 873 223.1 22.31
2 1150 823 973 214.0 21.40
3 1150 823 1023 177.4 17.74
4 1400 298 298 301.5 30.15
Propane
5 2450 423 773 502.5 50.25
6 2450 423 873 410.2 41.02
7 2450 423 973 220.0 22.00
26
 When extracting the obtained soots,
it was revealed that from 20 to 50 % of
the substance transmit into the extract.
The analysis of soot extracts is carried out
by IR- and UV-spectroscopic methods.
Such compounds as anthracene, anthan-
trene, 1-12-benzpyrelene, pyrene, coro-
nene, phluorantene are rather reliably
identified. In the absorption spectra there
are a number of bands, which the iden-
tification of which is difficult (Table 5).
That is in extracts of the obtained soots
there are mixtures of different poly-
aromatic compounds, the ratio of which
depends on the conditions of obtaining
soot.
The fluorescence spectra of solutions
of soot extract in iso-octane (exc=294 1 – at exc.= 340 nm, 2 – at exc.= 405 nm
and 405 nm) are taken. In fluorescence
spectra (Fig. 3) the intensive bands with a Fig.3. Fluorescense spectra of a solution of
precise oscillatory structure and main soot extract (solvent – iso-octane)
maxima at 429, 456, 487 and 510 nm are
observed. The quantum output of the observed fluorescence is close to 0.3. The
reference of fluorescence bands to individual compounds is difficult, as in the solution
there is a mixture of polyaromatic products. Nevertheless, it is possible to assume that
in soot formation process there takes place synthesis of PAH having high quantum
fluorescence outputs in blue and green area of the spectrum.

Table 5

PAH identified in soot extracts of low temperature propane combustion

PAH Structural formula Molecular weight

Anthracene
C14H10 178

Pyrene
C16H10 202

Fluoranthene
C16H10 202

Triphenylene
C18H12 228

27
 PAH Structural formula Molecular weight

Anthanthrene
C22H12 264

1.12-benzperylene
C22H12 264

Coronene
C24H12 288

2.4. Soot formation in electric field

Dependence of soot yield on applied potential for the first series (curves 1 and 2)
and the second (curve 3) series of experiments is shown in Fig. 4, a. A decrease in
soot yield from 11.2 % (without potential) to 3 % (at 2,2 kV) occurs at application of
positive potential to the upper electrode (relatively to the lower one) (Fig. 4, a, curve
2). There is a greater decrease in
soot yield when the upper elec-
trode is charged positively rela-
tively to the lower one (Fig. 4, a,
curve 1).
In the second series of expe-
riments (Fig. 4, a, curve 3) two
times decrease in soot yield was
observed even at 0.1 kV under
similar conditions, obtained wit-
hout field application. Further in-
crease in potential causes negli-
gible decrease in soot yield. Ac-
cording to the data of electron
Fig. 4. Dependence of soot yield Р (а) and soot particle microscopy the radius of soot par-
size (b) on the applied potential value at polarity on the ticles, obtained at methane com-
electrodes: 1 - ±; 2 - ; 3 - ± (distance between the
electrodes of 90 mm for 1 and 2, 205 mm for 3)
bustion in electric field of diffe-
rent strength, depends also on the
value of applied potential (Fig. 4, b). It is noting that the particle size value spread
decreases with the increase in field strength; more homogeneous soot forms.
When the electric field direction speeds up positively charged particles a decrease
in interplanar d002 occurs in obtained soot (Table 6). For example, value of d002 is
equal to 3.50 Å at 2.2 kV, but at application of field with inverse polarity the change
in interplanar distance is negligible (d002 = 3.57 Å).
In the second series of experiments, as it can be seen in the table, the decrease in
d002 from 3.58 to 3.52 Å was observed with the increase in U from 0 to 0.4 kV.
The qualitative explanation of the observed phenomena may be made using
28
conceptions of the electric influence on soot forming flame due to the presence of
charged particles in it [7, 9, 13]. Such influence is determined by thermoionization
processes in large hydrocarbon molecules and processes of charge exchange between a
soot particle and the surrounding it cloud of other charged particles.
Table 6
X-ray structural characteristics of soots under the effect of electrical field

U, kV Electrode polarity Quantity of formed soot, d002 , Å

%
The field series of experiments
— 11.21 3,58
0.5 ± 5.9 3.52
1.0 ± 4.01 3.52
1.5 ± 2.64 3.52
2.0 ± 2.02 3.52
2.2 ± 2.02 3.50
0.5 ± 8.82 3.57
1.0 ± 6.62 3.57
1.5 ± 3.06 3.57
2.0 ± 3.03 3.57
2.2 ± 3.00 3.57
The seconds series of experiments
11.21 3.58
0.1 ± 5.7 3.575
0.2 ± 5.6 3.55
0.3 ± 5.2 3.54
0.4 ± 5.5 3.52
0.5 ± 5.3 3.52
1.0 ± 5.0 3.52
1.5 ± 4.6 3.52

In case of electric field application in the way, when the positive electrode is
positioned below, the formed soot particles are charged under thermal electron
emission effect and, having got the positive charge, are sped up by the external field.
Due to the faster passing through the combustion zone, they do not have enough time
to grow to the sizes which are typical for the particles, formed in the absence of
field; it results in the decrease in soot weight. This direction of electric field exerts an
influence on the earliest stages of soot particle formation; it is confirmed by the first
discovered change in interplanar distance d002.
In case of an opposite direction of electric field (positive electrode is above) soot
particles are trapped by the field in the negative charge area. At the simultaneous
effect of thermal electron emission and recharging, the latter is the prevailing
mechanism of charge obtaining. And soot particles, formed in the positive charge area
and charged positively, are recharged and speeded up by external electric field after
negative charge area passing.

2.5. Paramagnetism of soot particles

Investigating the kinetics of soot formation during the combustion of a

hydrocarbon has significant scientific and practical interest. One problem is that soot
from the low-temperature combustion of hydrocarbons has polycyclic aromatic
29
hydrocarbons (PAHs) adsorbed on its surface and consequently has carcinogenic
activity. In spite of all the work [11, 16-18] there is still little clarity on the embryos
participating in the formation and growth of soot particles. Quantitative data on the
paramagnetism of soot particles are lacking, because of the experimental difficulties of
sampling; also the high reactivity of the active centers on soot particles creates
ambiguity about their identification. It is important [19] to study the kinetics of the
initial stages of soot formation; so far it is clear [1] that “young” soot particles are
more reactive than the “old” ones, with the amplitude of an electron spin resonance
(ESR) signal having been taken as an indicator of reactivity. This chapter investigates
the paramagnetism of soot particles sampled from different points of the flame front,
when burning propane and oxygen.
This chapter uses ESR spectroscopy
to study soot particles, produced in rich
flames of propane and oxygen, when
stabilized in a quartz reactor, with two
separately heated sections. The reactor
and flame stabilization on a matrix have
been described before [11]. The experi-
ments were done at 1 atm with the first
section’s temperature being T1 = 423 K;
the temperature, T2, in the second section
varied from 923 to 1023 K. The total
consumption of fuel and oxidizer was
6800 cm3/min with [C3H8]/[O2] = 1. The
residence time in the second section was
1.35 s. Sampling was done with a quartz
probe, inserted into the second section of
Fig. 5. The temperature profiles along a reactor the reactor; the inlet diameter of the
with a flame of propane and oxygen [C3H8]/[O2] = probe was 50 mm. ESR spectra were
1 and T1 = 423 K [20] registered using the spectrometer JES-
Me-3x with the sample in a special
cuvette cooled to the boiling point of nitrogen.
Measurements of the g-factor and splitting were done
in comparison with Mn+2 ions against a background
MgO. The number of spins was determined in compa-
rison with a monocrystalline sample of CuSO4  5H2O
of known mass.
The temperature profiles along the reactor’s length
with a flame of propane and oxygen are shown in Fig.
5. Soot particles were sampled from the luminous zone
represented by the shading. The ESR spectra of low-
temperature soot sampled from various points in a
flame front are given in Fig. 6. These ESR spectra are
symmetrical singlet signals, characterized by a g-factor
of 2.003 ± 0.001. Its shape for all samples is
T1=423 К, Т2=923 К; the values
of z were 1: 4.3cm, 2: 4.8 cm,
symmetrical about the center and is described well by
3:5.3 cm, 4:5.8cm a Lorentz function. The width of the modulation was
0.2 mTl. The value of the g-factor and the small width
Fig. 6. The ESR spectra of low- of the modulation indicate that the signal was caused
temperature soot [20] by organic paramagnetic compounds; moreover, the
30
closeness of the g-factor to that for a free electron (2.0023) shows that the signals are
associated with the products of low-temperature (< 973 K) pyrolysis of organic
compounds [14]. According to Ingram [14] these signals belong to unpaired electrons
on molecules of PAH. The concentration of paramagnetic centers (PMC) in a sample
is C ~ 1019–1020 spin/g as calculated from ESR spectra. The value of C recorded at
various times after soot sampling is
shown in Fig. 7 for different values of
z, the distance from the matrix of the
reactor. It appears that C decreases
with time; moreover initially (~0-25
h) it decreases faster than
subsequently, but eventually (after
~200 h) this concentration becomes
stabilized. Therefore it may asserted
that the initial soot particles, i.e. the
so-called “young” soot, are more
active, possibly pollute the
environment faster and age with time.
The concentration of paramagnetic
centers decreases, i.e. the ESR signal
intensity decreases, is possibly due to
radical recombination on forming
cyclic dimers [15]. Fig. 7. Plot of concentration of paramagnetic
The profiles of the concentration of centers against the time after sampling [20]
radical centers in soot sampled from
different points along the flame front
are shown in Fig. 4. The samples were taken from the lower boundary of the luminous
zone (z = 4.3 cm), from the luminous zone (z = 4.8 cm and 5.3 cm), and from the
upper boundary of luminous zone (z = 5.8 cm) of the propane and oxygen flame. The
radical concentration profiles for T2 varying from 923 to 1023 K, passed through a
maximum, as shown in Fig. 8. The maximum concentration of paramagnetic centers in
this propane–oxygen flame was observed at a distance from the matrix of the reactor,
z, of ~4.8 cm and the maximum concentration decreased if T2 was raised.
The maximum concentration of
radical centers was observed at T2 =
923 K. One can see from Fig. 8 that the
number of radical centers falls when the
second section’s temperature, T2, is
increased with T1 = 423 K. If the probe
is moved to a larger z, the dependence
of C, the concentration of radical
centers on temperature becomes less
sharp (Fig. 8). Thus the radical
concentration decreased sharply with
increasing temperature at z = 4.3 cm;
for example at 923 K it is 2.9 x 1020
spin/g, and at 1023 K it is 2 x 1019
spin/g. At z = 5.8 cm Fig. 8 shows that
C practically did not change any more. Fig. 8. The concentration profiles of paramagnetic
centers for different temperatures, T2

31
 2.6. Mesophase pitches

In [14] PAH tendency to form liquid-crystalline phase at unusually high

temperatures of 753-773 K was discovered. Apparently, it can be explained by the fact
that soot particles with PAH adsorbed at their surface pass through flame front of 973-
1273 K, i.e. they undergo high-temperature quenching.
The possibility to obtain mesophase pitches based on low-temperature hydro-
carbons combustion products in sooting regime, changes in thermopolycondensation
process in chemistry and the structure of resins and pitches was investigated in [15].
The determination of thermopolycondensation optimal conditions was also studied.
The study of propane and propane-benzene soot extraction resins, the same resins
overprecipitated by hexane and isooctane with the help of IR-spectroscopy method
showed that with a few exceptions all of them had similar absorption bands
The intensive absorption bands of aromatic structures were observed in the area of
700-800, 1540-1560, 1610-1630, 3050 cm-1 as well as the intensive absorption bands
of methylene (2930 cm-1) and methyl (2960 cm-1) groups. Methylene groups were
presented by long chainlike structures. The observed weak band of 1720 cm-1 allowed
supposing that oxygen containing compounds with C = O bond were in soot extract.
Besides, there were significant differences in band intensity distribution. The
absorption band intensity of aromatic structures increased for benzene resins in
comparison with propane ones. The intensity of these bands also increased after resins
over precipitation by isooctane and hexane. The exception was propane soot extraction
resin. Its spectra showed higher intensity of methyl groups absorption and lower
intensity of methylene groups. Consequently, it had either short chains or strongly
branched aliphatic structures. Another distinguishing feature of propane resin IR-
spectrum is in intensive band presence at 1100 cm-1 which corresponds to ether bond
C – O.
After propane resin over precipitation by hexane aliphatic hydrocarbon group
content changes and pitch at 1100 cm-1 disappears.
It was found out that resin over precipitation led to resin chemistry change which
influences mesophase pitches structure obtained at resins thermopolycondensation.
The extraction soot is the initial raw material for mesophase pitches obtaining as it
satisfies the requirements for the raw material, i.e. it has organic molecules of planar
structure including several aromatic cycles.
The work on thermopolycpndensation process optimal conditions determination
was held. For pitches obtaining from soot extraction resin a special device was used,
its scheme is given in [15].
Fig. 9. shows the dependence of pitch softening temperature derived from propane
soot extraction resin and propane-benzene soot extraction resin over thermal treatment
temperature. Extraction soot thermopolycondensation was held at temperatures of
373-673 K in argon medium.
From the given data it follows that with the thermal treatment temperature increase
pitch softening temperature also increases, i.e. PAH with high molecular masses
content rises.
From this figure it can be seen that resin treatment with hexane extraction
increases pitch softening temperature in comparison with initial one. It is also seen
that the temperature of pitch softening which was derived from propane soot
extraction resin is significantly higher than that of the pitch derived from benzene soot
extraction resin. It can be explained by the difference of pitches group composition.

32
 Carbonization process resides in continuous enlargement of a number and sizes of
polymating aromatic regions with the following straightening out of their over
molecular structure. In each moment
of carbonization process the formation
of paramagnetic centers (PMC) can
take place as well as their recombina-
tion. However, during carbon ization
products cooling to the room tem-
perature only the certain part of para-
magnetic centers stabilises which is an
integral part of polycondensed aro-
matics. Thus, ESR signal parameters
depend on their quantity, size and
mutual location. That is why the study
of pitches derived from soot extraction
resin was carried out with the help of 1 – propane soot extraction resin; 2 – benzene
soot extraction resin overprecipitation by hexane;
ESR method.
3 – benzene soot extraction resin
Pitch samples and 1-fraction
spectra (1 is the fraction characterised Fig. 9. The dependence of pitch softening
as pure mesophase) are symmetrical temperature from thermal treatment temperature
singlet signal with a line width from
4.5 to 5.6 Gs for different samples with g-factor which is equal to 2.0027  0.0003.
The values of g-factor are close to theoretical magnitudes: for isolated aromatic -
radicals it is 2.00266 and for model mesophase cokes obtained at aromatic
hydrocarbons pyrolysis it is 2.002743 [16]. This is a strong argument to refer ESR
pitches signals, obtained from soot extraction resin to aromatic -radicals.

Table 7
Some characteristics of soot extraction resin

Factors Resin extracted Resin extracted over precipitated by

isooctane
Density (kg/m3 ) 998.2 994.4
Coking capacity (% ) 10.8 11.4
Ash content (% ) - -
Group chemical composition
(mass.%):
Paraffin and naphten hydrocarbons 6.2 -
Easy and middle aromatic 16.0 11.3
hydrocarbons
Heary aromatic hydrocarbons 61.3 74.8
Resins 15.4 12.9
Asphaltenes - -
Element composition (mass.%):
Carbon 92.15 91.26
Hydrogen 7.64 7.77

The concentration of PMC in pitches depends on the initial raw materials and
softening temperature. Thus for propane-benzene resin pitch PMC concentration is
1.631020 PMC/g, the concentration of PMC for the same pitch with 10 % of phthalic
anhydride is 1.91020 PMC/g. As the pitch softening temperature increases from 95 to
378 K PMC concentration grows from 2.721020 to 3.01020 PMC/g and a mesophase
33
extracted from the pitch contains 8.101020 PMC/g. The higher content of
paramagnetic centers in the mesophase is determined by the presence of the most
developed mate systems in it, which consists of plane molecules.
When the characteristics of soot extraction resins were studied, they were found to
be similar to resins of pyrolysis by their factors, however the content of aromatic
hydrocarbons in it was higher considerably. Some characteristics of the soot extraction
resins are presented in Table 7.
The rate of aromatic features is one of the important characteristics of resins.
For increasing the rate of aromatic features of the soot extraction resin we treated
resin with hexane and isooctane by estimating the rate of aromatic features of
unsolved fraction. The rate of aromatic features (η) was determined according to the
IR-spectroscopy data, η expresses the ratio of integral intensity of absorption in the
region of 3045 cm-1 to that in the region 3000 - 2750 cm-1, i.g. it is a ratio of
hydrocarbon group contents which are absorbed in these spectrum regions.
The studies of group chemical composition of thermopolycondensation products
have shown that pitches obtained from the initial resin and from resin treated with
solvents differ noticeably in their composition. The data on group chemical
composition of pitches have been presented in Table 8. Pitches obtained on the base of
resin extraction resettled by hexane have the least content of a1 -fraction. Pithches
obtained on the base of propane-benzene resin with activying addition of phthalic
anhydride have the highest value of α1 -fraction characterized as pure mesophase.

Table 8

Group chemical composition of hermopolycondensation products (%). The

thermopolycondensation period is 4 hours, temperature is 653 K

Sample α-fraction α1-fraction α2-fraction β-fraction γ-fraction

MP-1 61.4 26.7 34.7 34.1 4.5
MP-2 34.0 18.8 15.2 52.2 12.7
MP-3 38.4 30.0 8.4 58.0 3.6
MP-4 35.2 12.1 23.1 52.8 12.0
MP-5 53.4 26.9 26.5 32.2 15.8
MP-6 64.9 48.2 18.7 29.0 5.1
α-fraction unsoluble in benzene; α1-fraction unsoluble in pyridine; α2-fraction soluble in pyridine
and unsoluble in benzene; β-fraction unsoluble in isooctane, but soluble in benzene; γ-fraction soluble in
all listed solutions.

The structure of mesophase pitches obtained from resin extraction of low

temperature soot has been compared by radiographic quantitative phase analysis.
All the samples studies are characterized by complex but similiar type of
diffractions, which is the superposition of reflexes of several radiographic phases.
After desintegration of diffractional spectrum the reflexes of the following
components: graphite-like phase (Gf) with d ~ 3.48 - 3.86 Å and polynaphtene phase
(Nf ) with d~ 4.7 Å have been extracted. On a level with the determination of
quantitative radios of the pointed phases for Gf the values of d002, characterizing
interlayer distance Lc corresponding to the height of piles of graphite-like molecules ,
La corresponding to the layer diameteder and the graphitization coefficient Kg have
been calculated. The latter is calculated with a formula [17]:

Kg = ( L002 .10-2)/(d002-3.35)
34
The data have been presented in Table 9.
Table 9
X-ray characteristics of pitches

Radiographic phases (%) For Gf nm Number Kg

Samples Gf (002) Nf (γ) d002 Lc La of layers
MP-1 88 12 0.381 2.1 3.07 5.5 0.45
MP-2 87 13 0.378 2.0 2.85 5.3 0.47
MP-3 89 11 0.380 1.8 3.21 4.7 0.40
MP-4 84 16 0.386 1.6 2.54 4.2 0.31
MP-5 85 15 0.383 1.6 - 4.2 0.33
MP-6 91 9 0.359 2.3 2.86 6.4 0.96

The following conclusions can be drawn. X-ray structural characteristics of the

“organized” part of pitches and its fractions are close to those of the oil and coal
pitches and asphaltenes which have the layer-block overmolecular structure
characterized by the association of polynuclear plates in compounds consisting of 5-6
layers [20].
Table 10
The dependence of x-ray pitch characteristics on temperature. The
thermopolycondensation period is 4 hours

Thermal Radiographic Nf (γ) For nm Number Kg

treatment phases (%) d002 Lc of layers
temperature
(K)
453 87 13 0.383 1.9 4.9 0.40
533 88 12 0.380 1.9 5.0 0.42
593 88 12 0.377 1.9 5.0 0.45
673 89 11 0.364 1.8 4.9 0.64

The process of structural changes, which take place during the resin
thermopolycondensation when the temperature of treatment increases have been
studied. The data obtained have been presented in Table 10.
When temperature of thermal treatment increases the structure ordering takes
place. Parameter d002 characterizing interlayer distance decreases while the Lc value
corresponding to the height of piles of graphite-like molecules increases, therefor a
number of layers in piles increases.
A possible scheme of a cold-flame soot formation was constructed on the basis of
the literature and experimental data:

the diffusion from the combustion zone to the preflame region of H atoms and e−;

the formation of radicals as well as ethane and ethylene ions;

the growth of radicals and ions and the formation of acetylene, propylene, hexane,
benzene, benzene substituents, and carbenes on the basis of the olefins;

the formation of PCAHs.

35
CONTROL QUESTIONS

1. What do you think of cool-sooting combustion of hydrocarbons?

2. What do you think of the structure of the flame front?
3. What do you know about formation of РАH in low-temperature sooty flames, soot formation in an
electric field?
4. Can you explain the temperature profiles of cool-sooting combustion of hydrocarbons?
5. Can you explain roentgenographic analysis results of soot samples?

REFERENCES

1. H. Wagner, Gr. in Particulate Carbon: Formation During Combustion (D. C. Siegia and G. W. Smith,
Eds.), Plenum, New York, 1981. P. 3-29
2. R.Y. Mayo, F.G. Weinberg // Proc. Roy. Soc. A. London. 1970. V. 319. P. 351-371
3. V.Ja. Shtern. The mechanism of oxidation of hydrocarbons in a gas phase. AN USSR. 1960. P. 493
4. The patent of USA. № 2706740. P. 1955
5. M.D. Pogosyan, T.G. Simonyan, S.D. Arsentev, A.A. Mantashyan Laws of heat release and
accumulation of radicals in the field of cool flame and high-temperature oxidation of propane //
ITPM AN USSR. 1988. P. 59
6. H. Bohm, H. Wagner. et al. // 22-nd Symp. (Int). on Combustion. 1988. P. 403-411
7. В.L. Wersborg, A.C. Yeung, J.B. Howard // 15 th Symp. (Int). on Combustion. Pittsburg: The Comb.
Institute.1975. P. 1439-1448
8. К.H.Herman. Ber. Bunz. // Phys. Chem. 1979. V. 83. P. 733-745
9. F.Weinberg // Far. Soc. Symp. 1973. P. 120-132
10. A.A. Merculov, A.A. Ovsyanikov, L.S. Polak, V.T. Popov, Chem. Plasma Processing. 1989. V. 9,
№ 95.
11. Z.A. Mansurov, V.I. Pesterev, G.O. Turesheva, V.T. Popov. Archivum Combustionis. 1990. V. 10,
№ 209
12. Z.A. Mansurov, A.A. Merkulov, V.T. Popov, B.K. Tuleutaev, N.S. Almasov. The formation of
ultradispersive soot at combustion of methane in an electrical field // Chemistry of a solid fuel. 1994.
V. 3. P. 83-86
13. Dg. Lauton, F. Weinberg. Electrical aspects of burning // M.: Energy. 1976. P. 268
14. Ingram, D. ESR in biology, Mir, Moscow, 1965 (in Russian).
15. Waring R., and Sloan G., Chem. Phys. 40:772 (1964)
16. Merculov, A. A., Ovsyanikov, A. A., Polak, L. S., and Popov, V. T., Chem. Plasma Processing 9:95
(1989)
17. Haynes B. S., and Wagner, H. G., Prog. Energy Combust. Sci. 7:229 (1981)
18. Lahaye J., and Prado, G., in Chemistry and Physics of Carbon, New York, 1978, p. 167.
19. Tesner P. A., Fiz. Gor. Vz. 3:48 (1972).
20. ZA. Mansurov, E.K. Ongarbaev, T.T. Tutkabaeva. The paramagnetism of soot particles in propane-
oxygen flames // Combustion and Flame 118:741–743 (1999).

36
 Chapter 3

MECHANISM OF SOOT FORMATION

The soot formation process is a complicated physicochemical phenomenon. Until

present time there is not generally accepted concept of soot formation and soot
particles structure and as the consequence the microscopic model of the process. Such
outstanding authorities as H. Wang [1], P.A. Tesner [2], A.V. Krestinin [3] J. Howard
[4], M. Frenklach [5], H. Bockhorn [6], A. Violi [7], D. Anna [8], and J. Lahaye [9]
are the experts in the field of technical carbon production, they have made an
significant contribution to the understanding of soot formation process, and for the last
30 years the positive progress has been achieved in the field of soot particles
formation mechanism. Today, several models of soot formation are well known, that
describes the detailed mechanism of it formation. In this chapter we consider the most
important soot formation mechanisms such as: polyyne model (Krestinin et al., 2000),
polyaromatic model (Frenklach et al., 1991) and ion model (Calcote et al., 1990).
Any theoretical model of soot formation has to answer the following key
questions:
1. Which species are the gaseous precursors of soot particles, i.e. which gas-phase
molecules form primary aerosol particles?
2. What are the nuclei for soot particles and why they do grow irreversibly at the
stage of nucleation?
3. How can the high rate of surface growth of soot particles during nucleation be
explained?
Many researchers believe that polycyclic aromatic hydrocarbons are precursors of
the nucleation of soot particles [10]. For example, in [11], two types of polyaromatic
molecules were detected n a diffusion acetylene-oxygen flame:
1) polycyclic aromatic compounds with no side chains: naphthalene,
acenaphthenyl, coronene, phenanthrene;
2) polycyclic aromatic compounds with side chains (a molecular mass of 150–
500).
The concentration of the first-group of PCAHs in the soot-formation zone was
practically unchanged, while the concentration of the second-group PCAHs increased
sharply.
In the zone of intensive soot formation, yet another large group of molecules –
polyacetylenes, representing the intermediate substances in the formation of
polyaromatic compounds, was detected. The mechanism of this process was
considered in detail in [12].
As is known, the soot particles acquire a charge at the early stages of soot
formation. Investigations carried out in [13] have shown that, at the initial instant of
time, the soot particles are charged mainly negatively. In this case, the charging
process is predominantly determined by the interaction of the soot particles with
electrons having a small mass, a high mobility, and, consequently, a large adhesion
coefficient. With time, as a result of the interaction of the mainly negatively charged
soot particles with positive ions (C2H3+ and C3H3+ for the most part), the charge
distribution of the nanoparticles becomes more symmetric. In this case, the positively
charged particles account for approximately 36 %, the neutral particles 24 %, and the
negatively charged particles 40 % of the soot particles [13]. It should be noted that
37
recently a symmetric charge distribution of the soot particles was detected in a
propane-air flame; the maximum charge of these particles did not exceed ± 10e [14].
These results were obtained for the atmospheric pressure flame and for the flame in
the combustion chamber of a power plant and support the assumption that a soot is
formed by the ion model [15].

3.1. Polyyne model of soot formation

Polyyne model of soot formation was proposed A.V. Krestinin [16]. The
beginning of the creation of this model lies “the acetylene pathway” to soot, first put
forward by Berthelot [17] and later supported by others [18, 19]. Proponents of this
hypothesis emphasize that the thermodynamic stability of acetylene (and the whole
family of polyynes (PYs) C2nH2, n = 2, 3,...) grows at higher temperatures, while the
stability of all other hydrocarbons decreases [20]. Having discovered polyynes in the
sooting zone of a flame, Bonne, Homann and Wagner [11, 21] suggested that polyynes
play a key role in soot formation. Only quite recently this idea received further
development in a “polyyne model”, which describes soot nucleation as a fast
polymerization of a “supersaturated polyyne vapor” [20, 23-24].
On the basis of the schematic is shown the polyyne model of soot formation, in
Fig. 1 [25]. During the induction time the pyrolysis of initial fuel results in the
formation of polynuclear aromatic hydrocarbons and polyyne molecules. Compared to
a rather slow, multi-stage increase in the number of aromatic rings in the PAHs, the
polyynes grow in a simple and fast way [26, 27], typically in:
C2nH2 + C2H = C2n+2H2 + H , k+ = 2×1013 cm3/(mol s), n = 1, 2, 3, ...
Consequently, the concentrations of the polyynes C4H2 , C6H2 , C8H2 ... are high
enough to be detected as major hydrocarbon intermediates in both pyrolysis and
combustion processes [20, 27-31]. This, together with the high reactivity of these
species in polymerization reactions [32], suggests that polyynes can be assumed to be
the gaseous precursors of soot. This is the basis of the polyyne model [25].

Fig. 1. An outline of the soot formation mechanism adopted in the polyyne model [25]

38
 It is assumed that polyyne molecules are capable of forming a polymer as the
product of a surface polymerization process involving radicals. Such polymerization is
assumed to occur only if the growth rate of the polymeric layer on the surface exceeds
the rate of carbonization of this polymer. In the model, two different mechanisms for
the formation of carbonaceous material on a surface are accordingly distinguished.
The first mechanism produces pyrolytic carbon, when Wgr is less than Wcarb, where Wgr
and Wcarb are, respectively, the rates of surface growth and carbonization of polymeric
material. The second mechanism forms polymeric material (coke), when Wgr is greater
than Wcarb. The surface of the polymeric layer is supposed to consist of C-H groups,
which are capable of forming local radical sites after the elimination of H-atoms.
Surface growth of the polymeric layer occurs predominantly on those radical centers.
Polyyne polymerization on the radical centers is assumed to be irreversible.
Further, it is assumed in the model that polyynes are capable of causing
“breeding” to their chain polymerization process, as soon as a polyyne complex is
formed on a surface radical site; this is schematically shown in Fig. 2. In this case, two
additional radical centers are created by cyclization taking place in such a complex.
Such so called “breeding” of radical sites during polyyne polymerization is a key
assumption, which brings nucleation into the model. In fact, nucleation begins when
the condition:

Wbr >Wterm + Wcarb (1)

holds and continues so long as this inequality is valid. Here Wbr is the rate of breeding
radical sites (mol/(cm2 s) and Wterm + Wcarb is the rate of their decay. During this
period the population of radical sites on the surface of a growing particle is equal to its
maximal value; therefore this process is denoted as fast polymerization.

Fig. 2. Formation of a polyyne complex on a surface radical site, and its transformation into
a more stable structure, followed by “breeding” of radical sites [25]

The inequality (1) may be used as a definition of supersaturation of the polyyne

vapor. Indeed if this condition is fulfilled, the probability of breeding at every surface
radical site is greater than the probability of its decay; consequently, every gas phase
radical capable of maintaining the formation of a polyyne complex becomes a center
for polymerization and grows irreversibly to a polymeric globule - a primary soot
particle. In the polyyne model the nuclei for soot particles are radical centers in
polyyne polymerization at the stage of fast polymerization [25]. Based on this, it is
assumed that the nucleation rate is equal to the rate of formation of such polyyne
complexes, capable of transformation into stable multiradicals. These complexes are
assumed to be formed in reactions between a polyyne molecule and a polyyne radical
or between two polyyne molecules [33].

39
 3.2. Polyaromatic model of soot formation

The "aromatic" pathway of soot formation was theoretically investigated for

determining the mechanism of coagulation of aromatic components with formation of
primary aerosol particles [34]. It was inferred in
[35] that polycyclic aromatic hydrocarbons
(PAHs) and soot can be formed in successively
increasing cycles of the reaction with the
participation of acetylene (the mechanism of
hydrogen-abstraction – carbon-addition), Fig. 3.
The significance of this reaction was substan-
tiated in the first attempt to numerically simulate
Fig. 3. Scheme of the soot formation in a the detailed kinetics of growth of aromatic
hydrocarbon flame [35] compounds.

3.2.1. Hydrogen-abstraction – carbon-addition mechanism

The term “hydrogen-abstraction – carbon-addition” (HACA) mechanism was

introduced pioneered by Frenklach and Wang [5, 36, 37], it captures the essence of the
thermodynamic and kinetic requirements for the sooting process. Soot is a ubiquitous
byproduct of fuel-rich flames above a certain fuel-to-oxidizer threshold. Hence, it is a
logical first postulate that the kinetic processes that drive the flame phenomenon must
be related intimately to the sooting process. Factors important to both flame chemistry
and soot formation are: acetylene – the most abundant hydrocarbon intermediate – as
the molecular building block, the H atom – the driving force behind chain branching
and flame propagation – as the “catalyst”, and the high temperature that facilitates fast
kinetics, but at the same time limits the molecular mass growth because of fragmen-
tation of the precursor and possibly soot itself [37].
The main essence of this mechanism implied a repetitive reaction sequence of two
principal steps: (I) abstraction of a hydrogen atom from the reacting hydrocarbon by a
gaseous hydrogen atom,

Ai + H → Ai- + H2 (2)

followed by (II) addition of a gaseous acetylene molecule to the radical site formed,

Ai- + C2H2 → products (3)

Ai is an aromatic molecule with i peri-condensed rings, and Ai- is its radical. The
significance of this (2)-(3) reaction-path feature was identified in the very first attempt
at numerical modeling of detailed reaction kinetics of the growth of aromatics [38].
The full description HACA mechanism is shown in Fig. 4.
The essence of the two-step feature is as follows: the first step activates a molecule
to further growth by converting it to a radical. Of course, this can be accomplished in
many different ways, and many such possibilities have indeed been tested. It turns out
that, under the conditions of typical shock-tube and flame experiments, which provide
the majority of the present training experimental data, it is the H abstraction by a
gaseous H which typically dominates [38-40]. Obviously, if other reactions are to
contribute to the process of converting an aromatic molecule to a radical, either as a
result of different reactor conditions or by discovery of new chemistry, they should be
40
included. The key feature of the first step of HACA, however, is its reversibility. The
reverse steps can be the reverse direction of the H abstraction itself,

Ai + H2 → Ai- + H (4)

or other reactions, such as the combination with a gaseous H,

Ai + H → Ai- (5)

The contribution of reaction (5) as compared to the reverse of (4) increases with
pressure and molecular size, i.e., as the rate coefficient of (5) approaches its high-
pressure limit.

Fig. 4. A version of the hydrogen-abstraction – carbon addition (HACA) reaction mechanism.

The standard Gibbs function per carbon atom is plotted relative to an initial system with
benzene:acetylene:H atom = 1:5:1 at 1800 K [37]

The reversibility of the acetylene addition step, reaction (3), or, more precisely, the
degree of its reversibility, is what determines whether this step will contribute to
molecular growth. For a simple addition, due to the entropy loss, the reaction is highly
reversible, and often runs in reverse. Forming a hydrogen atom as a product,

Ai- + C2H2 → products + H (6)

recovers some of the entropy but still, in many cases, the reaction is still highly
reversible, like in the case of reaction

Ai- + C2H2 ↔ AiC2H + H (7)

Only when, in addition to the recovery in entropy, the decrease in energy is large
enough, does the reaction become more irreversible, and in the formation of
particularly stable aromatics, referred to as islands of stability [38] or stabilomers [41]
the reaction becomes practically irreversible [42].
A notable advance in the kinetic development of aromatics chemistry came from
the work of Dean [43] who demonstrated the importance of chemically activated
reactions in combustion kinetics. This led to the work of Westmoreland et al. [44] who
elucidated a number of critical chemically activated pathways to the first aromatic ring
formation. Further work by Wang and Frenklach examined the thermochemical [45]
and transport properties of aromatics [35], and the kinetics of chemically activated
41
reactions for aromatics [46]. This work led to the development of a kinetic model for
PAH formation [42] (which was later improved, e.g., [47]) and many variations used
for the detailed modeling of soot formation.

3.2.2. Formation of the first aromatic ring

The formation of the first aromatic ring in flames of nonaromatic fuels begins with
the formation of a vinyl-acetylene compound. At high temperatures, this ring arises as
a result of the combination of acetylene with the n-C4H3 radical formed as a result of
the detachment of H from the vinylacetylene [48]. At low temperatures, n-C4H5 is
formed in the process of interaction of acetylene with vinyl; this radical gives rise to
the formation of benzene. Benzene and phenyl radicals can be transformed in one and
the same reaction of detachment of H, in Fig. 5 [36].
The calculations were carried out in the temperature range 1500÷2500 K at fuel-
oxidizer ratios of 4:1, 3;1, 2:1, 1:1, and 1:2. The radical-chain mechanism of methane
combustion, including 87 elementary reactions, was investigated. The values of the
preexponential factors, the activation energy, and the reaction order were taken from
[49]. For reactions 60-87, these constants are unknown, and they were varied such that
a maximum concentration of PAHs was obtained. In this case, the preexponential
factor was changed within the range 1012÷1014, which corresponds to the change from
the radical mechanism to the ionic mechanism of formation of polycyclic
hydrocarbons.

High-temperature pathway

Low-temperature pathway

Fig. 5. Scheme of formation of the first aromatic ring in a nonaromatic-fuel flame [36]

Slavinskaya and Frank [50] have developed a relatively short kinetic mechanism
(93 particles and 729 reactions) for description of the formation of polyaromatic
hydrocarbons and their growth with formation of five aromatic rings in methane and
ethane flames on the basis of analysis of the literature data published during the last 30
years. The PAH reactions proceeding by the low- and high-temperature (lower and
higher than 1550 K) mechanisms were analyzed in detail. Fig. 6 presents the most
significant reactions of formation of molecules of the first aromatic ring at
temperatures lower than 1550 K, among which the key reaction is H2CCCH+C2H2 =
C5H5. At this temperature, the propargyl radical reacts with C4H3 and, in doing so,
forms the first C5H5 radical.
The reactions of C4H3, C4H4, and C4H5 with acetylene and the ethynyl radical play
a large role in the formation of benzene in this temperature regime. The other
molecules with one aromatic ring and the radicals with side chains are formed in the
42
reaction of cyclopentadienyl with acetylene and propargyl from phenyl and benzene.
The possible reaction pathway of sterol and phenylacetylene is the oxidation of
indenyl in the main reaction zone rich with the radicals O, OH, and HO2. Indenyl can
be formed in a large amount as a result of the recombination C5H5 C4H2 even at a
temperature of 1300 K.

Fig. 6. Main reaction pathways of formation of a molecule of one aromatic ring

at T 1550 K in methane and ethylene flames. H2CCCH is denoted as C3H3 [50]

In [41], the energy of transformation of some PCAHs was estimated with the use
of the thermochemical group method and, on the basis of the data obtained, the
constants of equilibrium between the PCAHs and H2 and C2H2 were calculated. An
important result of this work is the deduction that certain strongly condensed "critical"
PCAHs, the concentration of which is minimum under external conditions varying
within a wide range, are present in an equilibrium homogeneous system at T > 1700
K. This deduction allowed the conclusion that a thermodynamic barrier exists for the
formation of soot in a high-temperature homogeneous system.
The results of the simulation performed in [51] point to the fact that benzene can
form in the fuel-rich diffusion premixed flames mainly in two ways with the use of
resonance-stable radicals: in the process of the reaction of allyl and propargyl radicals
and in the spontaneous reaction of propargyl radicals.
The formation of aromatic compounds and PAHs in an industrial combustion was
simulated with the use of the detailed kinetic model of [52]. A high-turbulence, high-
reaction zone of a typical industrial diffusion combustion was represented with the use
of a perfect-mixing reactor model. With this model, the emission of aromatic compo-
unds and PAHs was determined at a varying equivalent ratio. In the full-scale
industrial furnace at the Sandia National Laboratories, the temperature in the perfect-
mixing reactor was approximately 1500 K. The composition of the fuel was selected
such that it corresponded to the fuel used in the experiments: 16 % of H2, 7.3 % of
propane, and 76.7 % of natural gas. The time of precipitation was taken to be equal to
10 s for approximate determination of the benzene concentration in the exhaust gas.
The results of comparison of the calculation with the experiment are shown in Fig. 7.
It is seen that detailed chemical kinetic models can be useful for the interpretation of
experimental data [48].

43
 The soot formation in the process of oxidation of hydrocarbons in shock waves
was experimentally and theoretically investigated in [53].

3.3. Ion model of soot formation

In the middle of 60s of last century, along with polyyne and polyaromatic model
for soot particles formation the ion model was developed too.
Different mechanisms have been
proposed to explain the existence of
charged carbon clusters and soot
particles in hydrocarbon/air flames. For
example, Calcote and Keil [54]
hypothesized that charged soot particles
grow from positive heavy hydrocarbon
ions due to ion-induced nucleation. This
mechanism can explain the appearance
of positively charged particles only. In
order to explain the formation of neutral
and negatively charged particles,
additional mechanisms must be invoked
in the model. The hypothesis of Calcote
and Keil was based on the observed
strong decrease in positive hydrocarbon
Fig. 7. Comparison of experimental data (light ion number densities during soot
marks) with the results of a detailed chemical- particle formation. This hypothesis will
3
kinetic simulation (dark marks). Ci, Cair, g/cm
be considered below.
The ionic mechanism by Calcote and
Keil is shown in Fig. 8 [54]. The ionic
mechanism starts with the chemiion,
C3H3+, which rapidly grows by the
addition of acetylene, or other small
neutral species. As an ion becomes
larger, its electron recombination
coefficient increases and it is more likely
to be removed from the system by
recombination with the electrons
produced in the initial chemiionization
process. The growing ions either become
charged soot particles or following
Fig. 8. Chemiions to soot [54]
recombination with electrons (produced
in the chemiionization step) yield neutral
species which can continue to grow, as in
the neutral mechanism, to become neutral particles. These particles, neutral or charged,
continue to grow by the addition of acetylene. In a hot flame, or one in which alkali metals
have been added to produce ions, the particles can become charged by thermal ionization
or by diffusive charging. The charge on the particle can play an important role in
determining the rate of coagulation. The latter stages of soot particle growth, by acetylene
addition reactions and by coagulation are common to both the neutral free radical
mechanism and to the ionic mechanism. The differences between the two mechanisms are
44
in the details of the mechanism of nucleation, i.e., in the initial reactions by which the
chemical species change from small molecules to very large molecules, i.e., from a system
characterized as chemical, in which chemical kinetics dominates, to one characterized as
particulate, in which particle dynamics dominates.
The ion mechanism assumes the precursor of soot formation to be C3H3+, the
dominant ion in fuel rich hydrocarbon flames. The source of this ion i s [55, 56]:

СН* + О → СНО+ + е- (8)

followed by several reaction paths to produce C3H3+, e.g.:

СНO+ + СН2 → СН3+ + CO (9)

СН3+ + С2Н2 → С3Н3+ + H2 (10)

and
СНO+ + Н2O → Н3O+ + CO (11)

Н3O+ + С3Н2 → С3Н3+ + H2O (12)

There are two isomeric structures of C3H3+, a stable cyclic structure,

cyclopropenyl, and a linear structure, propargyl. Measurements near room temperature
demonstrate that reactions of propargyl are fast, k = 1 x 10-9 cm3 s-1, generally equal to
the Langevin rate, while the rate coefficients for comparable cyclopropenyl reactions
are much smaller [57, 58]. Eyler and associates [59] found the condensation reaction
of propargyl with C2H2 to be slow, k < 5 x 10-12 cm3 s-1, although they confirmed that
its reaction with C4H2 to produce C5Hx, is fast, k = 1.4 x 10-9 cm3 s-1. In contrast, at a
higher pressure (40-100 Pa), Smith and Adams [60] observed rapid reaction of C2H2
wlth both propargyl, k = 1.1 x 10-9 cm2 s-1, and cyclopropenyl, k ≤ 1 x 10-11 cm3 s-1.
Thus the value of the rate coefficient for the reaction of the propargyl ion with
acetylene appears unsettled.
The above rate coefficients were measured near room temperature and the validity
of extrapolating them to high temperatures is not clear. Langevin theory [61], which
does not predict a temperature dependence for ions reacting with non-pol ar
molecules, has been well tested at ambient temperatures and is generally consistent
with experiments [57, 58, 60]. However, Langevin theory accounts only for the rate of
production of an ion-molecule complex and not the fate of the nascent complex; thus
at higher temperatures the dissociation paths may differ from those at room
temperature.
The ionic mechanism posits that the precursor ions react with neutral species, e.g.,
acetylenes to produce larger ions, e.g.:

С3Н3+ + С2Н2 → С5Н5+ + H2 (13)

С3Н3+ + С4Н2 → С7Н5+ (14)

These ions sequentially and rapidly add low molecular weight neutral species, e.g.,
acetylenes, to produce increasingly larger ions [55, 62]. All of the individual ions
involved in the postulated mechanism, up to mass 557, have been observed in sooting
flames [55, 63, 64].
45
 Some electrons produced in Reaction (8) produce negative ions by attaching to
large molecules; these reactions are favored by low temperature and increasing
molecular weight. Ion-electron recombination rate coefficients are about 2 x 10-7 cm3
s-l for small ions at flame temperatures, and are about two orders of magnitude smaller
for ion-ion recombination. The ion-electron recombination rate coefficients increase
with increasing molecular weight so that as ions grow larger, positive ions are
recombined more rapidly to form neutral incipient soot particles or precursors of
neutral species.
Balthasar et. al [65] assumed that the ionization of primary neutral soot particles
due to thermoemission is responsible for the appearance of charged clusters. A third
mechanism of the formation of charged particles was suggested in [66] and is based on
the attachment of ions and electrons to neutral particles. As was demonstrated, this
mechanism can explain the appearance of both positively and negatively charged
particles in a flame. It should be noted that later Maricq hypothesized that such an
attachment can play an important role in the formation of charged and neutral soot
particles in hydrocarbon flames [67]. Now, we will considered this mechanism.
Kinetic model, which al-
lows to determine the dy-
namics of the formation of
polydisperse ensemble of
charged particles from the
birth of primary clusters
should include both the ki-
netics of formation of heavy
molecular components – the
building material for the
formation of these clusters.
Also, the ions and electrons
in the gas phase processes
and the formation of these
clusters, their growth
surface, interaction with ions
and electrons, and coagu-
lation neural charged par-
ticles and the ability of
electron emission particles at
high temperatures. The sche-
matic diagram of the forma-
tion of a polydisperse en-
Fig.9. The schematic diagram of the formation of a polydisperse semble of charged and
ensemble of charged soot particles by combustion of hydrocarbon neutral particles in hydro-
fuels [68] carbon/air flames is shown
in Fig. 9 [68].
As is believed, soot particles form from polyyne molecules and polyaromatic
hydrocarbons (PAH) [6]. Pyrene molecules (with the Lennard-Jones diameter dp =
0.724 nm), which are composed of four aromatic rings, are considered as the main
soot precursors. The primary clusters with d = 2 nm, which serve as the inception
particles, are constructed from PAH via PAH-PAH coalescence. In sooting flames the
concentration of pyrene molecules can be as large as 1012-1014 cm-3 and the
concentration of primary clusters attains 1011-1012 cm-3.
46
 As was pointed out above, there exists a large number of ions and electrons in
flames. Particles are polarized in the ion or the electron electric field. The appearance
of image charge onto particles significantly enhances the ion-particle interaction.
Because the ions produced in hydrocarbon flames are singly charged, the potential of
the ion-particle interaction may be expressed in the form

( )= − (15)
(

where the parameter P is the product of the ion and particle charges, r is the distance
between the ion and the soot particle, e is the elementary charge and a is the radius
of the particle. The first term in equation (15) describes the conventional Coulomb
potential of a point charge on the particle, whereas the second is the potential of the
image charge. The fundamental difference between potential (15) and the Coulomb
potential is that an ion can be captured by the particle at the distance r = rc > a. This
distance depends on P and is usually called the ion (electron) capture radius.
Further, let us assume that all ions are singly charged. Once an ion approaches the
particle more closely than rc, it moves along a convoluting spiral trajectory. Note
that, in the case of purely Coulomb interaction, the ion trajectory in the field of the
particle is hyperbolic and the distance of the closest approach is equal to the particle
radius [69].
Despite some simplifications concerning the neglect of the surface growth
mechanism by reactions with gaseous species and mechanism of the surface oxidation
the developed model of charged soot particle formation makes it possible to describe
the particle charge distribution for the C2H4/air atmospheric sooting flame detected in
[67]. Fig. 10 shows the predicted and the measured particle fractions with different
charges for various time instants (t = 0 is related to the flame front and t = 100 ms
corresponds to the 34 mm distance downstream the flame front) for small particles
with d = 5 nm, for middle size particles with d = 15 nm and for larger particles with d
= 25 nm (the experimental data on charge distribution were obtained only for particles
with d ≥ 13 nm). As seen, the calculations fit satisfactorily with the experiment [67]
and exhibit the symmetrical distribution of particle charge centered at q = 0. This
means that the main mechanisms of charged particle formation in flames are the
charging of clusters and particles due to attachment of ions (electrons) and coagulation
of charged-charged and charged-neutral particles. From the plots shown in figure 10 it
follows that for small particles (d ≤ 15 nm) there does not exist the characteristic
Boltzmann charge distribution. Conversely, for larger size particles (d > 15 nm) the
charge distribution is close to the Boltzmann one, especially at large time instants or at
large distances (≥ 30 mm)from the flame front. Based on the good consistency in the
predicted and the measured charge distributions for particles it may be concluded that
the mechanisms of the surface growth and the surface oxidation by reactions with
environment molecules do not play a significant role in the formation of relatively
small size (d ≤ 25 nm) charged particles in flames. This may be caused by the greater
rate of the coagulation of charged-charged and charged-neutral particles as compared
with the rate of conventional coagulation of neutral particles. For the ensemble of
charged particles the rate of surface growth for relatively small particles is noticeably
smaller than the coagulation rate. Therefore, the coagulation process is mostly
responsible for the size evolution of such particles. Certainly, this thesis must be
confirmed (or not) on the basis of an accurate simulation using a more complete model

47
of soot particle formation including additionally the mechanisms of surface growth
and oxidation both for charged and for neutral particles.

Fig.10. Predicted and measured normalized charge

distribution of 5 (a), 15 (b) and 25 nm (c) soot
particles for different time instants. Filled symbols
represent the experimental data [67]. The dashed
line shows the Boltzmann distribution at the
calculated flame temperature T = 1850 K. For
particles with d = 15 nm the experimental data are
shown for t = 100 ms and for particles with d = 25
nm the experimental data are related to t = 30 ms.
The predictions were done without taking into
account thermoelectron emission [69]

It is believed that surface growth and surface oxidation should come into play for
larger size particles. As was demonstrated in [11] large particles with d ≥ 50 nm can
accumulate a significant charge (up to 30-40 elementary charges). The existence of
such a large charge onto particles must enhance the rate of surface growth
significantly and, therefore, the process of surface growth should affect the formation
of large charged particles considerably. Moreover, in order to properly describe the
evolution of size distribution of the whole ensemble of particles in sooting flames and
inside the combustor it is needed to take into account both the surface growth and the
surface oxidation processes. Besides these processes it seems very important to extend
the model toward the coagulation of non-spherical charged particles. Fig. 11
summarizes the processes involved in the kinetic model and shows the temporal
method of charged particle formation.
А common disadvantage of the ionic mechanism of soot formation is the fact that
the ratio C/O in the molecules of the precursor of soot nucleation from the very
beginning of the process increases steadily, while up to a certain point in time is the
ratio of the data mechanisms close to unit [70, 71].
Despite all of the above mechanisms, until now a comprehensive analysis of
charged carbon cluster and soot particle formation, including the kinetics for gaseous
neutral species, ions and electrons, as well as particulate kinetics and interaction of
gaseous and condensed plasma compounds, has not been conducted.

48
 3.4. Detailed simulation of the formation of PAHs and soot

It is believed that polycyclic aromatic hydrocarbons are the key intermediate

compounds in soot formation [5, 72, 73]. At the present time, the following scheme of
formation and evolution of soot is widely used:
1. Nucleation of particles. PAHs of large size are formed mainly in the process of
successive chemical reactions of the radicals of the small-sized PAHs with acetylene
and of the PAHs with the PAH radicals. Certain PAHs interact with each other, while
the other PCAHs continue to grow, with the result that a particle is formed.
2. Growth of the surface of particles. The reactions occurring on the surface of the
growing particles substantially in-
crease the carbon mass. It is sug-
gested that these reactions are simi-
lar to the reactions of PAHs [5, 72,
73], activated by the detachment of
hydrogen, despite the fact that the
growth of particles competes with
their oxidation with the participation
of O2, OH, and O. It is conceivable
that acetylene and PAHs are particle-
growth reactants, and their con-
tribution to this process is approxi-
mately determined by concrete con-
ditions [74–77]. The decrease in the
rate of growth of particles with
increase in their size corresponds to
the decrease in the experimentally
determined ratio H/C [9, 78–80].
3. Coagulation of particles. Then
the growing particles increase due to
their collisions. Initially, the col- Fig. 11. The principal scheme of processes resulting
in the formation of neutral and charged particles
liding particles coalescence comp- involved in the model
letely with formation of new spheri-
cal structures, while at a later time they agglomerate into fractal clusters representing
chain-like structures [5].
In [81], a detailed kinetic model describing the formation of soot and the
consumption of PAHs and soot in the combustion of fuel-rich mixtures was proposed.
In accordance with this model, the indicated processes include 1102 gas-phase
reactions and 5202 reactions of growth of a soot particle with the participation of 295
PAH particles. With the use of the sectional approach, the large PCAH and carbon
particles with a diameter of up to 70 nm were rated in the BIN classes, including
substances with certain mass numbers. The amounts of carbon and hydrogen atoms,
rated in concrete BIN classes, in accordance with their average masses, are determined
by the H/C ratio, which decreases with increase in the particle size. The model was
successfully tested in a fuel-rich premixed benzene–oxygen–argon flame (H/C = 2.4,
CAr = 10 %, u = 25 cm/s, P = 5.33 kPa). The results of a preliminary calculation with
the use of this model were compared with published experimental data, including

49
profiles of the mole fractions of individual PAHs and the concentrations of soot
particles. It was established that the reactions of the PAH-radicals with the PAHs and
between the PAHs radicals are the main pathways to the soot nucleation. The surface
growth of the soot particles provides ≈ 75 % of their final mass, and the reaction of
acetylene with the PAH radicals is the main pathway of development of the surface-
growth reaction – it leads to a decrease in the concentration of PAHs in the afterflow
zone. The coagulation of the BIN particles and their radicals leads to the formation of
particles of increasing size, while the oxidation of particles by the OH radicals counts
very little in the decrease in their size.
To obtain more complete quantitative information on the processes being
considered, the detailed chains of the reaction, in which PAHs are formed and their
concentration decreases [82, 83], were extended to the formation of particles [35, 84].
Many models are based on the assumption that PAHs of certain sizes are soot nuclei.
At the same time, it was conceived that a rapid polymerization of acetylene leads to an
increase in the carbon structure [3]. Two main principles based on the moments
method [35] and the sectional approach [84, 85] were used. A union of particles in the
section reaction is described with the use of the BIN classes. A distinguishing feature
of the section approach is that the directly-related gas-phase and colloid-chemical
systems are described similarly in the common form as A+B = C+D.

Table 1
BIN molecular classes enclosing large pahs and soot particles

BIN Mass, a.m.u СxНy d, нм H/C

1 201–400 С24Н12 0.85 0.500
2 401–800 С48Н24 1.07 0.500
3 801–1600 С96Н48 1.34 0.500
4 1601–3200 С193Н84 1.69 0.435
5 3201–6400 С388Н144 2.13 0.371
6 6401–12800 С778Н264 2.68 0.339
7 12801–25600 С1560Н480 3.37 0.308
8 26601–51200 С3124Н912 4.24 0.292
9 51201–102400 С6256Н1728 5.35 0.276
10 102401–204800 С12528Н3264 6.73 0.261
11 204801–409600 С25088Н6144 8.48 0.245
12 409601–819200 С50240Н11520 10.69 0.229
13 819201–1638400 С100608Н21504 13.46 0.214
14 1638401–3276800 С201472Н39936 16.96 0.198
15 3276801–6553600 С403456Н73728 21.36 0.183
16 6553601–13107200 С807936Н135168 26.91 0.167
17 13107201–26214400 С1617920Н245760 33.91 0.152
18 26241401–52428800 С3239936Н442368 42.72 0.137
19 52428801–104857600 С6483968Н835584 53.83 0.129
20 104857601–209715200 С12972032Н1622016 67.82 0.125

The numbers of equal-mass particles and structural isomers increases rapidly with
increase in their molecular mass. Therefore, the BIN classes of very large PAHs and
50
particles with definite masses were determined. A BIN class of these substances is
determined by their average molecular mass and the number of carbon and hydrogen
atoms in them. The characteristics of BIN classes are presented in Table 1. The
diameter d of the particles was determined on the assumption that they are spherical
and their concentration is 1.8 g/cm3 [86]. It is suggested that the transformation of
gaseous particles into solid ones in a flame is determined by both the temperature of
the flame and its pressure. The results of detection of the compounds, including as
large as 160 carbon atoms in the extract of a flame-generating condensed material
[87], allow the conclusion that BIN classes with number 5 or larger include
"particles," while BIN classes with numbers 1-4 include "large PAHs." This
representation agrees with the definition of particles with a molecular mass of ≈ 2000
a.m.u and a diameter of ≈ 1.5 nm as incipient soot particles [74, 88] and the
consideration of soot as nonextractable particles with a gravimetrically determined
concentration [89]. It has been established that the decrease in the H/C ratio and the
resulting increase in the curvature of the particle surface are due to the presence of
five-membered rings and correlate, in many cases, with the formation of fullerene
structures [90, 91]. In [81], the maximum particle diameter, measured in a low-
pressure one-dimensional homogeneous mixture in an acetylene flame [20, 92], was
related to the H/C ratio determined under indentical conditions [9].

3.5. Mechanism of soot formation in rich flames

The flame front structure of soot formation flames can be considered in

close connection with mechanism of fuel conversion. A scheme that suggested
by Bockhorn [7] in 1994 is known, where polycyclic aromatic hydrocarbon
(PAH) are precursors of soot particles. Since then, in rich fuel flames the
nanosized particles such as fullerenes and graphenes are found.
The fullerenes formation is generally observed at pressures below 60 Torr,
because for fullerenes formation is necessary an adherence of steric spatial
factor, but in flames at atmospheric pressure this factor is prevented by triple
collisions. Therefore, the pressure coordinate can be introduced to the general
scheme for soot formation. At low pressure the formation of fullerenes from
polycyclic aromatic hydrocarbons (PAHs) are occurred, but with pressure
increase the PAHs are coagulates to soot particles. From PAHs follows the
graphene formation as intermediate product between PAH and soot particles,
which is confirmed by the formation of multi-layered graphene films at
atmospheric pressure and single-layered at the pressures below 60 Torr. On the
basis of the data on synthesis of fullerenes, carbon nanotubes, superhydrophobic
soot and graphene in the flame it is possible to modify the general scheme
proposed by H. Bockhorn [7] for rich fuel flames, namely to make a pressure-
coordinate, which allows the formation of fullerenes at low pressures, and soot
at high pressures. In addition the scheme was completed by graphene
formation as an intermediate product stage of graphene formation (Fig. 12) [93].

51
 Fig. 12. Modified scheme for soot, fullerenes and graphene formation process in flame [93]

CONTROL QUESTIONS

1. How would you describe polyyne model of soot formation?

2. Can you give detailed modeling of soot formation and PАH?
3. What is the idea of polyaromatic model of soot formation?
4. How formation of the first aromatic ring is conducted?
5. Can you give the description of ion model of soot formation?
6. How would you describe some observations beginning of the formation of soot?

52
 REFERENCES

1. H. Wang. Formation of nascent soot and other condensed-phase materials in flames // Proceedings of
the Combustion Institute, 2011. – Vol. 33. – P. 41–67.
2. P.A. Tesner. Kinetics of pyrolytic carbon formation // Chemistry and physics of carbon. 1984. V. 19.
P. 65-161
3. A.V. Krestinin. Polyyne Model of Soot Formation Process // 27th Symp. (Int) on Combustion. The
Combustion Institute. Pittsburg. 1998. P.1557-1563.
4. J.B. Howard, A. L. Lafleur and other, Carbon, 30, 1183-1201 (1992)
5. M. Frenklach. Reaction mechanism of soot formation in flames // Phys. Chem. Chem. Phys. 2002.
V. 4. P. 2028-2037
6. H. Bockhorn. Soot Formation in Combustion: Round Table Discussion / Ed. Heidelberg Springer
Verlag, 1994.
7. A. Violi, A., D’Anna, A., D’Alessio A., “A Kinetic Model of Particulate Carbon Formation in Rich
Premixed Flames of Ethylene and Benzene,” paper 47.2, in Fifth International Conference on
Technologies and Combustion fo r a Clean Environment, Lisbon, Portugal, in press (2000).
8. D’Anna, A., Violi, A. Proc. Combust. Inst. 27:425 433 (1998).
9. A. Garo, J. Lahaye, G. Prado. Mechanism of formation and destruction of soot particles in a laminar
methane-air diffusion flame // Twenty-first (International) on Combustion/The Combustion Institute,
1986. P. 1023-1031
10. C.S. McEnally, L.D. Pfefferle, B. Atakan, K. Kohse-Hoinghaus. Studies of aromatic hydrocarbon
formation mechanisms in flames: progress towards closing the fuel gap // Prog. Energy and
Combust. Sci. 2006. V. 32. P. 247–294
11. K.H. Homann, H.G. Wagner. Some aspects of soot formation // Carbon and Breach Publishers. 1996.
P. 151–184
12. K.H. Homann, H.G. Wagner. Some new aspects of the mechanism of carbon formation in premixed
flames // 11th Int. Symp. Combust., The Combustion Inst., Pittsburg. 1967. P. 371–379
13. A.M. Savel’ev, A.M. Starik. Features of the interaction of ions and electrons with nanopartilces in
the plasma formed in the combustion of a hydrocarbon fuel // Zh. Tekh. Fiz. 2006. V. 76. Issue 4.
P. 53–60
14. A.A. Onischuk, S. Di Stasio, V.V. Karasev, et al. Evolution of structure and charge of soot
aggregates during and after formation in a propane air diffusion flame // J. Aerosol Sci. 2003. V. 34,
№ 4. P. 383-403
15. H.F. Calcote. Mechanisms of soot nucleation in flames // A critical review, Combust. Flame, 1981.
V. 42, № 2. P. 215–242
16. A.V. Krestinin. Kineticheskaya model sazheobrazovaniya iz acetilena v rasbavlennih smesiyah pri
temperature vishe 1600 К // Chemicheskaya fizika. 1987. V. 6, № 3. P. 342-349 [in Russian]
17. M.P.E. Berthelot. Ann. Chim. Phys. 1866. V. 9. P. 471
18. G. Porter. 4th Symposium (Int.) on Combustion, The Williams and Wilkins Co., Baltimore. 1953.
P. 248
19. P.A. Tesner, I.M. Gutor, V.G. Knorre, N.V. Lutsik, M.P. Mulyava. Khim. Fizika. 1983. V. 2. P. 110
20. A.V. Krestinin. Detailed Modeling of Soot Formation in the Hydrocarbon Pyrolysis Process //
Proceedings of Joint Meeting of the British, German and French Sections of the Combustion
Institute. Nancy, France. 18-21 May, 1999. P. 109-111
21. Bonne, U., Homann, K.H., and Wagner, H.Gg., 10th Symposium (Int.) on Combustion, The
Combustion Institute, Pittsburg. 1965. P. 503-512
22. A.V. Krestinin. In: Advanced Computation & Analysis of Combustion. (G.D. Roy, S.M. Frolov,
P. Givi, Eds.), ENAS Publishers, Moscow. 1997. P. 38
23. A.V. Krestinin. // Chem. Phys. Reports, 1994. V.13. P. 191
24. A.V. Krestinin. // Chem. Phys. Reports, 1998. V. 17 P. 1441
25. A.V. Krestinin. Detailed Modeling of Soot Formation in Hydrocarbon Pyrolysis // Combust. Flame.
2000. V. 120. P. 513-524
26. M. Frenklach, D.W. Clary, W.C.Jr. Gardiner, S.E. Stein. 20th Symposium (Int.) on Combustion, The
Combustion Institute, Pittsburg, 1984. P. 887
27. J.H. Kiefer, W.A. Von Drasek, Int. J. Chem. Kinet. 1990. V. 22. P. 747
28. R.D. Smith, A.L. Johnson. Combust. Flame. 1983. V. 51 P. 1
29. A.N. Hayhurst, H.R.N. Jones. J.Chem.Soc., Faraday Trans. 1987. V. 2. P. 83
30. F.W. Lam, J.B. Howard, J.P. Longwell. 22nd Symposium (Int.) on Combustion, The Combustion
Institute, Pittsburg. 1988. P. 323.
31. J.A. Cole, J.D. Bittner, J.P. Longwell, J.B. Howard. Combust. Flame. 1984. V. 56. P. 51
32. I. Dale. In: Chemistry of acetylenes. (Viehe,H.G., Ed.), Marcel Dekker, N.Y. 1969. P. 11

53
33. A.V. Krestinin, M.B. Kislov, A.V. Raevskii, et al. On the mechanism of formation of soot particles //
Kinet. Katal., 2000. V. 1. P. 102-111
34. Z.A. Mansurov. Formation of soot from polycyclic aromatic hydrocarbons as well as fullerenes and
carbon nanotubes in the combustion of hydrocarbon // Journal of Engineering Physics and
Thermophysics. 2011. V. 84, №1. P. 125-159
35. M. Frenklach, H. Wang, Detailed mechanism and modeling of soot particle formation soot formation
in combustion, mechanism and models, in: H. Bockhorn (Ed.), Soot Formation in Combustion,
Springer Series in Chemical Physics, Springer Verlag, Berlin. 1994. V. 59. P. 165
36. M. Frenklach, H. Wang. Detailed modeling of soot particle nucleation and growth // Proceedings
Combust. Institute. 1991. V. 23. P. 1559-1566
37. H. Wang. Formation of nascent soot and other condensed-phase materials in flames // Proceedings of
the Combustion Institute. 2011 V. 33. P. 41-67
38. M. Frenklach, D.W. Clary, W.C. Gardiner, S. E. Stein. // Proc. Combust. Institute. 1985. V. 20.
P. 887
39. H. Wang, M. Frenklach. Combust. Flame. 199. V. 110. P. 173
40. M. Frenklach, J. Warnatz. Combust. Sci. Technol. 1987. V. 51. P. 265
41. S.E. Stein, A. Fahr. High temperature stabilities of hydrocarbons // J. Phys. Chem. 1985. V. 89.
P. 3714-3725
42. H. Wang, M. Frenklach, Combust. Flame. 1997. V. 110. P. 173–221.
43. A.M. Dean, J. Phys. Chem. 89 (1985) 4600–4608.
44. P.R. Westmoreland, A.M. Dean, J.B. Howard, J.P. Longwell. // J. Phys. Chem. 1989. V. 93.
P. 8171–8180
45. H. Wang, M. Frenklach. // J. Phys. Chem. 1993. V. 97. P. 3867–3874.
46. H. Wang, M. Frenklach . // J. Phys. Chem. 1994. V. 98. P. 11465–11489.
47. J. Appel, H. Bockhorn, M. Frenklach. // Combust. Flame. 2000. V. 121. P. 122–136.
48. Z.A. Mansurov, N.G. Prikhodko, A.V. Saveliyev. Obrazovanie PCAU, fullerenov, uglerodnih
nanotrubok, I sazhi v processah goreniya. Almaty: Kazhak university, 2012. P. 383
49. J. Warnatz, U. Maas, R. Dibble. Combustion. Physical and Chemical Fundamentals, Modelling and
Simulation, Experiments, Pollutant Formation. Fizmatlit: Moscow (2003) [Russian]
50. N.A. Slavinskaya, P.A. Frank modeling study of aromatic soot precursor formation in laminar
methane and ethane flames // J. Combust. Flame. 2009. V. 156. P. 1705–1722
51. J.A. Miller, C.E. Melius. Kinetics and thermodynamic issues of aromatic compounds in flames of
aliphatic fuels, Combust. Flame, 1992. V. 91. P. 21–39
52. N.M. Marinov, W.J. Pitz, C.K. Westbrook. The formation of aromatics and PAH’s in laminar flames
// Joint meeting of the British, German and French sections. 1999. P. 7.
53. P.A. Vlasov, J. Warnatz. Kinetic modeling of the soot formation in the pyrolysis of various aliphatic
and aromatic hydrocarbons in shock tubes // Khim. Fiz. 2004. V. 23, № 10. P. 39-46
54. H.F. Calcote, D.G. Keil. The role of ions in soot formation // Pure and App. Chern. 1990. V. 62,
№ 5. P. 815-824
55. H.F. Calcote. // Combust. Flame. 1981. V. 42. P. 215
56. A.N. Eraslan, R.C. Brown. // Combust. Flame. 1988. V. 74. P. 191
57. J.R. Eyler. The Chemistrv of Combustion Processes, T.M. Sloane, Ed.; Advances in Chemistry,
Series American Chemical Society, Washington, DC. 1984. V. 249. P. 49
58. K.C. Smyth, S.C. Lias, and P. Ausloos, Combust. Sci. Techn. 1982. V. 28. P. 147
59. F. Ozturk, G. Baykut, M. Aoini, J.R. Eyler. // J. Phys. Chem. 1987. V. 91. P. 4360
60. D. Smith, N.G. Adams. // Int. J. Mass Soect. Ion Proc. 1987. V. 76. P. 307
61. R. Patrick, D.M. Golden. // J. Phys. Chem. 1985. V. 82. P. 75
62. H.F. Calcote, D.B. Olson, D.G. Keil. // EnerQv & Fuels. 1988. V. 2. P. 494
63. D.G. Keil, R.J. Gill, D.B. Olson, H.F. Calcote. Twentieth Symposium (International on Combustion,
The Combustion Institute, Pittsburgh). 1985. P. 1129
64. P. Michaud, J.L. Delfau, A. Barassin. // Eighteenth Symposium (International on Combustion, The
Combustion Institute, Pittsburgh). 1981. P. 443
65. M. Balthasar, F. Mauss, H. Wang. // Combust. Flame. 2002. V. 129. P. 204-216
66. A.M. Savelev, A.M. Starik, N.S. Titova, O.N. Favorskii. Dokl. Phys. 2004. V. 49. P. 441-446
67. M. Maricq. // Combust. Flame. 2006. V. 144. P. 730-743
68. А.М. Saveliyev, А.М. Starik, N.S. Titova, О.N. Favorskii. Osobennosti formirovaniya zaryazhennih
I neitralnih nanochastic v uglevodorodno-vozdushnih plamenah // DAN. 2008. V. 420, № 6. P. 752-
757
69. А.М. Starik, А.М. Saveliyev, N.S. Titova. Formation of charged nanoparticlesin hydrocarbon
flames: principal mechanisms // Plasma Sources Sci. Technol. 2008. V. 17. P. 1-8
70. P.A. Tesner. Obrazovaniya sazhi pri goreniya // Fizika goreniya i vzriva. 1979. V. 15, № 2. P. 3-14

54
71. Z.A. Mansurov. Cool sooting flames of hydrocarbons // Journal of Thermal Science. 2001. V. 10,
№ 3. P. 269-280
72. H. Richter, J.B. Howard. Formation of polycyclic aromatic hydrocarbons and their growth to soot – a
review of chemical reaction pathways // Prog. Energy Combust. Sci. 2000. V. 26. P. 565-608
73. J.B. Howard. Carbon addition and oxidation reactions // Proc. Combust. Inst. 1990. V. 23. P. 1107–
1127
74. S.J. Harris, A.M. Weiner. A picture of soot particle inception // Proc. Combust. Inst. 1988. V. 22.
P. 333–342
75. T.G. Benish, A.L. Lafleur, K. Taghizadeh, J.B. Howard. C2H2 and PAH as soot growth reactants in
premixed C2H4–air flames // Proc. Combust. Inst. 1996. V. 26. P. 2119–2326
76. S. Macadam, J.M. Beer, A.B. Hoffmann. Soot surface growth by polycyclic aromatic hydrocarbon
and acetylene addition // Proc. Combust. Inst. 1996. P. 2295–2302
77. A. Kazakov, M. Frenklach. On the relative contribution of acetylene and aromatics to soot particle
surface growth // Combust. Flame 1998. V. 112. P. 270–274
78. A. Ciajolo, R. Barbella, A. Tregrossi, L. Bonfanti. Spectroscopic and compositional signatures of
PAH loaded mixtures in the soot inception region of a premixed ethylene flame // Proc. Combust.
Inst. 1998. V. 27. P. 1481-1487
79. S. Irle, G. Zheng, Z. Wang, K. Morokuma. The C60 formation puzzle "solved": QM/MD simulations
reveal the shrinking hot giant road of the dynamic fullerene self-assembly mechanism // J. Phys.
Chem. 2006. V. 110, № 30. P. 14531–14545
80. L.G. Blevins, R.A. Fletcher, B.A. Benner, et al. The existence of young soot in the exhaust of inverse
diffusion flames // Proc. Combust. Inst. 2002. V. 29. P. 2325–2333
81. H. Richter, S. Granata, W.H. Green, J.B. Howard. Detailed modeling of PAH and soot formation in
laminar preliminary mixture benzene/oxygen/argon at low pressure flame // Proc. Combust. Inst.
2004.
V. 30. P. 1397–1405
82. H. Wang and M. Frenklach, A detailed kinetic modeling study of aromatics formation in laminar
premixed acetylene and ethylene flames // Combust. Flame. 2004. V. 110. P. 173-221
83. H. Richter, W.J. Grieco, J.B. Howard. Formation mechanism of polycyclic aromatic hydrocarbons
and fullerenes in premixed benzene flames // Combust. Flame. 1999. V. 119. P. 1-22
84. C.J. Pope, J.B. Howard. Simultaneous particle and molecule modeling: An approach for combining
sectional aerosol equations and elementary gas-phase reactions //Aerosol Sci. Technol. 1997. V. 27.
P. 73-94
85. R.J. Hall, M.D. Smooke, M.B. Colket (F. L. Dryer and R. F. Sawyer Eds.) // Physical and Chemical
Aspects of Combustion. A Tribute to Irvin Glassman, Combust. Sci. Technol. Book Series, Gordon
and Breach Sci. Publ., Amsterdam. 1997. V. 4. P. 189
86. J. B. Donnet, R.C. Bansal, M.J. Wang. // Carbon Black: Science and Technology, 2nd ed., Dekker,
New York. 1993
87. A.L. Lafleur, K. Taghizadeh, J.B. Howard, et al. Characterization of flame-generated C10 to C160
polycyclic aromatic hydrocarbons by atmospheric pressure chemical ionization mass spectrometry
with liquid introduction via heated nebulizer interface // J. Amer. Soc. Mass Spectrom. 1996. V. 7.
P. 276-286
88. J.T. McKinnon, J.B. Howard. The roles of PAH and acetylene in soot nucleation and growth // Proc.
Combust. Inst. 1992. V. 24. P. 965-971
89. W.J. Grieco, A.L. Lafleur, K.C. Swallow, et al. Fullerences and PAH in low-pressure premixed
benzene/oxygen flames // 27th Int. Symp. Combust. Inst. 1998. P. 1669-1675
90. A.L. Lafleur, J.B. Howard, K. Taghizadeh, et al. Identification of C30H10 dicy-dopentapyrenes in
flames: correlation with corannulene and fullerene formation // J. Phys. Chem., V. 100. P. 17421-
17428
91. A. Goel, P. Hebgen, J.B. Van der Sande, J.B. Howard. Combustion synthesis of fullerenes and
fullerenic nanostructures // Carbon. 2002. V. 40. P. 177-182
92. B.L. Wersborg, J.B. Howard, G.C. Williams. Physical mechanisms in carbon formation in flames //
14th Int. Symp. Combust. The Combustion Inst., Pittsburgh. 1973 P. 929
93. Z.A. Mansurov. Soot and nanomaterials synthesis in the flame // Journal of Materials Science and
Chemical Engineering. 2014. V. 2. P.1-6

55
 Chapter 4

CONTROL OF THE FORMATION OF PCAHS

AND SOOT IN A FLAME

The formation of soot in a flame is a complex multistage process, where the

reactions of formation and transformation of PCAHs play a great role. Since the
reactions of soot formation are rapid and complex, of importance is control of
processes leading to the formation of PCAHs and soot.

4.1. Influence of chemical additives on the formation of polycyclic

aromatic hydrocarbons

We now consider some experimental data on the influence of chemical additives

on soot formation. Metal additives containing, e.g., iron and manganese, are usually
considered as good inhibitors of soot in the combustion processes [1–3]. On the other
hand, experimental investigations have shown that certain metal-containing additives
increase the PCAH concentration and the yield of soot in hydrocarbon flames [4].
In Table 1, data on the influence of some chemical compounds on the formation of
PCAHs and soot, taken from a number of ublications, are presented. Feitelberg et al.
[5] have established that, under different conditions, additions of ferrocene and
manganese increase the soot formation in premixed ethylene flames, even though they
do not change the concentrations of the C1–C4 particles. Assuming that the iron on the
surface of the growing soot particles catalyzes the formation of PCAHs from
acetylene, Hahn and Charalampopoulos [6] have established that soot formation
increases in the case where iron pentacarbonyl is added to a premixed propane flame.
They argue that the volume soot fraction increases due to the action of the iron-oxide
ring on the initial area of the soot surface. These authors also show that, in the
combustion chambers used in practice, iron additives can catalyze the oxidation of
carbon through the formation of Fe2O3 (burning-out of soot).
Bulewicz [7] and Salooja [8] have shown that a change in the concentration of
additives in diffusion flames can change the character of their action, beginning with
suppression of smoke and ending with increasing its effect. Salooja [8] has established
that the introduction of metal additives into different regions of diffusion flames can
give rise to different effects. The lowest height of introduction of an additive into a
flame corresponds to the lowest yield of soot [9]. In the case of introduction of
additives into the higher-lying regions of the flame, their action changes from
suppression of the soot formation to its enhancement. The available experimental data
show that metal additives introduced into the premixed diffusion flames change the
soot yield (Table 1).
The influence of HBr and Fe(CO)5 on the concentration of PCAH molecules and
their precursors in an acceptor-radical flame was determined on the basis of model
experiments for premixed flames and flames in a mixing reactor [10]. The results
obtained suggest a kinetic model of soot formation, including reactions determining
56
the inhibition of the flame by iron pentacarbonyl and HBr. Near the stoichiometric
conditions, the influence of the radical-acceptor is very large. An increase in the
equivalent ratio decreases the relative influence of these impurities. The influence of
the additions of gas-phase acceptors decreases with increase in the equivalent ratio.
The kinetic model shows that gas-phase additives increase the yield of PCAHs.
The results of the present work support the earlier supposition that the experimentally
determined decrease in the yield of microparticles is probably due to the
heterogeneous reactions of the iron oxides, caused by the increase in the rate of
oxidation of the soot formed. A simulation has shown that the contributions of the
propargyl compounds and the reactions of benzene formation to the process being
considered are determined by the equivalent ratio. The first of them dominates in a
rich system, where the propargyl compound corresponds to benzene formation, while
the second is more important at a small equivalent ratio.

Table 1

Influence of metal-containing agents on the formation of pcahs and soot

4.2. Influence of an electric field

The influence of an electric field on the combustion of hydrocarbons is of great

interest for chemical synthesis and control of soot formation for environmental
protection. It has been established [11] that even a small-strength electric field
substantially influences the kinetics of a combustion. The process of soot formation in
a hydrocarbon flame is determined to a large extent by the circumstance that the soot

57
particle acquires a positive charge in the flame [12-15]. This process can be controlled
with the use of external electric fields [16].
The action of an electric field on combustion was investigated by many
researchers from the standpoint of its influence on the structure of soot particles and
their yield. It has been established that in the case where a constant electric field is
applied to an atmospheric pressure diffusion flame, the yield of soot particles
decreases markedly and, at an electric-field strength of 200 kV ⁄ m, the decrease in the
soot yield comprises 90 % at a negative polarity and 70 % at a positive polarity [17].
In this case, the shape of the flame changes substantially with increase in the voltage
applied and in the polarity. At a negative polarity, the temperature in the upper part of
a flame is approximately 500 oC higher than that in the absence of an electric field and
comprises 1800 oC, which was not detected at a positive polarity. It was inferred that
the decrease in the yield of soot under the action of an electric field is due to the
oxidation of the soot particles.
A detailed analysis of the sizes of soot particles, carried out with the use of
electron microphotography, has shown that particles having a diameter of 50 nm
without an electric field decreased to a diameter 10 nm under the action of an electric
field of several kilovolts [17]. In this case, as was shown in [18], soot particles become
more homogeneous as the applied voltage increases. It is known that an electric field
acts on the nucleation of soot particles and their growth and that the soot particles in a
flame mainly acquire a positive charge [17, 19]. In accordance with the ion theory
[20], ions of the type CnHn+ are active centers of growth of soot particles, which also
substantially influences the kinetics of the soot formation under the action of an
electric field. At an electric-field strength of 200 kV ⁄ m and higher, the Joule heat
released becomes comparable to the heat released as a result of the chemical reaction,
which increases the temperature of the flame to 1800 oC and higher [17]. In [21], the
influence of the parameters of a constant electric field – the voltage applied and the
polarity – on the shape of a flame as well as on the yield and structural characteristics
of the soot particles, fullerenes, and PCAHs formed was investigated.

1) flame front; 2) central glowing part; 3) burner; 4) upper needle electrode; 5) glow-discharge
filament

Fig. 1. The Shape and dimensions of a flame without an electric field (a) and with electric fields of
negative (c) and positive (b) polarity (U = 10 kV)

The investigations were carried out in the process of combustion of a benzene

vapor premixed with oxygen (at an atomic ratio C ⁄ O = 1.0) and an argon additive
(10% of the mixture volume) on a plane burner at a pressure P = 40 torr.
58
 A needle - plane electrode system with an interelectrode distance H = 18 cm,
positioned in a burner apparatus made from a quartz glass, was used. The matrix of the
burner served as a plane electrode. A high constant voltage U of negative ("minus" at
the needle electrode) or positive ("plus" at the needle electrode) polarity of value 0.5–
20 kV was applied to the needle electrode positioned in the upper part of the burner
apparatus. The exposure of the flame to the longitudinal electric field gave rise to a
negative or positive electric charge influencing the flame.
The action of an electric field on the flame was observed visually, beginning with
U ≥ 1.35 kV for the negative polarity and with U ≥ 3.0 kV for the positive one. This
effect manifested itself as compression, vibration, stretching, and flattening of the
flame, its transformation into a tulip-like shape, and other changes. In the absence of
an electric field, the flame was stable with no vibrations and had a cylindrical shape
(identical to the burner shape) with a diameter D = 2.5 cm; the height of the glow zone
was L = 4.5 cm and the distance between it and the burner was δ = 0.5 cm.
In the case where a voltage of V = 10 kV or higher was applied to the flame, the
upper part of the flame front took the shape of a wave (Fig. 1), following the wave-
like action of the glowing winding cord of the glow discharge on it; however, the
distance between the flame front and the burner remained unchanged. In the case
where a positive-polarity electric field was applied to the flame, beginning with U ≥
10 kV, the action of the electric field on the flame increased to the point where it died
out, i.e., the positive ions and the neutral particles led by them were carried out from
the flame front to the preflame zone and then to the negative electrode — the burner,
which extinguished the flame. The negative-polarity electric field did not extinguished
the flame, which shows once again that the positive ions are of primary importance in
the organization of the combustion process [16, 19].
An analysis of the microphotographs of soot samples has shown that the soot
obtained is a polydisperse substance and its particles are of near-spherical form. The
arithmetic mean weighted diameter of the soot particles dm as well as their maximum
dmax and minimum dmin diameters are presented in Table 2.
Table 2

Soot-particle size depending on the voltage applied and the electric-field polarity

An electric field is useful in increasing the amount of soot particles due to the
change in the duration of their stay in the growth zone. A positive-polarity electric
field causes a more rapid increase in the sizes of the soot particles. However, a longer
stay of soot particles in the combustion zone and in the zone of action of an electric
field leads to their oxidation.
The soot formation in the combustion of a methane-oxygen mixture in an electric
field was investigated with the use of different-polarity electrodes [18] with a potential
difference changing from 0 to 2.2 kV. It has been established that the soot particles
and the soot yield decrease under the action of an electric field. In the case where the
59
positive electrode is positioned at the input of the second section, the soot particles
formed are charged due to the thermal-electron emission and they, having obtained a
positive charge, are accelerated by the external field (Fig.2). Since, in this case, soot
particles are transported more rapidly through the combustion zone, they have no time
to grow to the sizes
characteristic of the process
in the absence of a field,
which explains the decrease
in the soot mass and in the
sizes of the soot particles.
An opposite-direction
electric field decelerates the
soot particles in the
negativecharge region.
Thus the experimental
results show that the
Fig. 2. Dependence of the soot yield (a) and the sizes of the soot processes occurring in a
particles on the polarity of the electrodes: 1) í; 2) ±; 3) í; the soot-forming flame can be
distance between the electrodes is 90 (1, 2) and 205 mm (3) [148].
U, kV; β, %; r, Å.
controlled with the use of an
external electric field that
can influence the
combustion stability, the soot formation, the chemical synthesis, and the yield of
fullerenes.

CONTROL QUESTIONS

1. How would you describe control of the formation of PCAHs and soot in a flame?
2. How influence of an electric field on the formation of PCAHs and soot?
3. How influence of metal-containing agents on the formation of PCAHs and soot?
4. How formation of the first aromatic ring is conducted?
5. How would you describe some observations beginning of the formation of soot?

REFERENCES

1. B. S. Haynes, H. Junder, and H. G. Wagner, Proc. Combust. Inst., 18, 1365 (1980).
2. J. B. Howard and W. J. Kausch, Soot control by fuel additives, Prog. Energy Combust. Sci., 6, 263–
276 (1980)
3. J. Lahaye, S. Boehm, and P. Ehrburger (H. Bockhorn Ed.), Soot Formation in Combustion, Springer-
Verlag, Berlin–Heidelberg.
4. V. I. Babushok, W. Tsang, and K. L. McNesby, Additive influence on polycyclic aromatic
hydrocarbon formation, Proc. Combust. Inst., 29, 2315 (2002).
5. A. S. Feitelberg, J. P. Longwell, and A. F. Sarofim, Metal enhanced soot and PAH formation,
Combust. Flame, 92, 241 (1993).
6. D. W. Hahn and T. T. Charalampopoulos, The role of iron additives in sooting premixed flames,
Proc. Combust. Inst., 24, 1007 (1992).
7. E. M. Bulewicz, D. G. Evans, and P. J. Padley, Effect of metallic additives on soot formation
processes in flames, Proc. Combust. Inst., 15, 1461 (1974).
8. K. C. Salooja, Combustion control by novel catalytic means, Nature, 240, 350 (1972)
9. R. A. Blevins, B. A. Fletcher, E. B. Benner, and G. W. Steel, The existence of young soot in the
exhaust of inverse diffusion flames, Proc. Combust. Inst., 29, 2325–2333 (2002)
10. D. Tanke, H. G. Wagner, and I. S. Zaslonko, Mechanism of the action of iron-bearing additives on
soot formation behind shock waves, Proc. Combust. Inst., 27, 1597 (1998).

60
11. B. L. Wersborg, A. C. Yeung, and J. B. Howard, Concentration and mass distribution of charged
species in sooting flames, Proc. 15th Int. Symp. Combust., The Combustion Inst., Pittsburg (1975),
pp. 1439–1448
12. R. Y. Mayo and F. G. Weinberg, On the size, charge and number-rate of formation of carbon
particles in flames subjected to electric fields, Proc. Roy. Soc. A, 319, 351 (1970)
13. D. G. Keil, B. J. Gill, D. B. Olson, and H. F. Calcote, Ionization and soot formation in premixed
flames, 20th Symp. (Int.) Combust., The Combustion Institute, Pittsburg (1984), No. 7, 1129–1137.
14. F. Weinberg, Smokes, droplets, flames and electric fields, Far. Soc. Symp., 120–132 (1973).
15. Z. A. Mansurov, A. A. Merkulov, V. T. Popov, B. K. Tuleutaev, and N. S. Almazov, Synthesis of
ultradisperse soot in the combustion of methane in an electric field, in: Abstracts of papers submitted
to the 3rd All-Union Conf. "Chemical Synthesis on the Basis of Monocarbohydrate Molecules,"
Moscow (1991).
16. J. Lawton and F. Weinberg (V. A. Popov Ed.), Electrical Aspects of Combustion [in Russian],
E′nergiya, Moscow (1976).
17. S. Masahiro, A. Toshihiro, and A. Masataka, Control of soot emitted from acetylene diffusion flames
by applying an electric field, Combust. Flame, 119, 356–366 (1999).
18. Z. A. Mansurov, A. A. Merkulov, V. T. Popov, B. K. Tuleutaev, and N. S. Almazov, Formation of
ultradisperse soot in the combustion of methane in an electric field, Khim. Tverd. Topl., No. 3, 83–
86 (1994).
19. E. M. Stepanov and B. G. D’yachkov, Ionization in a Flame and the Electric Field [in Russian],
Metallurgiya, Moscow (1968).
20. P. A. Tesner, Formation of Carbon from Gas-Phase Hydrocarbons [Russian translation], Khimiya,
Moscow (1972)
21. Z. A. Mansurov, N. G. Prikhodko, T. T. Mashan, and B. T. Lesbaev, Study of soot formation at
combustion of hydrocarbons at low pressure, Proc. 7th Asia-Pacific Int. Symp. Combustion and
Energy Utilization, December 14–19, 2004, Hong Kong (2004).

61
 Chapter 5

SYNTHESIS OF FULLERENES IN FLAME

The discovery of fullerenes representing a new form of existence of one of the

most abundant elements on the Earth - carbon - has been recognized as one of the
most significant achievements in 20th century science. Despite the fact that it has long
been known that carbon atoms have a unique ability to bind into complex three-
dimensional branched structures, which forms the basis of the whole subject of
organic chemistry, the possibility of only one carbon forming stable skeleton
molecules was, nevertheless, unexpected.
The fullerenes C60 and C70 were identified in 1985 [1] and were obtained in
macroscopic amounts in 1990 in the process of evaporation of graphite by an arc
discharge. The C60 and C70 fullerene ions were detected in flames in 1987 and
identified by the mass-spectrometric method [2-4]. Howard et al. [5-6] have obtained
large amounts of C60 and C70 in laminar premixed soot forming flames of benzene and
oxygen at low pressures. Unlike the an evaporation of graphite, in the fullerenes
formed in flames the ratio C70/C60 changes from 0.26 to 8.8 (in the case of evaporation
of graphite, this ratio changes from 0.02 to 0.18).

5.1. The history of fullerenes

The icosahedral C60H60 cage was mentioned in 1965 as a possible topological

structure [7]. Eiji Osawa of Toyohashi University of Technology predicted the
existence of C60 in 1970 [8, 9]. He noticed that the structure of a corannulene molecule
was a subset of an Association football shape, and he hypothesized that a full ball
shape could also exist. Japanese scientific journals reported his idea, but it did not
reach Europe or the Americas.
Also in 1970, R. W. Henson (then of the Atomic Energy Research Establishment)
proposed the structure and made a model of C60. Unfortunately, the evidence for this
new form of carbon was very weak and was not accepted, even by his colleagues. The
results were never published but were acknowledged in Carbon in 1999 [10, 11].
Independently from Henson, in 1973 a group of scientists from the USSR, directed
by Prof. Bochvar, made a quantum-chemical analysis of the stability of C60 and
calculated its electronic structure. As in the previous cases, the scientific community
did not accept the theoretical prediction. The paper was published in 1973 in
Proceedings of the USSR Academy of Sciences [12].
In mass spectrometry, discrete peaks appeared corresponding to molecules with
the exact mass of sixty or seventy or more carbon atoms. In 1985, Harold Kroto (then
of the University of Sussex), James R. Heath, Sean O'Brien, Robert Curl and Richard
Smalley, from Rice University, discovered C60, and shortly thereafter came to discover
the fullerenes (Fig.1) [1]. Kroto, Curl, and Smalley were awarded the 1996 Nobel
Prize in Chemistry [13] for their roles in the discovery of this class of molecules. C60

62
and other fullerenes were later noticed occurring outside the laboratory (for example,
in normal candle-soot).

a b
Fig. 1. The suspension of C60 fullerene in toluene (a) and results of mass spectroscopy,
obtained by Kroto H. W. and Smalley R. E. in 1985 (b) [1]

By 1991, it was relatively easy to produce gram-sized samples of fullerene powder

using the techniques of Donald Huffman, Wolfgang Krätschmer and Konstantinos
Fostiropoulos. Fullerene purification remains a challenge to chemists and to a large
extent determines fullerene prices. So-called endohedral fullerenes have ions or small
molecules incorporated inside the cage atoms. Fullerene is an unusual reactant in
many organic reactions such as the Bingel reaction discovered in 1993. Carbon
nanotubes were recognized in 1991 [14].
Minute quantities of fullerenes, in the form of C60, C70, C76, C82, and C84
molecules, are produced in nature hidden in soot and formed by lightning discharges
in the atmosphere [15]. In 1992, fullerenes were found in a family of minerals known
as Shungites in Karelia, Russia [16].
In 2010, fullerenes (C60) were discovered in a cloud of cosmic dust surrounding a
distant star 6500 light years away. Using NASA's Spitzer infrared telescope the
scientists spotted the molecules' unmistakable infrared signature. Sir Harry Kroto, who
shared the 1996 Nobel Prize in Chemistry for the discovery of buckyballs commented:
"This most exciting breakthrough provides convincing evidence that the buckyball
has, as I long suspected, existed since time immemorial in the dark recesses of our
galaxy [16].

5.2. Methods of obtaining fullerenes

Methods of producing fullerenes are quite varied, and in particular, include an arc
discharge in an inert gas atmosphere using a graphite electrode [17], laser vaporization
of a graphite surface [18], laser pyrolysis of hydrocarbons [19], combustion of
aliphatic and aromatic hydrocarbons at higher equivalence ratio [20], pyrolysis of
hydrocarbons in a reactor [21], an arc plasma jet at atmospheric pressure [22].
Analysis methods for obtaining fullerenes on an industrial scale have shown that not
all of them are suitable for industrial production because of high energy costs and low
productivity. Recently, however, they are all, to some extent, have been developed.

63
 At present, the predominant way of industrial production of fullerenes is arc
evaporation of graphite proposed in 1990 by B. Krechmer [23, 24]. Optimization of
the production process of the arc is a certain relation between the carbon
concentration, temperature and rate of gas flowing out of the electrode gap. However,
in the arc it is almost impossible to obtain such optimum ratio, since the arc current
changes the three parameters simultaneously. The most comprehensive study was
carried out in [25], where the authors obtained the rate of erosion of the graphite
anode, the growth rate of carbon formation on the cathode and the content of
fullerenes in the soot on the arc current, type and pressure of the gas, the inter
electrode gap. For a graphite anode with a diameter of 6 mm maximum values of the
fullerene yield of 10-15% are obtained at a current of 80 A, a helium pressure of 100
Torr, and the gap between the electrodes 3-5 mm. With the decrease and increase of
current and pressure at the optimal electrode spacing a drop in the yield of fullerenes
was observed [26].

5.2.1. Arc vaporization of graphite

The first method of production of fullerenes used laser vaporization of carbon in

an inert atmosphere, but this produced microscopic amounts of fullerenes. In 1990, a
new type of apparatus using an arc to vaporize graphite was developed in Germany by
Kratschmer and Huffmann [24].
The fullerene field was revolutionized by the discovery that simple resistive
vaporization of graphite rods could produce fullerenes in substantial yield [23]. This
procedure, which is often termed the Kratschmer- Huffman method, is a
straightforward and low cost method for generating large quantities of fullerene-
containing carbon soot. A typical example of an arc vaporization fullerene generating
apparatus is shown in Fig. 2. Key components of the fullerene generating apparatus
include (1) high-current electrical feedthroughs; (2) at least one movable electrode;
and (3) a water-cooled surface to collect the vaporized carbon products. These
components are housed within a vacuum chamber that is connected to a pumping/gas
handling system that can evacuate the chamber to ≤10-3 torr and can control the
pressure between 10-500 torr. Lastly, a high current (100-200 A), low voltage AC or
DC power supply is needed to drive the evaporation; an arc welder is an inexpensive
but effective power supply for this purpose.

Fig. 2. The schematic illustration of the arc vaporization apparatus

for generating fullerene-containing soot

64
 In a typical preparative experiment, two high purity graphite rods are clamped to
the high current feedthroughs, and the chamber is pumped down to ≤10-3 torr and
refilled with He gas to a pressure of 150-250 torr. As both oxygen and water
significantly inhibit formation of fullerenes, it is important to evacuate the chamber
carefully and refill it using purified helium. The electrodes are positioned so that the
carbon rods are just touching, and then vaporization is initiated by passing a high
current through the rods. For 6.25-mm-diameter rods, currents between 100 and 200
A lead to efficient fullerene formation. Under these conditions, the 6.25-mm rods
are consumed at a rate of about 5-10 mm/min. The crude carbon product or
soot produced by this vaporization collects on the water-cooled inner surface of the
fullerene apparatus (Fig. 2) and is readily removed from the walls and collected
using a stiff brush. This soot contains a variety of carbon products including C60 and
larger fullerenes.
There are several experimental factors that can be varied to maximize the yield
of fullerene clusters in the soot produced from vaporization of graphite electrodes.
These factors include the vaporization current density and helium partial pressure.
The importance of these two experimental variables in determining the yield of
fullerenes can be understood by considering the environment in which the clusters
form. Specifically, vaporization of the carbon rods is not driven by ohmic heating
but rather by an arc discharge between the ends of the graphite rods. These latter
conditions produce a plasma, and hence, control of temperature and cooling of the
carbon plasma will determine the yield and size distribution of the fullerene
clusters. Qualitatively, increases in the current density will increase the plasma
temperature, while increases in the buffer gas (He) pressure will increase the
cooling and growth rates of carbon species. A particularly dramatic example of the
effects of He pressure is found in comparing the carbon products produced from
discharges in 150 and 500 torr of helium. When using the former conditions,
fullerene clusters are observed in high yield, while at higher He pressures carbon
nanotubes are the principle product. Unfortunately, however, systematic studies of
how the current density and the buffer gas pressure affect the yield and cluster size
distribution have not been carried out.

5.2.2. Laser ablation

A second very useful and powerful technique for producing fullerene clusters
involves laser ablation of graphite in a helium atmosphere. Historically, laser ablation
was the first technique used to generate fullerene' clusters in the gas phase [1, 27-29],
although initial attempts to prepare isolable
- quantities of C60 were unsuccessful with
this technique. The early attempts to prepare fullerenes by laser ablation relied on
ablation in an inert gas buffer at room temperature. The failure of these early attempts
to yield macroscopic quantities of fullerenes are almost certainly due to the fact that
the dense carbon plasma created during laser vaporization cools too quickly. The
rapidly cooling plasma does not provide sufficient time for the growing carbon
fragments to rearrange into stable, closed fullerene structures. In support of this
scenario, Smalley and coworkers first showed that laser ablation of graphite carried
out at elevated temperatures could in fact lead to the efficient production of
macroscopic quantities of C60 and other fullerenes [30, 31].

65
 A typical laser ablation apparatus that has been used to prepare macroscopic
quantities of fullerenes in our laboratory is shown in Fig. 3. A key feature of this
apparatus is that graphite is ablated within a high temperature furnace. By carrying out
the ablation at an elevated temperature, the plasma cools more slowly and growing
carbon clusters have sufficient time to rearrange (anneal) into stable fullerenes. We
have found that fullerene production is most efficient at 1200 °C, the upper limit of
our furnace. At lower temperatures, the yield of fullerene is reduced [30, 31]. It is
possible, however, that ablation at higher temperatures (i.e., > 1200 °C) could lead
to further enhancements in the yields of fullerenes produced by this technique.

Fig. 3. The schematic view of a laser ablation apparatus for production of C60 and other fullerenes.
In this apparatus graphite is ablated inside a quartz tube held at 1200 °C using the green (532 nm) line
from a pulsed Nd-YAG laser

There are several features of the laser ablation method that make it an excellent
technique for production of fullerenes. First, it is possible to systematically vary the
characteristics of the carbon plasma through variations in the laser pulse energy and
wavelength [32]. Second, by controlling the buffer gas pressure and the furnace
temperature it is also possible to control systematically the growth of clusters from the
carbon plasma. Control of the growth and annealing should increase the efficiency and
selectivity of the fullerenes produced. To date, these experimental variables have not
been carefully studied, although such investigations could advance significantly our
understanding of how to grow rationally new clusters. Finally, since the laser light is
easily positioned and no external electrical connections to the sample are needed, it is
possible to produce fullerenes from very small amounts of carbon precursors.

5.3. Methods of investigation of fullerenes

5.3.1. Mass spectroscopy of fullerenes

One of the most common and reliable methods for identification of fullerenes is
spectral analysis. This method serves as a source of the most complete and reliable
information on the structure and other distinctive features of this class of compounds.
The time-flight mass spectrometers with a laser ionization of the analyte are most
frequently used for investigations.

66
 In nature, carbon is mainly represented by two isotopes: 12C (98.99 %) and 13C
(1.11%). Thus, for 99 isotope atoms 12C one atom 13C is fitted. Consequently the
molecula weight of C60 may be:
– 720 a.m.u. (in case it consists only of atoms 12C) with a probability of 60 : 60 ·
98.99 = 98.89 %;
– 721 a.m.u. (59 atoms 12C and one 13C) with a probability of 60 : 1 · 1.11 = 66.6 %;
– 720 a.m.u. (58 atoms 12C and two 13C) with a probability of 60 : 2 · 1.11 = 33.3 %.
The first experimental confirmation of the existence of molecules C2n with an even
number of atoms, where 20 < n < 90 was mass spectrum, was published in 1984 (Fig.
4) [33].
Characteristic mass spectrum of
positively charged ions of fullerenes in
acetylene – oxygen flame is shown in
Fig. 5 [5].
Fig. 6 shows a typical view of the
mass spectrum of ions of one of the
samples of C60 [34]. In this case, the
sample was placed at a distance of 5
mm from the axis of the mass
spectrometer, and the time delay
between the laser pulse Texp exposure
and detection time was 10 μs. In the
spectrum, ions can be roughly divided
into three groups: a – parent ions
corresponding to the mass of C60 (720
a.m.u.); b – fragment ions formed from Fig. 4. Mass spectrum of carbon clusters in
parent ions C60 and having a mass supersonic beam, which is formed by laser
evaporation of carbon target in a pulsed nozzle
corresponding to that of C58 (696 a.m.u.), at a helium atmosphere [33]
C56 (672 a.m.u.), C54 (648 a.m.u.), etc., as
well as associative (cluster) ions formed in the process of clustering of parent ions and
fragment ions of C60 with the corresponding masses.

Fig. 5. Mass spectrum of positively charged ions of fullerenes in acetylene - oxygen flame [5]

67
 Fig. 6. Mass-spectra of C60 powder [34]

Note the feature characteristic of the fullerene molecule: the fragment and cluster
ion formation differ from each other by the amount of C2, indicating the specificity of
both the processes of fragmentation and the structure of the molecule. Mass extract of
fullerene soot produced by combustion of benzene-oxygen mixture, is shown in Fig. 7
[35]. Apart from C60 and C70, in the mass spectrum molecules of PAH are present.

Fig. 7. Mass-spectra of soot produced in the flame of benzene and oxygen [35]

5.3.2. Transmission electron microscopy of fullerenes

High resolution transmission electron micros-

copy (HRTEM) is used to obtain images of indi-
vidual fullerene molecules that reveal the shape and
size of the molecules. It is advantage of the HRTEM
and disadvantage of the chemical analysis.
In work [36], individual fullerene molecules
and their size were observed using high resolution
transmission electron microscopy. The molecular
diameter could be specified and measured to within
±0.05 nm. Fig. 8 shows a representative HRTEM
image from the analysis of C60 precipitated from
Fig. 8. HRTEM image of a pure solution directly onto the TEM grid. The C60
C60 sample [36] molecules have taken a crystalline form with a
68
twofold symmetry that is visible. Comparison of the length scale with the black
centers of the individual molecules reveals a diameter of about 0.7 nm as expected for
C60 molecules. The measured center-to-center distance between the molecules is 1.01
nm along two of the crystallographic directions and 1.14 nm along the third direction.

5.4. Synthesis of fullerenes in flame

Fullerenes C60 and C70 were discovered in carbon vapor produced by laser
irradiation of graphite and later were produced in macroscopic quantities by graphite
vaporization with resistive heating. The possibility that fullerenes could also be
formed in sooting flames had been suggested, and all-carbon ions having the same
mass/charge ratios as fullerenes in carbon vapor had been detected in flames and
assumed to have the closed-cage structure of fullerenes [37].
In 1987, Homann and co-workers reported that positive and negative ions of
fullerenes such as C60, and C70 had been found in low-pressure premixed hydrocarbon
sooting flames [3, 38–46]. After that, macroscopic quantities of fullerenes were
recovered from soot made in low-pressure benzene flames [47, 48]. Howard and co-
workers showed that the fullerene yields and composition were changed by the flame
conditions such as C/O ratio, chamber pressure, residence time, diluent concentration,
and gas velocity in combustion of benzene in a low-pressure premixed flat flames [4,
5, 37, 49-52] and low-pressure benzene/argon/oxygen diffusion flames [53, 54]. They
also showed concentration profiles of fullerenes and polycyclic aromatic hydrocarbons
(PAHs) (up to 300amu) in a heavily fullerene forming premixed benzene/oxygen
flame [55].
In 1995, Sattler and co-workers detected fullerenes in a methane/argon diffusion
flame at atmospheric pressure by on-line multiphoton ionization time-of-flight mass
spectroscopy [56, 57]. These fullerene production technologies are known as the
combustion method. As the combustion method is a continuous and easily scalable
process similar to that employed for commercial carbon black production, it was
considered to have more potential for fullerene production than the arc process. In
1998, Alford et al. presented a combustion apparatus large enough for industrial use
and could produce fullerenes continuously without halting the fullerenes synthesis
[58]. The key component of the apparatus is a porous refractory plate burner, which
improves efficiency and overcomes many of the limitations of the previous
combustion machines that employed watercooled burner plates.

5.4.1. Synthesis of fullerenes under the conditions of combustion of

a benzene–oxygen–argon mixture at a low pressure in an electric field

As is shown in [59, 60], it is practically impossible to increase the yield of

fullerenes in combustion using a traditional method. In [61, 62], it was proposed to
increase the fullerene yield by the action of an electric field on a flame. The authors of
[63, 64] have shown that synthesis of fullerenes formed in the plasma of an electric arc
increases with the increase in the electron density and in the ionization instability,
causing a change in the electron concentration from 1010 to 1011 cm–3. Despite the fact
that the conditions of formation of fullerenes in a flame differ from those in an electric
arc, they share a common trait – the initial products are ionized. An electric field

69
applied to a flame changes the electron concentration and the flame temperature, with
the result that conditions necessary for the formation of fullerenes are established.
A series of experiments on the study of the yield of fullerenes in a premixed
benzene–argon–oxygen flame exposed to a longitudinal electric field in the region of a
dark corona discharge and a glow discharge at C/O=1.0, P = 40 torr, benzene-flow
rate Q1 = 250 cm3/min, oxygen-flow rate Q2 = 758 cm3/min, argon-flow rate Q3 = 101
cm3/min (10 % of the combustion-mixture volume), and velocity of the cold mixture
inflow from a perforated stabilizer v = 18.4 cm/s have been carried out [61]. A
principal scheme of the experimental plant is given in the Fig. 9.
To study the effect of an electric field on the flame, we chose an electrode system
“needle-flat surface’ with interelectrode distance L=18 cm, located in the burner unit
of quartz glass (Fig. 10). The burner was chosen as a flat surface electrode. The time
of one experiment was t = 20 min. Constant high voltage U of negative ("minus" on an
electrode "needle") or positive ("plus" on an electrode "needle") polarity in the range
from 0.5 to 20 kV was applied to the needle electrode that was placed in the upper part
of the burner unit. As a result of applying a longitudinal electric field, negative or
positive crown discharge which influenced the flame was created. After completing
the experiment, the soot was collected from the combustion chamber and the soot-
collector filter and weighed; then the soot was extracted in benzene and investigated
by physicochemical methods.

1 – receiver; 2 – batcher- crane; 3 – reducer; 4 – crane; 5 – comsumption indicator; 6 – kilo voltmeter

C 50; 7 – evaporator-mixer agitator; 8 – vapour overheater buffer capacity; 9 – burning equipment from
quartz glass; 10 –high-voltage electric fuser of alternating current; 11 – high-voltage source of constant
current BV-9,0-1,5; 12 – vacuometer VDG-1; 13 – filter – soot collector; 14 – low-temperature catch
tank; 15 – three-throw crane; 16 – vacuum pump VN-2MG; 17 – batcher of electromechanical pump
liquid batcher; 18 – microampere meter M95

Fig. 9. The scheme of experimental plant [61]

It was stated in [62, 65] that the upper electrode with a negative-polarity exerts a
large influence on the fullerene formation as compared to the one with positive-
polarity. However, in experiments conducted with a negative-polarity upper electrode
in the region of dark, corona, and glow discharges at an interelectrode distance H = 18
cm with the use of a needle-plane electrode system, a significant increase in the yield
of fullerenes was not detected. At an interelectrode distance H = 18 cm, the upper
70
electrode was positioned at a height of 10-11 cm above the upper edge of a flame.
Under the indicated conditions, a discharge acted on the upper part of the flame, where
a smaller number of ions are found [66-68] and the formation of the output
combustion products is practically absent. The glow discharge occurs in the form of a
glowing vibrating string of bluish colour with the diameter of 3-10 mm, connecting
the top electrode with the bottom, through the flame, Fig. 11 [69].

Fig. 10. The burner scheme and photograph

In this case, the calculated electron density did not exceed ~105÷107 cm-3. It was
stated in the earlier investigations that the optimum density of the free electrons,
providing a maximum yield of fullerenes, is ~109 cm-3 [64]. It is known [70] that, if
the density of the charged particles in a gas is very small, they interact mainly with
neutral particles.

Fig. 11. Type of glow discharge at an interelectrode distance of H = 18 cm [69]

This interaction is short-lived and the pair collisions are of first importance in such
an ionized gas. When the density of the charged particles increases, the role of their
interactions with each other increases gradually too. It is apparent that, under the
above-described experimental conditions, the density of the charged particles is
insufficient to effectively influence the fullerene formation at an interelectrode
distance of 18 cm.
Investigations carried out under the conditions where the upper electrode was
positioned directly above a flame have shown that, in this case, the yield of fullerenes
increases [62]. The influence of the type of electrode and the height of its position
71
above a flame on the formation of fullerenes was investigated for determining the
conditions providing their maximum yield. Electrodes in the form of a needle and a
ring were used. The investigations were carried out at a negative polarity of the upper
electrode under the conditions where U = 7 kV and H = 1-9 cm (with a step of 1 cm).
A glow discharge appeared independently of the type of an electrode at different
heights of its disposition above a flame; this discharge caused an intense glow of the
whole flame (Fig. 12). In this case, the average temperature of the flame increased to
1200 oC (without a field, this temperature was equal to 950 oC).

a b
a) a flame without a discharge; b) the flame exposed to a discharge (U = 7 kV)

Fig. 12. Action of a glow discharge on the yield of fullerenes in the case where a ring cathode is
positioned above the central part of a flame

Thus, in the case where the upper electrode is placed directly above a flame, the
near-cathode region of the glow discharge exerts a stronger influence on the processes
occurring in the flame. In this case, in the IR spectra of soot extracts, there appeared
all the spectral lines corresponding to the fullerenes C60 ( - 528, 577, 1183, and 1429
cm-1), Fig. 13 [62].
0,220
1462,23 794,93 526,61
1414,50 1182,73 674,13
0,20 1430,17 565,96
1134,14 641,78 577,42 455,78
0,18

0,16 535,57
459,25

0,14

0,12
535,57
0,10

0,08
A

0,06 
0,04

0,02
458,67
0,00

-0,02

-0,04
534,55

-0,06 0
1500,0 1400 1300 1200 1100 1000 900 800 700 600 500 450,0
CM-1

1 – reference sample C60; 2 – electrode needle (Ux.x. = 7 kV, I = 7x10-3 A); 3 – electrode ring (Ux.x. =
7 kV, I = 7x10-3 A)

Fig. 13. IR spectra of the dry extracts of soot at d = 4 cm

72
 Table 1

Influence of an electric field on the formation of fullerenes in the combustion of the

benzene–oxygen–argon mixture

Electrode Soot yield, mg Soluble part yield, % Yield of C60, %

Н = 6 сm, I = 6.6 mА, Uн = 7 kV
Needle 0.5879 65.64 5
Ring 0.3825 47.54 13
Н = 4 сm, I = 7 mА, Uн = 7 kV
Needle 0.3421 42.15 10
Ring 0.2245 62.56 15
Н = 6 сm, I = 4.5 mА, Uн = 12.5 kV
Ring 0.4909 36,68 2
Н = 4 сm, I = 4 mА, Uн=7 kV
Ring 0.3623 40.81 5
Н = 1 сm, I = 3.8 mА, Uн = 5 kV
Ring 0.5284 24.79 1
Without field
No electrodes 0.5958 22.55 1

It is stated that a ring electrode offers an advantage over a needle one, and the
maximum yield of the C60 fullerene (β = 15 % of the soot formed) was obtained in the
case where this electrode was positioned above the peripheral part of the central
reaction zone of the flame (Table 1) [62].
The increase in the yield of fullerenes in the case where a ring electrode is
positioned above the peripheral part of the reaction zone of a flame is due to the glow-
discharge conditions providing an effective synthesis of fullerenes.
This supposition is based on the results of the earlier investigations [44], showing
that, if the peripheral zone of a benzene flame is heated by any external heat source,
e.g., a laser beam, which does not simply burn the soot but forms conditions identical
to those at the center of the flame, the concentration of fullerenes increases.
The photographs of soot samples obtained with the use of a high-resolution JEM-
3010 electron microscope of the JEOL firm with a magnifying power of up to
1,500,000 times at an accelerating voltage of 300 keV, show that these samples are
highly structurized, which is characteristic of fullerene formations. It is known that the
fullerenes formed in a flame are adsorbed, along with the polycyclic hydrocarbons, on
the soot particles. The indicated photographs are presented in Fig. 14 [71-73].
The soot samples shown on them comprise agglomerates consisting of spherical
and nonspherical particles and amorphous soot (Fig. 14 a). In these samples, spherical
nanostructures with a diameter of up to 2 nm are also present (Fig. 14 b). In (Fig. 14
c), straight and bent carbon structures are seen. In photographs obtained with high
magnifying power (Fig. 14 d), one can see a layered carbonaceous material.
The mass-spectrometric investigations of benzene extracts of soot were carried out
with the use of an ion-cyclotron-resonance Fourier spectrometer of the FT ICR MS
Bruker "Apex Qe" type with an electric-spray ionization. They have shown that the
fullerene oxides C60O, C60O2, C60O3, C70O, and C94O as well as the higher fullerenes
C74, C78, C90, and C94 are present in the indicated extracts [74]. The results of
identification of fullerenes by their mass spectra are presented in Table 2.

73
 а b

c d

Fig. 14. Microscopic photographs of fullerene-containing soot samples obtained

with different magnifying powers [69]

Table 2

The results of identification of fullerenes by the mass-spectra of benzene soot extracts

Formula of a substance Calculated mass, Mass determined by the

a.m.u spectrum, a.m.u
С24 288.26 288.3
С16Н10 (pyrene, fluoranthene) 202.26 202.4
С18Н10 (cyclopentapyrene 226.28 226.4
benzofluoranthene)
С20Н10 (corannulene) 250.3 250.3
С60О 736.66 736.5
С60О3 768.66 768.5
С60Н2 722.68 722.5
С70О 856.77 856.7
С70Н2 842.78 842.6
С74 888.81 888.6
С78 936.8 936.8
С90 1080.99 1081.7
С94 1129.03 1128.9
С94О 1145.59 1145.5

74
 Thus, it has been stated that the yield of fullerenes increases under the action of a
glow discharge in the case where an electrode (a needle or a ring) is positioned
directly above a flame, and the maximum yield of fullerenes is attained with the use of
a ring electrode positioned above the central region of the flame front. The maximum
yield of the fullerene C60 was β = 15% of the soot formed.

5.5. Properties of fullerenes

Before describing fullerene properties, special attention should be given to the

fullerene structure. For the full concept of fullerene structure let us consider an
example: if the C60 molecule has the size of a soccer ball, then the soccer ball in turn
would be about the size of the earth. This is (a) and the structure of fullerene (b) are
shown in Fig. 15.

a b

Objects diameter: Earth: 12,750 Km = 12.75×106 m, Soccer ball: 22 cm = 2.2×10-1 m, C60 molecule:
7Å = 7.0×10-10 m

Fig. 15. Comparison of C60 molecule with the size of a soccer ball (a) [75]
the structure of fullerene C60 (b)

Fullerenes consist of 20 hexagonal and 12 pentagonal rings as the basis of

icosohedral symmetry of a closed cage structure. Each carbon atom is bound to three
others and is sp2 hybridized. The C60 molecule has two bond lengths - the 6:6 ring
bonds can be considered "double bonds" and are shorter than the 6:5 bonds. C60 is not
"superaromatic" as it tends to avoid double bonds in the pentagonal rings, resulting in
poor electron delocalisation. As a result, C60 behaves like an electron deficient alkenes
and reacts readily with electron rich species. The geodesic and electronic binding
factors in the structure account for the stability of the molecule. In theory, an infinite
number of fullerenes can exist, their structure is based on pentagonal and hexagonal
rings, constructed according to the rules for making icosahedra [75].

5.5.1. Solubility of fullerenes

Fullerenes have appreciable solubility in a broad class of organic solvents. The

data on solubility of fullerenes C60 and C70 in various organic solvents are shown in
75
Tables 3 and 4. This feature is explained by the unusual structure of fullerenes, with
increased reactivity. The interest to investigations on the behavior of fullerenes in
solutions is aroused by the fact that fullerenes are the only soluble form of carbon that
is of both applied and fundamental importance. The uniqueness of the fullerene is
revealed in the fact that in solutions they form clusters of several molecules.
Interaction of the solvent with fullerenes is accounted for van der Waals forces.

Table 3

Solubility of C60 in various solvents [76]

Solvent Solubility, mg/ml

isooctane 0.026
N-pentane 0.005
N-Hexane 0.043
octane 0.025
dioxane 0.041
dean 0.070
decalin 4.6
dodecane 0.091
tetradecane 0.126
cyclohexane 0.036
carbon tetrachloride 0.32
Trichloroethylene 1.4
tetrachlorethylene 1.2
1,1,2,2-tetrahloetilen 5.3
methylene chloride 0.245
methanol 0.000
ethanol 0.001
N-methyl-2-pyrrolidine 0.89
benzene 1.7
toluene 2.8
xylenes 5.2
mesitylene 1.5
tetralin 16.0
bromobenzene 3.3
anisole 5.6
chlorobenzene 7.0
1,2-dichlorobenzene 27.0
1,2,4-trichlorobenzene 8.5
1-methylnaphthalene 33.0
dimethylnaphthalenes 36.0
1-phenylnaphthalene 50.0
6-chloronaphthalene 51.0
carbon disulfide 7.9

76
 Table 4
Solubility of С70 [76]

Solvent Solubility, mg/ml

toluene 1.067
benzene 1.302
1,2-dimethylbenzene 15.624
1,4-dimethylbenzene 3.982
1,3,5-trimethylbenzene 1.47
1,2-dichlorobenzene 29.988
1,3-dichlorobenzene 18.732
N-pentane 0.002
N-Hexane 0.0126
N-heptane 0.047
N-octane 0.042
H Dean 0.053
dodecane 0.101
cyclohexane 0.0798
isopropanol 0.0021
carbon tetrachloride 0.118
carbon disulfide 15.288
tetralin 12.264
dichloromethane 0.0798
water 1.344  е-10

5.5.2. Physical properties

Due to the high reactivity of higher fullerenes and a small percentage of their
output in synthesis of fullerenes, their properties are very poorly elucidated in
literature. The properties of fullerenes C60 and C70 are more widely studied and widely
reported. Affinity of C60 to electron makes up 2.65 eV, the ionization potential is 7.61
eV, the binding energy per carbon atom is 7 eV, the polarizability is close to 80 Å3. A
molecule can take up to 12 electrons and release one electron, i.e. the charge on C60
can vary from 1 to - 12. In the condensed state, a fullerene is molecular crystals -
fullerites, in which the binding energy between molecules is 1.86 eV and is realized by
van der Waals forces. Fullerites have orientational disorderness characterized by the
ability of heated molecules to change the relative spatial orientation under the
conditions of the of the crystalline environment, which is typical for low energy
barriers between the different positions of the molecules of the order of 0.2-0.3 eV.
The physical constants properties of fullerene C60 are presented in Table 5.
Table 5

The physical constants of fullerene C60

Average C-C distance 1.44 Å

FCC Lattice constant 14.17 Å
C60 mean ball diameter 6.83 Å
C60 ball outer diameter 10.18 Å

77
C60 ball inner diameter 3.48 Å
Tetrahedral site radius 1.12 Å
Octahedral site radius 2.07 Å
Mass density 1.72 g/cm3
Molecular density 1.44 x 1021/cm3
Compressibility (-d in V/dP) 6.9 x 10-12/cm3/dye
Bulk modulus 14 Gpa
Structural phase transitions 255K, 90K
Binding energy per atom 7.4 eV
Electron affinity (pristine C60) 2.65 eV
Ionization potential (1st) 7.58 eV
Ionization potential (2nd) 11.5eV
Vol. Coeff. of thermal expansion 6.2 x 10-5 cm3/K
Band gap (HOMO-LUMO) 1.7 eV
Spin- orbit splitting C (2p) 0.0022 eV
Velocity of sound vt 2.1 x 105 cm/sec
Velocity of sound vl 3.6 x 105 cm/sec
Debye temperature 185 K
Thermal conductivity (300 K) 0.4W/mK
Phonon mean free path 50 Å
Static dielectric constant 4.0 – 4.5
Standard heat of formation 9.08 k cal mol-1
Index of refraction 2.2 (600nm)
Boiling point Sublimes at 800K
Resistivity 1014 ohms m-1
Vapor density N/A
Vapor pressure 5 x 10-6 torr at room temperature
Crystal form Hexagonal cubic
8 x 10-4 torr at 800K

5.6. Applications of fullerenes

The uniqueness of the fullerene structure accounts for their unusual physical and
chemical properties which are greatly promising for their practical use. Fullerenes are
similar in their properties to graphite, and therefore inherit its high heat resistance,
extraordinary electrical properties as well as its anti-friction properties. Specific three-
dimensional structure of fullerene molecules makes it possible to obtain a previously
unknown class of heteroatomic compounds - doped fullerenes, and these compounds
have begun a new turn in chemical (biochemical) science. Here are the most
promising areas of application of fullerenes and their derivatives, some of which
received a practical embodiment. Most patents abroad are devoted to application of
fullerenes in semiconductor technology and nanoelectronics as photodiodes,
transistors and solar cells. C60, having high electro negativity, readily forms charge-
transfer complexes, where acts as an electron acceptor. In diodes consisting of
fullerene as an acceptor covalently bound to a photoactive donor, photo induced
electron transition occurs under the effect of light. High efficiency of light absorption
and charge separation is a fundamental of principle operation of solar cells. In the near
future, this property of fullerenes may be used for creation of photo energy converters
78
and other devices using photoconductivity. Application of fullerenes for growing
diamond films also proved to be promising. In the process of destruction by
microwave radiation, molecules of C60 form carbon pair C-C. Being deposited on the
surface of SiO2, they form diamond microcrystals forming smooth diamond films.
There are developments on the use of fullerenes and their endohedral derivatives,
including inside the carcass atoms with a half-integer spin such as hydrogen or
phosphorus, as the quantum computer parts [5].
Fullerenes are able to take over a large number of free electrons, so they are strong
oxidizing agents and form many new chemical compounds. The possibilities of
application of fullerenes in pharmacology are perhaps most numerous. At present,
antiviral and neurotropic drugs, adsorbents for hemosorption as well as methods of
treatment with their use are being developed on the basis of fullerenes.

CONTROL QUESTIONS

1. Can you tell about history of fullerenes?

2. What kind of methods for synthesis of fullerenes do you know?
3. Can you describe the method of synthesis of fullerenes in the arc vaporization of graphite?
4. Which type of methods for identification of fullerenes do you know?
5. What do you think of synthesis of fullerenes in flame?
6. What is difference between synthesis of fullerenes in flames with electric field and without the
electric field?
7. What are properties of fullerenes?
8. Can you describe physical properties of fullerenes?
9. Can you list solubility of C60 in various solvents?

REFERENCES

1. H.W. Kroto, J.R. Heath, S.C. O’Brien, et al. C60: Buckminsterfullerene, Nature, 318, 162–163 (1985).
2. A. Hirsh (Ed.), Fullerenes and Related Structures, Springer, Berlin (1999).
3. P. Gerhardt, S. Loffler, K.-H. Homann. The formation of polyhedral carbon ions in fuel-rich
acetylene and benzene flames // Proc. 22nd Int. Symp. Combust., The Combustion Inst., Pittsburgh
(1988), pp. 395–401.
4. J.B. Howard, J.T. McKinnon, Y. Makarovsky, et al. Fullerenes C60 and C70 in flames, Nature, 352,
139–141 (1991).
5. J. B. Howard, A. L. Lafleur, Y. Makarovsky, S. Mitra, C. J. Pope, and T. K. Yadav, Fullerenes
synthesis in combustion, Carbon, 30, 1183–1201 (1992).
6. H. Richter, A. J. Labrocca, W. J. Grieco, K. Taghizadeh, A. L. Lafleur, and J. B. Howard, Generation
of higher fullerenes in flames, J. Phys. Chem. B, 101, 1556–1560 (1997).
7. H.P. Schultz. Topological Organic Chemistry. Polyhedranes and Prismanes. // Journal of Organic
Chemistry 30 (5): 1361. (1965).
8. E. Osawa, Kagaku 1970, 25, 854.
9. B. Halford (2006). "The World According to Rick".Chemical & Engineering News 84 (41):
13. doi:10.1021/cen-v084n041.p013.
10. P.A. Thrower. (1999). "Editorial". Carbon 37 (11):1677
11. R.W. Henson. (2010). The History of Carbon 60 or Buckminsterfullerene
12. D.A. Bochvar, E.G. Galpern. (1973). Dokl. Acad. Nauk SSSR209: 610.
13. The Nobel Prize in Chemistry 1996. Retrieved 7 February 2014.
14. S.J. Mraz (2005). A new buckyball bounces into town. Machine Design.
15. The allotropes of carbon. Interactive Nano-Visualization in Science & Engineering Education.
Retrieved 2010-08-29.
16. Stars reveal carbon 'spaceballs', BBC, July 22, 2010.
17. Alekseyev N.I., Dyuzhev G.A. Fullerene formation in an arc discharge // Carbon. – 2003. – Vol. 41. –
№ 7. – P.1343-1348.
18. Kasuya D., Kokai F., Takahashi K., et. al. Formation of C60 using CO2 laser vaporization of graphite at
room temperature // Chem. Phys. Lett. – 2001. – Vol. 337. – № 1-3. – P. 25-30.

79
19. Petcu S., Cauchetier M., Armand X., et. al. Formation of fullerenes in the laser pyrolysis of benzene //
Combust. Flame. – 2000. – Vol. 122. – № 4.– P. 500-507.
20. Takehara H., Fujiwara M., Arikawa M., et. al. Experimental study of industrial scale fullerene
production by combustion synthesis // Carbon. – 2005. – Vol. 43. – Р. 311-319.
21. A.F. Bubliyevskiy, A.A. Galinovskiy, A.V. Gorbunov et al. Plazmennyy piroliticheskiy sintez
uglerodnykh nanostruktur v smesi azota i propan - butana // Inzhenerno - Fizicheskiy Zhurnal. –
2006 - T. 79. - № 2 - S. 3-9 [in russian]
22. R. Dubrovsky, V. Bezmelnitsyn. Enhanced Approach to Synthesis Carbon Allotropes by arc plasma //
Rev. Adv. Mater. Sci. – 2003. – V. 5. – P. 420 – 424.
23. W. Kratschmer, L.D. Lamb, K. Fostiropoulos, D.R. Huffman. Solid C60: a new form of carbon // Nature.
– 1990. – Vol. 347. – P. 354–358.
24. W. Kratschmer, K. Fostiropoulos, D.R. Huffman. The success in synthesis of macroscopic quantities of
C60 // Chem. phys. let. – 1990. – Vol. 170.– P. 167-175.
25. R. Dubrovsky, V. Bezmelnitsyn. New Approach in Synthesis of Carbon Allotropes in Large Quantities //
Fullerenes, Nanotubes, and Carbon Nanostructures. – 2003. – V.12. – Is.1-2. – P. 17–24.
26. Laplaze D., Bernier. P., Journet C., et. al. The use of solar energy for the production of fullerenes and
porous silicon // Journal de Physique. – 1997. – Vol. 7. – № 3. – P. 463-472.
27. S.C. O'Brien, R. Heath, R.F. Curl, R.E. Smalley. (1988). Photophysics of buckminsterfullerene
and other carbon cluster ions // J. Chem. Phys. 88, 220-230
28. R. Heath, R. F. Curl, R. E. Smalley. The UV absorption spectrum of C60 (buckminsterfullerene): A
narrow band at 3860 Å J. Chem. Phys. 87, 4236 (1987).
29. R. F. Curl and R. E. Smalley. Probing C60 // Science 242, 1017 (1988).
30. H. Kroto. "Space, Stars, C60 and Soot", Science, 1988, 242, 1139.
31. R. E. Haufler, Y. Chai, L. P. F. Chibante, J. Conceicao, C. Jin, L. S. Wang, S. Maru-
yama, and R. E. Smalley, Mater. Res. Soc. Symp. Proc 206, 627 (1991).
32. J. T. Cheung and H. Sankur, CRC Critical Rev. in Solid State and Mat. Sci. 15, 63 (1988).
33. E.A. Rohlfing, D.M. Cox, A.J. Kaldor. Production and characterization of supersonic carbon cluster
beams. Journal Chemical Physics, 81, 3322−3330 (1984).
34. G.N. Churilov, N.V. Bulina, A.S. Fedorov. Fullereny: Sintez i teoriya obrazovaniya. – Novosibirsk:
Izd - vo SO RAN, 2007 – 230 [in russian].
35. V.V. Brazhkin, A.G. Lyapin, S.G. Lyapin. Novyye kristallicheskiye i amorfnyye modifikatsii
ugleroda, poluchennyye iz fullerita pri vysokom davlenii // UFN. – 1997 – T.167. – № 9 – S. 1019-
1022 [in russian].
36. A. Goel, J.B. Howard, J.B.V. Sande. Size analysis of single fullerene molecules by electron
microscopy // Carbon 42 (2004) 1907–1915
37. J.B. Howard, J.T. McKinnon, M. E. Johnson, Y. Makarovsky, A.L. Lafleur Production of C60 and C70
Fullerenes in Benrene-Oxygen Flames // J. Phys. Chem. 1992, 96,6657-6662
38. Gerhardt PH, Loffler S, Homann KH. Polyhedral carbon ions in hydrocarbon flames // Chem Phys
Lett 1987; 137:306–10.
39. Gerhardt PH, Homann KH. Ions and charged soot particles in hydrocarbon flames // J Phys Chem
1990; 94:5381–91.
40. Baum T, LoЁffler S, WeilmuЁnster P, Homann KH. Polycyclic aromatics, fullerenes and soot
particles as charged species in flames // Prepr Pap Am Chem Soc Div Fuel Chem 1991; 36: 1533–
40.
41. Baum T, LoЁffler S, LoЁffler P, WeilmuЁ nster P, Homann KH. Fullerene ions and their relation to
PAH and soot in low-pressure hydrocarbon flames // Ber BunsenGes Phys Chem 1992; 96: 841–57.
42. Ahrens J, Kovacs R, Shafranovskii EA, Homann KH. Online multi-photon ionization mass
spectrometry applied to PAH and fullerenes in flames // Ber Bunsen Ges Phys Chem 1994; 98: 265–
8.
43. Ahrens J, Bachmann M, Baum T, Griesheimer J, Kovacs R, WeilmuЁnster P, et al. Fullerenes and
their ions in hydrocarbon flames // Int J Mass Spectrom Ion Proc 1994; 138: 133–48.
44. Bachmann M, Wiese W, Homann KH. Fullerenes versus soot in benzene flames // Combust Flame
1995; 101: 548–50.
45. Bachmann M, Wiese W, Homann KH. PAH and aromers: precursors of fullerenes and soot. Proc
Combust Inst 1996; 26: 2259–67.
46. Ahrens J, Keller A, Kovacs R, Homann KH. Large molecules, radicals, ions, and small soot //
Particles in fuel-rich hydrocarbon flames // Ber Bunsenges Phys Chem 1998; 102: 1823–39.
47. Howard JB, McKinnon JT, Makarovsky Y, Lafleur AL, Johonson ME. Production and
characterization of fullerenes in flames // Prepr Pap Am Chem Soc Div Fuel Chem 1991;
36(3):1022–5.
48. McKinnon JT, Bell WL, Barkley RM. Combustion synthesis of fullerenes // Combust Flame 1992;
88: 102–12.
49. Pope CJ, Howard JB. Fluxes and net reaction rates of flame species pertinent to fullerenes // Prepr
Pap Am Chem Soc Div Fuel Chem 1991; 36(4):1541–6.

80
50. Mitra S, Pope CJ, Gleason KK, Makarovsky Y, Lafleur AL, Howard JB. Synthesis of fullerenes (C60
and C70) by combustion of hydrocarbons in a flat flame burner // Proc Mater React Soc Symp 1992;
270: 149–54.
51. Howard JB. Fullerenes formation in flames // Proc Combust Inst 1992; 24: 933–46.
52. Howard JB. Combustion synthesis of fullerenes // Combust Fundam Appl, Jt Tech Meet, Cent East
States Sect Combust Inst 1993:10–9.
53. Hebgen P, Goel A, Howard JB, Rainey LC, Sande JBV. Synthesis of fullerenes and fullerenic
nanostructures in a low-pressure benzene/oxygen diffusion flame // Proc Combust Inst 2000; 28:
1397–404.
54. Goel A, Hebgen P, Sande JBV, Howard JB. Combustion synthesis of fullerenes and fullerenic
nanostructures // Carbon 2002; 40: 177–82.
55. Grieco WJ, Lafleur AL, Swallow KC, Richter H, Taghizadeh K, Howard JB. Fullerenes and PAH in
low-pressure premixed benzene/oxygen flames // Proc Combust Inst 1998; 27:1669–75.
56. Hepp H, Siegmann K, Sattler K. Multiphoton ionization mass spectroscopy of fullerenes in methane
diffusion flames // Proc Mater Res Soc Symp 1995; 359: 517–22.
57. Siegmann K, Hepp H, Sattler K. High-resolution height-profile analysis and laser-ionization
characterization of a wide range of fullernes in laminar diffusion flames // Surf Rev Lett 1996; 3(1):
741–5.
58. Alford JM, Diener MD, Wright JD. Large-scale production of fullerenes by combustion // Book of
abstracts. 215th ACS national meeting, Dallas, March 29–April 21, 1998.
59. A.A. Bogdanov, D. Deininger, G.A. Dyuzhev. Prospects for development of industrial methods for
production of fullerenes // Zh. Tekh. Fiz., 70, Issue 5, 1–7 (2000) [in russian]
60. I.V. Zolotukhin and A.V. Gustov. Analysis of methods for production of fullerenes // Perspek.
Materialy, No. 2, 5–12 (2002).
61. Z.A. Mansurov, N.G. Prikhodko, T.T. Mashan, B.T. Lesbaev. Formation of PAH, fullerenes,
nanoparticles and soot at combustion of hydrocarbons in electric field // Proc. 20th Int. Colloquium
on the Dynamics of Reactive Systems, July 31–August 5, 2005, Montreal, Canada. CD, P. 5.
62. N.G. Prikhodko, B.T. Lesbaev, Z.A. Mansurov. Formation of fullerenes in a flame subjected to a gas
discharge // Izv. Nats. Akad. Nauk RK, Ser. Khimicheskaya, No. 2, 15–18 (2006).
63. G.N. Churilov, A.S. Fedorov, P.V. Novikov. Influence of electron concentration and temperature on
fullerene formation in a carbon plasma // Carbon, 41, No. 1, 173–178 (2003).
64. K.L. Stepanov, Yu.A. Stankevich, L.K. Stanchits, et al. Influence of the electron density on the
kinetics of formation of fullerenes in a carbon plasma // Pis’ma Zh. Tekh. Fiz., 29, Issue 22, 10–15
(2003) [in russian]
65. N.G. Prikhodko, B.T. Lesbaev, T.T. Mashan, Z.A. Mansurov, Soot formation in conbustion of a
benzol–oxygen mixture in an electric field at a pressure of 40 Torr // Gorenie Plazmokhim., 2, No. 1,
59–71 (2004) [in russian]
66. E.M. Stepanov and B. G.D’yachkov, Ionization in a Flame and the Electric Field // Metallurgiya,
Moscow (1968) [in russian]
67. J. Lawton, F. Weinberg (V. A. Popov Ed.), Electrical Aspects of Combustion // Energiya, Moscow
(1976) [in russian]
68. H.F. Calcote, R.J. Gill. Comparison of the ionic mechanism of soot formation with a free radical
mechanism // in: H. Bockhorn (Ed.), Soot Formation in Combustion. Mechanisms and Models,
Springer Series in Chem. Phys., Springer, Berlin (1994), Vol. 59, pp. 471–484.
69. Z.A. Mansurov, N.G. Prikhodko, A.V. Saviliev. Obrazovanie PAH, fullerenov, uglerodnih
nanotrubok I sazhi v processax goreniya // Almaty: Kazak university, 2012. -383 s.
70. Yu.P. Raizer, Gas Discharge Physics // Nauka, Moscow (1987) [in russian]
71. Z.A. Mansurov, N.G. Prikhodko, B.T. Lesbaev, T. T. Mashan, Combustion of the premixed
benzeneoxygen mixture in electric field at low pressure, Proc. 31st Int. Symp. Combust., Heidelberg
(2006), P. 164.
72. Z.A. Mansurov, T.A. Shabanova, V.L. Levin, N.G. Prikhod’ko, A novel characteristic of a C60–C70-
fullerene containing substance (according to electronic-microscopy data) // Vestn. KazNU, Ser.
Khimicheskaya, 39, No. 3, 444–448 (2005).
73. Z.A. Mansurov, N.G. Prikhodko, B.T. Lesbaev, M. Auyelkhankyzy. Soot and fullerenes formation in
the premixed benzene-oxygen mixture in electric field at low pressure // Eurasian chemico-techn.
Journal. 2012. V. 14. 1. P. 13-18
74. Z.A. Mansurov, B.T. Lesbaev, D.I. Chenchik, et. al. Synthesis of fullerenes and carbon nanotubes in
flames // Book of abstracts, Int. Conf. on Carbon, Nagano (2008), pp. 134–139.
75. B.C. Yadav, R. Kumar Structure, properties and applications of fullerenes // International Journal of
Nanotechnology and Applications. ISSN 0973-631X Volume 2, Number 1 (2008), pp. 15–24
76. Bezmelnitsyn VN Eletskii AV, M. Okun Fullerenes in solutions // Physics-Uspekhi. – 1998. – T. 168.
– № 11. – S. 1195-1220 [in russian]

81
 Chapter 6

SYNTHESIS OF CARBON NANOTUBES IN FLAME

Carbon nanotubes (CNTs) are allotropes of carbon with a cylindrical

nanostructure. Nanotubes have been constructed with length-to-diameter ratio of up to
132,000,000:1 [1], significantly larger than for any other material. These cylindrical
carbon molecules have unusual properties, which are valuable for nanotechnology,
electronics, optics and other fields of materials science and technology. In particular,
owing to their extraordinary thermal conductivity and mechanical and electrical
properties, carbon nanotubes find applications as additives to various structural
materials. For instance, nanotubes form a tiny portion of the material in some
(primarily carbon fiber) baseball bats, golf clubs, or car parts [2].
Nanotubes are categorized as single-walled nanotubes and multiple walled
nanotubes. Techniques have been developed to produce nanotubes in sizeable
quantities, including arc discharge, laser ablation, chemical vapor deposition, silane
solution method and flame synthesis method. The properties and characteristics of
CNTs are still being researched heavily and scientists have barely begun to tap the
potential of these structures. They can pass through membranes, carrying therapeutic
drugs, vaccines and nucleic acids deep into the cell to targets previously unreachable.
Overall, recent studies regarding CNTs have shown a very promising glimpse of what
lies ahead in the future of medicines.

6.1. The history of carbon nanotubes

The history of filamentous micro-and nanosized particles of carbon has more than
a hundred years, it is stated in monography [3]. Hughes and Chambers in 1889 [4] was
issued the first patent for obtaining of a tubular form of carbon formed during the
pyrolysis of a mixture of CH4 and H2 in an iron crucible. After that periodically
appeared new publications on fila-
mentous carbon: 1890 – Shultzen-
berger [5], 1928 – Fisher, 1931 –
Hoffman, who described the for-
mation of catalytic filamentous car-
bon in the disproportionation reaction
of CO on Fe at 673-973 K, accom-
panied by an intense dispersion of the
catalyst and the formation of carbon,
such graphite [6]. In 1948, ils and
Railey discovered carbon fibers by
pyrolysis of CH4, C2H6 and C2H4 on a
quartz substrate at 1473 K.
Fig. 1. Carbon nanofilaments obtained by the A paper by Soviet scientists
thermal decomposition of carbon monoxide on iron
catalysts (x20000) [7] Radushkevich and Lukyanovich
82
published in 1952 [7] clearly showed electron-microscopic observation of carbon
filaments with a diameter of 100 nm, obtained by the thermal decomposition of carbon
monoxide on iron catalyst.
First was used transmission electron microscopy for the study of the produced
particles [7]. The resolving power of the microscope was, by modern standards, low
and reached only 5-6 nm, but allowed us to make important observations. In [7] work
was established the presence of particles having a diameter of 30 nm and a length of 7
m at the ends to the presence of the catalyst particles (Fig. 1). It was noted in the
inside of the particle availability a bright band, which similar to the channel and it was
extending along the entire length of the particle.
However, it was not officially registered as a carbon
nanotube. Since that time the interest to carbon
nanofilaments has not been decreased. One of the
main work is work of Endo [8], this work describes
the growing of carbon fibers in the vapor phase. The
further research of carbon fibers led to opening of
nanotubes.
Carbon nanotubes had opened by Iijima
unexpectedly in 1991, after the catalyst-free
formation of nearly perfect concentric multiwall
carbon nanotubes (c-MWNTs) was reported [9] as
by-products of the formation of fullerenes via the
electric-arc technique. But the real breakthrough
Fig. 2. Microscopic image
occurred two years later, when attempts to fill the multiwalled nanotubes of five (a),
nanotubes in situ with various metals led to the two (b) and seven (c) graphite
discovery – again unexpected – of single-wall carbon layers and their model of cross-
nanotubes (SWNTs) [10]. sectional [9]

6.2. Methods of obtaining carbon nanotubes

There is a whole range of methods for synthesizing of carbon nanotubes. Arc

discharge [11], laser ablation [12], and chemical vapor deposition (CVD) [13] are the
main methods for production of carbon nanotubes.

6.2.1. Arc discharge

Arc discharge is the most widespread method for production of carbon nanotubes
[11, 14-16]. There were first discovered carbon nanotubes in the products of arc
synthesis [7]. The latest version of the arc-discharge method is shown in Fig. 3 [10].
Arc discharge belongs to the methods that use higher temperatures (above 1700 0C)
for CNT synthesis which usually causes the growth of CNTs with fewer structural
defects in comparison with other techniques.
In this method, carbon source with Fe, Ni, Co catalysts located at discharge region
is vaporized due to discharge heat. The vaporization carry out in a chamber with
reduced pressure (~ 600 mbar), and vapours which include CNTs deposit both on the
surface of cathode and on the wall of the chamber. The disadvantages of this method
are high temperature of process, reduced pressure and using expensive noble gasses.

83
 Electric arc discharge in liquid environment was first
used by Hsin et al [17] in water in 1991 in order to
synthesize CNTs. This simplified method does not require
expensive noble gases, high temperature furnace and
vacuum equipments [18]. Since then, benzene, toluene
[19] and the liquid nitrogen [20]-[22) were also used as
liquid environment. Sano et al produced a fine quantity of
carbon nano onions and CNTs from arc discharge in
deionised water [20]-[24]. Arc discharge was not stable in
deionised water due to its electrical insulating
characteristic. Therefore, salt solution was used as the
liquid environment in order to improve electrical
conductivity of solution [25].

6.2.2. Pulsed laser vaporization

The pulsed laser vaporization method was originally

developed by Smalley and co-workers for the production
Fig. 3. The scheme of the
production of carbon of fullerenes and CNTs [16, 26, 27]. The quantity of
nanotubes the arc- material formed in this method was so small that a
discharge method
structural identification by electron microscopy was not
feasible instead identification took place only by mass
spectrometry [28]. In order to produce larger amounts of fullerenes, the same group
later developed the laser-furnace method [29]. But the method proved to be more
suitable for the formation of SWNTs rather than fullerenes. In PLV the target is
vaporized in a high temperature Ar buffer gas, Fig. 4. The PLV technique presented
various advantages including SWNT formation, diameter control, allowed research on
formation mechanism, and new materials production. TheSWNTdiameter can be
controlled by adjusting the temperature, catalyst composition, and gas flow rate [30-
32]. At higher temperatures, SWNTs of larger diameters are produced. Equally, the
introduction of a NieY catalyst enlarges the nanotube diameter, while applying a
RhePd catalyst reduces the diameter [33, 74]. SWNTs with minimal defects were
synthesized using the PLV technique accompanied by purification processes [35-37].
Disadvantages of the PLV method include the limited scale of SWNT production.
Although the structures are wellcrystallized they often coexist with large amounts of
contaminants such as catalytic particles and amorphous carbon.

Fig. 4. The scheme of the production of carbon nanotubes the pulsed laser vaporization method

84
 6.2.3. Chemical vapor deposition (CVD)

While the arc discharge and pulsed laser vaporization can both produce well-
crystallized CNTs and fullerenes, these methods are limited by their production scale.
The chemical vapor deposition technique has been heavily used to synthesize carbon
fibers, filaments, and is now being widely used to produce CNTs. Various CVD
configurations have been developed and tested for the synthesis of CNTs, including
horizontal furnace, fluidized bed reactor, vertical furnace, and basic plasma enhanced
CVD [38]. In all cases CNTs are synthesized via hydrocarbon vapor decomposition in
direct contact with catalytic particles. The process involves heating the furnace tube
holding the catalytic material to a certain temperature, when the growth temperature is
reached, the carbon feedstock is introduced. Fig. 5 is a schematic of CVD with a
horizontal configuration, which is the most popular of the configurations. The
horizontal configuration is advantageous because there is no temperature gradient
within the heated zone. A variety of hydrocarbons in the form of liquid, gas, and solid
states can be employed as the source of carbon feedstock. In the case of liquid or solid
hydrocarbons, the vaporization process can first be achieved through the use of an
external evaporator. Various catalytic materials can be used in any state and can be
positioned inside the furnace or introduced from outside. It has been shown that CVD
can be ideally employed to grow a variety of CNTs including aligned carbon
nanotubes on preferred substrates, in powder form, thin or thick films, aligned or
entangled, straight or coiled [39]. The type of hydrocarbon, the catalyst type and size,
and the process temperature are the key parameter in carbon nanotube growth using a
CVD technique. For MWNT formation the growth temperature ranges from 600 to
900 0C [40]. Various light hydrocarbons such as methane, ethylene, and acetylene are
directly introduced in the CVD reactor. Heavier hydrocarbons such as benzene,
toluene, and xylene are introduced as liquids and evaporated during CVD heating [41].
For SWNTs higher temperatures are used, ranging from 900 to 1200 0C [39] and
methane is usually used as the carbon source. In addition, for SWNT synthesis a wide
range of catalysts are used including transition metals (Fe, Co, Ni), and solid
organometallocenes such as ferrocene, cobaltocene, nickelocene [42].

Fig. 5. The scheme of the production of carbon nanotubes the chemical vapor deposition method

Since the discovery of CNTs a large number of experimental works have been
devoted to understanding the growth mechanisms of CNTs using different methods
and parameters. Recently some numerical models have been created for understanding
the synthesis of CNTs dealing with the development of kinetic laws to simulate
growth process. The CNT growth model requires multistep processes that can predict
catalytic particle formation and the reactions of gas-phase species with the formed
85
nanoparticles. The ability to produce CNTs in large scale and with controlled
properties depends on the understanding of the growth pathways [43]. Well developed
models can provide rapid and important information for the growth of CNTs in a
certain process that can only be obtained through complex and careful experiments.
CVD synthesis of carbon nanotubes has been modeled, however, these works
represents only a fraction of experimental works. The growth models have been
attempted by various authors including models of catalytic nanoparticle formation
during CVD process [44, 45] and modeling of CNT growth [46, 47].
Large quantities of CNTs can be synthesized using CVD. Smalley and co-workers
[48] reported that a high pressure reactor can yield SWNTs in the order of ~ 10 g/day.
One of the principal disadvantages of CVD, however, is the long time required to
complete the synthesis process. Moreover, CVD is distinctive in that it affords
researchers the ability to synthesize CNTs with highly accurate controllable
morphologies directly on the desired substrates by using nanotemplates. The catalytic
particle size directly influences the CNT diameter which suggests that controlled
catalyst size will produce controlled diameter tubes [49]. However, production of
desirable substrates, such as nanotemplates, is time intensive as it requires various
anodization processes and the electrodeposition of the catalytic particles inside the
pores. Wu et al. reported a typical experimental process of 10 hours for fabricating
tiny ‘nano’ templates [50, 51].

6.3. Methods of investigation of carbon nanotubes

Among the various analytical for carbon nanotubes, transmission electron

microscopy (TEM) is the one by which carbon nanotubes were discovered when
observing carbon soot produced in an electric arc between graphite electrons [9].
In a transmission electron microscope, the electron wave mainly experiences two
basic processes during its propagation from the specimen to the image: interaction
with the specimen and modulation by the electron-optical lens system [52, 53]. During
the first process, the electron wave picks up the structure information of the specimen.
In the second process, the structure information is transferred in the microscope. Due
to the unique design of the TEM, the electron wave carrying the structure information
forms an electron-diffraction pattern (EDP) on the back focal plane of the objective
lens and an image on the imaging plane. Therefore, the EDP and the image can be
simultaneously observed in the TEM. By using an aperture at the back focal plane of
the objective lens to select the transmitted electron beam for imaging, a bright-field
(BF) image is formed. A dark-field (DF) image is created by using one of the
diffracted beams in the EDP for imaging. If a large aperture is selected to allow both
the transmitted electron beam and a set of diffracted beams to pass through, a high-
resolution TEM (HRTEM) image is formed. Mainly due to the spherical aberration
(Cs) of the objective lens in the microscope, the contrast of an HRTEM image is
seriously affected by the focusing conditions, and the image resolution is limited. As a
result, a HRTEM image is usually not straightforward to interpret and it is hard to
directly observe the atomic structure of carbon nanotubes (CTNs). Due to the recent
development of the Cs-corrected HRTEM [54], the resolution of the microscope has
dramatically improved to the sub-A level. It is worth remembering that to minimize
spherical aberration, the microscope is preferable operated art an acceleration voltage
lower than 120 kV to minimize the knock-on atomic displacement damage of the
CNTs by high-energy electrons, ad to increase the image contrast.
86
 Scanning electron microscopes (SEM) are frequently used to observe carbon
nanotubes due to their wider availability, relative ease of use, as well as simpler sample
preparation methods compared to TEM and scanning tunneling microscopy (STM). Both
nanotube powder samples as well as individual tubes on a bulk substrate (e.g., silicon
wafer) can be observed by SEM. A fine electron probe is scanned across the sample
surface while recording the current of the electrons emitted from the sample surface due to
the beam-surface interactions. The image is the intensity of the emitted electron current
versus beam location at the sample. An image can be formed by the secondary electrons
emitted from the sample surface as well as by the backscattered primary electrons. In
modern, field emission electron-source-based instruments, the energy filtering of emitted
electrons is also possible. Several electron detectors, including standard chamber
detectors, in – lens (or semi-in-lens) and backscattered detectors can be used in parallel.
The energy of the primary electrons is varied from 100 V to 30 keV. The secondary
electron imaging resolution of the state-of-the-art field emission source SEMs vary from
0.4 nm at 30 kV to 1.6 nm at 1 kV acceleration voltage. Accordingly, the walls of the
single and multiwall nanotubes cannot be observed with SEM. Similarly, whether small
tubes are isolated or bundled cannot be resolved.

6.4. Synthesis of carbon nanotubes in flame

In 1976, Endo observed the formation of carbon fibres in the pyrolysis of benzene
[2] and, in 1991, Iijima detected carbon nanotubes in graphite subjected to an arc
discharge [3], which was the beginning of the era of carbon nanotubes that, due to a
set of their unique properties, are of interest for both the applied and fundamental
sciences [4, 5].
Kennedy [6], Merchan-Merchan et al [7] has made a comprehensive survey of
investigations on the synthesis of carbon nanotubes in flames.
In a number of recent publications, results of the synthesis of carbon nanotubes in
a flame used as a heat source are presented. The production of carbon nanomaterials,
such as nanotubes and nanofibres, in a flame has a number of advantages over the
production of these materials with the use of electricity. First, in this case, there is no
need to use an expensive electric heat source. Second, the flame synthesis is in fact
simpler as compared to synthesis with the use of electricity [8].
The potential of the flames for synthesis of elongated carbon nanoforms was
demonstrated long before CNTs were discovered by Iijima in the process of
vaporization of graphite. The formation of elongated filamentous carbon on the
surface of a solid support inserted in a diffusion flame was reported in very early flame
studies. The formed filaments were hollow with metal granules at the tips. The
reported deposit structures closely resembled structures of carbon filaments formed by
CO and CH4 decomposition over metal catalysts [55]. In the same year when CNTs
were discovered by Iijima, Saito et al. [56] reported the formation of elongated
carbonaceous structures on the surface of a probe introduced in a sooting hydrocarbon
flame. The materials deposited on the wire probes were composed of elongated
structures that had lengths and diameters resembling the external morphology of what
we know today as CNTs or CNFs and was only observed in a certain flame position
using acetylene as a fuel. Howard et al. discovered the presence of multi-walled
carbon nanotubes coexisting with soot-like structures in condensable material
collected from a low-pressure premixed flame [57].

87
 Howard et al. showed the potential of the flame technology for the synthesis of carbon
nanotubes and fullerenes. These early findings were one of the contributing factors that
prompted scientists working in combustion to modify their existing combustion systems
and apply them to the synthesis of nanostructured materials. As a result, combustors of
various configurations have been applied for studying CNT synthesis that include co-flow,
flat flame, counter flow burners, and even the application of a simple paraffin wax candle
[58]. Table 1 contains a complilation of the different flame configurations that have
contributed to the research on CNT materials synthesis.
Various configurations of premixed, non-premixed (diffusion), partially premixed,
and inverse diffusion flames have been used as a hydrocarbon feedstock with a metal
catalyst for material synthesis applications.

Table 1
Flame configurations and generated CNT types [51]

88
 Сontinuation of Table 1

Yuan et al. studied the effects of fuel type, catalytic support type, its delivery
method, and residence time using a laminar co-flow diffusion flame for the synthesis
of MWNTs. The synthesis was performed using methane-air [59], ethylene-air, and
nitrogen-diluted ethylene diffusion flames [60] at atmospheric pressure conditions.
The catalytic supports used were in the form of wires and grids undergoing an
oxidation process, and grids pre-loaded with Co nanoparticles. The grid was pre-
oxidized using a lean premixed propane flame for about 5 s. The pre-oxidation
allowed for nanocatalytic particles to form on the grid’s surface before inserting it into
the flame for CNT synthesis. The cobalt pre-loaded grids were prepared by
electrodepositing the single Co nanoparticles on the surface. A catalytic support with a
chemical composition of Ni/Cr was introduced in the methaneeair flame system. The
meth-aneeair flame produced entangled MWNTs with a diameter range of 20-60 nm,
Fig. 6(a and b). It was shown that CNTs can grow based on the “base growth model”
mechanism as nickel oxide particles where found embedded inside some of the
nanotubes, Fig. 6(b).
In contrast to the exposure of the Ni-Cr wire in the meth-aneeair flame, the
insertion of the Ni-Cr wire in the ethyleneeair flame resulted in the formation of some
carbon fibers loaded with amorphous carbon particles, indicating the oxidation process
in the ethylene flame to be somewhat different from the methane flame and that the
shape of the catalytic support is essential.
The ethylene nitrogen-diluted diffusion flame synthesis performed using the pre-
oxidized stainless steel grid containing iron, chromium, and nickel oxide deposits on its
surface, resulted in structures similar to those grown in the methane-air flame on the
surface of a NiCr wire. That is, bundles of entangled and curved MWNTs with diameters
ranging from 10 to 60 nm coexisting with some carbon fibers, Fig. 6 (a and b).
A flame system formed using a nitrogen diluted ethylene diffusion flame
[C2H4]/[N2]=0.25 with the catalytic support prepared in the form of a stainless steel
grid with cobalt particles electrodeposited on its surface resulted in aligned MWNTs
with a monodisperse diameter size distribution, Fig. 6 (c and d). At a certain flame
position it was found that the introduction of N2 can decrease the flame temperature by
~ 300 0C. The authors claim that the nitrogen addition into the flame system was
found to be a key parameter to straighten the entangled tubes, likely due to lowering of
89
 the flame temperature resulting in
aligned MWNTs, Fig. 6 (c). In the
methane flame synthesis the
pyrolysis products were stable and
the pyrolytic carbon concentration
was much lower than in ethylene
flames.
The other most significant effect
observed from Xu’s experiments is
that for both, the binary and ternary
alloys, the lengths of the CNTs
decrease as the probe is positioned
closer to the flame tip. It was found
that the flame temperature and the
chemical species vary along the
axial and radial directions of the
flame. For the tested binary and
ternary alloys, flame zones were
identified for best synthesis of CNTs
within the radial range from 2 to 4.5
mm, although for some flame
heights this range was a bit com-
pressed or expanded due to flame
geometry.
Table 2 represents maximum
temperatures measured with a ther-
mocouple and species mole fractions
at the different flame heights. From
these measurements it can be
observed that the mole fractions of
the chemical species significantly
a and b) CNTs from an ethylene N2-diluted vary along the axial flame position.
flame on a pre-oxidized grid; c and d) CNTs from an Similar to the Xu et al. ex-
ethylene N2-diluted flame on a Co-coated grid. periments, it was shown that within
the flame environment there are
Fig. 6. SEM and TEM images of CNT from co-flow
flames at atmospheric pressures limits in the amount of gases such as
Adapted from Yuan et al. [59, 60] CO and H2 that can affect the
quantity and quality of the generated
CNTs [61]. The availability of H2 can promote carbon deposition by electronic
interactions with coadsorbates, in this case particularly CO. This occurs through
electronic charge donation from dissociated H2 to the CO ant bonding thereby
weakening the intermolecular bond while favoring metal carbon interaction [62, 63].
However, a heavy metal carbon interaction due to high concentration of H2 or CO can
prevent the catalyst to fully functionalize creating a blockage by the carbon molecules.
Vander Wal et al. [64] investigated the dependence of the structure of carbon
nanotubes formed in diffusion C2H2 flames on Al2O3, CaO, SiO2, and TiO2 substrates
with the use of Cu, Fe, and Ni as catalysts on the electron structure and chemical
composition of these flames. They noted that the electron interaction between the
metal-catalyst particles and the supporting substrate influences substantially the
density, homogeneity, and structure of the carbon nanotubes.
90
 Table 2

Gas-phase temperatures (thermocouple) and chemical spices (spontaneous Raman

spectroscopy) in the inverse diffusion flame at various flame locations [61]

Z (mm) Tmax (K) Co, % C2H2, % H, % Radial distance,

mm
6 1140 1.45 1.2 0 3.0
9 1140 2.3 2.6 1.0 2.0
12 1050 3.75 1.8 1.6 2.0
15 1045 1.5 0.0 0.35 2.0

6.4.1. Controlling MWNT growth in flames

The remarkable properties of CNTs allow for the development and identification
of emerging and novel potential applications. For certain applications it is required
that the CNTs are vertically aligned and free of metal particles and/or carbon by-
products. The ability to control CNT orientation, alignment, length, diameter and
diameter uniformity, purity, and internal structure is essential for the potential
fabrication of nano- and micro-electronic and nanoelectromechanic devices such as
capacitors, biological probes, field emission devices, electrocatalyst supports for use in
fuel cells, and anodes in Li-ion batteries, among others. For instance, CNTs due to
their remarkable property as electron emitters have been proposed for use in the
fabrication of a new generation of flat panel displays but the CNTs must first be
vertically aligned and grown on large flat surface/electrode areas [65]. Therefore, in
order for products based on CNTs to become a reality, an inexpensive synthesis
method is required that has the capability to produce CNTs in a controllable manner as
well as on desirable surface areas. Various control methods proposed for the synthesis
of vertically aligned CNTs are discussed in this section.
An efficient control method is required to improve uniformity and productivity of
flame-based synthesis and utilization of electromagnetic field control is one of the
most promising approaches. An effective control method must be applied to stabilize
growth of carbon nanomaterials with specified structure and morphology.
Recently, the electric field control was implemented by electrically isolating the
probe from the grounded support and operating it at floating potential mode (FPM) in
a counterflow flame medium [66]. Catalytic probes with similar physical and chemical
characteristics as previously used were employed to synthesize a variety of carbon
structures, Fig. 7. 73 % Ni + 73 % Cr + 10 % Fe wires with a diameter of 0.6-mm
were employed for the electric fields experiments. The probes were inserted in FPM
( 300 mV induced by flame generated ions and electron) at positions of 7.5, 8.5, and
9.5 mm from the edge of the fuel side nozzle. At 7.5 mm formation of a vertically
aligned carbon nanotube (VACNTs) layer averaging 40 microns in thickness is
observed, Fig. 7 (a and b). It was found that the thickness of the VACNT layer
decreases with the distance from the fuel nozzle. Fig. 7 (c-f) represent arrays of
nanotubes formed at 8.5 and 9.5 mm from the edge of the fuel nozzle; an average layer
thickness of 10 and 3 microns is measured for these positions. In order to study the
growth rates of the CNT arrays as a function of hydrocarbon species availability and
the effect of temperature within the flame sampling, CNT arrays synthesized on the
probe surface were mechanically disturbed before introducing them into the SEM, Fig.
7. The lengths of the CNT arrays formed in each tested flame position were obtained
by collecting several SEM images at different locations of the produced CNT layer.
91
 An average value of the CNT array thickness is obtained based on various collected
SEM images for each particular flame sampling.
The availability of carbon species near
the surface of the catalytic probe can be
controlled by repositioning the sampling
probe in the axial direction of the flame,
Fig. 5 (e and f). In the present set of
experiments the probes were exposed to the
flame medium for 15 minutes. Therefore,
for a sample position of Z=7.5, 8.5, and 9.5
mm an average layer thickness of 40, 9, and
3 mm were measured corresponding to the
growth rates of 2.7, 0.6, and 0.2 mm/min,
respectively. The variation of structure and
morphology of the formed carbon
nanomaterials is directly correlated to the
strong variation of temperature and
chemical composition present in the studied
flame region.
The distributions of several major
hydrocarbons (CH4, C2H2, CO, and C6H6)
that can contribute to the growth of carbon
deposits are shown in Fig. 7(e) along with
the temperature profile. It is easy to
conclude that all above components vary
significantly in the flame region of interest.
Thus, concentration of CH4 and C2H2
diminishes below 100 ppm for Z > 10.5
mm; the concentration of CO grows with Z
reaching its maximum at this point; C6H6 is
present in essential quantities only from 8
to 10 mm, maximizing at 9.5 mm. The
essential variations of temperature and
chemical composition are inevitable in a
variety of flame configurations. An
effective control method is necessary to
stabilize growth of carbon nanomaterials
with specified structure and morphology.
Fig. 7. Carbon nanostructures formed in various The electric field control method was
locations of opposed-flow oxy-flame. In the implemented and it shows to be an effective
presence of an electric field (FPM, Floating tool to control CNT growth.
Potential Method) the growth of vertically The introduction of the electric fields in
aligned carbon nanotubes is stabilized at all counterflow flames tended to stabilize the
tested positions. The distributions of several
major hydrocarbons species CNT growth resulting in uniformly
(CH4, C2H2, CO, and C6H6) that can contribute distributed CNT arrays on the probe’s
to the growth of carbon deposits are shown in surface. The disorganized CNT bundles,
(e) along with the temperature profile [51] nanofibers, helical coiled, and ribbon-like
structures growing in random directions
were not present when the probe was inserted in floating potential mode at any of the
tested flame positions.

92
 High-resolution TEM and SEM imaging analysis on the synthesized product
shows that the electric field method significantly improved the diameter uniformity of
the synthesized CNTs. The arrays composed of VACNTs shows a narrow distribution
of nanotubes diameter with average value of 35 nm. One of the interesting aspects
depicted in Fig. 7 (a-c) is a high-purity and well-ordered formation of nanotubes
showing that the electric field method can be successfully used to electrophoretically
keep the less ordered carbon black (soot particulates) from contaminating the
synthesized CNTs. The samples were never exposed to any kind of chemical and/or
physical cleaning. In some of the SEM images (Fig. 7) the CNTs look tilted. This is
due to the fact that arrays were disturbed and stripped off the substrates mechanically
in order to measure the array thickness under the SEM.
Various voltages have been employed in a flame electric field to control the
growth of the CNTs [67, 61]. Merchan-Merchan et al. applied voltages ranging from
0.3 to 25 V for studying the effect of the electric field on the synthesis of CNTs, Fig.
8. The increase of the applied voltage in a range from 0.3 to 1 V leads to a linear
increase of CNT array thickness while preserving layer purity and uniformity. The
effect of saturation is observed at voltages above 2 V. For voltages above 3 V, applied
electric fields begin to modify nanotube morphology and array uniformity, initially
generating helical nanotubes and then, at higher voltages, producing branched CNTs.
For applied voltages w3 V, straight as well as helically coiled spiral-like carbon
nanotubes were observed over the surface of the CNT layer (Fig. 8b).

a) from 0.3-2.0 V the arrays are composed of straight CNTs; b) from 2.0 to 5.0 V CNT arrays are
coiled and straight; c) from 5.0-12 V earlier branching of CNTs; d) from 12-25 V fully branched CNTs

Fig. 8. Influence of the applied voltage on morphological variation of the generated CNT arrays formed in
flame by a counterflow reactor

93
 Interestingly, these spiral-like structures grow very long and generally
perpendicular to the surface of the probe. As the applied voltage was increased, the
coil diameter tended to decrease (Fig. 8b). This effect can be interpreted using the
mechanical analogy of a spring whose diameter is reduced with an increase of the
applied axial force. The electrostatic force applied to the tip of the growing carbon
nanotube is acting as a tensile force producing a similar effect.
At 5 V, sprouting of the catalytic particle occurs which generates new potential
sites for CNT growth along the trunk of the existing CNT (Fig. 8c). At applied
voltages of 12 V, it is observed that the arrays are composed of structurally modified
CNTs. The CNTs achieved complete Y-branching, T-branching, and multi-branching
(Fig. 8d). It can be observed that the introduction of medium and high voltage
potentials can modify the internal and external morphology of the CNTs.
The application of an electric field was also studied by Xu and co-workers for
stabilizing CNT flame growth and controlling their morphology. The co-flow
configuration of an inverse diffusion methane flame was utilized by Xu et al. for those
studies [59]. In work [51], the effect of the electric field on CNT synthesis was tested
by introducing both negative and positive voltages. The results show that both types of
voltages significantly improved CNT alignment and growth rate effects, Fig. 9,
although negative volt-ages tended to yield synthesized CNT arrays that are more
compact and organized, Fig. 9(a and b). SEM analysis shows that the CNT arrays have
a thin layer of material covering the tips of the CNTs. This thin layer is composed of
catalytic particles corroborating that they were grown under the “tip-growth model”
mechanism. It is observed that under the influence of an electric field (negative
voltage) the thickness of the layer does not necessarily increase with an increase of
applied electric potential, Fig. 9(a and b). To the contrary, the thickness of the CNT
layer grown at the voltage potential of -2.0 V decreased by approximately 50 % from
the one grown at -0.24 V. The differences can be rationalized by consi-dering the
basic concept of the interaction between the CNT and electric field forces acting on its
tip. The electric force on the carbon nanotube tip can be affected by the amount of
catalytic material present at the tip of the CNTs, therefore, preventing the electric field
from effectively increasing the growth rates of the CNTs.

Fig. 9. Effect of voltage bias applied to a cylindrical catalytic probe inserted in a co-flow methane inverse
diffusion flame for the synthesis of CNT arrays.
Adapted from Xu et al. [61]
94
 Similar to the work of Merchan-Merchan et al. soot particulates were not observed
within the CNT arrays. The results of Xu et al. show that the introduction of voltage
bias on the substrate leads to aligning the CTNs, enhances the growth rates and
electrophoreti-cally keeps the soot particles from contaminating the CNT yield
compared to when the probe was in grounded mode.
Carbon nanotubes synthesized in the flame medium with the assistance of an
electric field have similar characteristics to CNTs synthesized in other electric-assisted
methods [61-63, 66, 68-70]. The elec-tric field method tends to affect the CNT
synthesis method by: (i) improving the diameter uniformity, (ii) stabilizing CNTs and
resulting in a more controllable and repeatable process, (iii) significantly improving
the growth rates, (iv) achieving higher purity of generated CNTs by repulsing other
carbon by-products, (v) forming CNT arrays with thickness controlled by utilizing
voltages of different strengths.

6.4.2. Synthesis single-walled carbon nanotubes in flame

Similar to the synthesis of MWNTs a combustion system tailored with an ideal

source of carbon, heat source, and appropriate cata-lytic material can result in the
production of single-walled carbon nanotubes. In the flame method the catalytic
precursors are generally introduced into the flame system in the gas-phase and
nucleate and condense to solidify into spherical metallic nano-particles. Flame
parameters can be used to obtain an appropriate flame environment that would allow
the formation of ideal sizes of catalytic particles for carbon nanotube inception and
growth. The available literature on the flame synthesis of SWNTs is scarce, in contrast
to flame synthesis of MWNTs, consisting of only a handful of experiments that have
been conducted on the synthesis of SWNTs. To some extent all products have
common morphological trends, even though they are synthesized in flames formed
using different burner configurations and conditions. These morphological trends
include: (i) SWNTs always coexist with metallic and/or soot particles, (ii) particles
often appear to be poisoned; even when ultra small catalytic particles, ideal for SWNT
inception, can be achieved, they can be heavily encapsulated with amorphous carbon
becoming inactive as catalysts for nanotubes, (iii) the presence of larger metallic
particles with very short SWNTs. In an effort to control these morphologies, various
flame and catalytic parameters can be employed. Recent works on flame synthesis of
SWNTs have shown that there are critical factors that influence the synthesis of
SWNTs, including: (a) type of reactant gas environment (concen-tration of gas
mixtures, gas additives, fuel types); (b) ratio of fuel to oxygen; (c) composition and
insertion rate of catalytic particle into the flame system; and (d) flame pressure (Table
3). The alteration of these flame and catalytic parameters can influence the source of
carbon and heat, catalytic particle formation, activation and deac-tivation of the
catalytic particle for SWNT synthesis.
The majority of the studies on the flame synthesis of SWNTs have been conducted
in the gas-phase. A continuous nano-structural change occurs through the flame
volume, from the formation of the bare catalytic particle at one part of the flame
volume to CNT inception and growth on the same catalytic particle in other parts of
the flame. Due to the structure of the flame as a non-isothermal medium, the
transformation can occur in milli-seconds. As mentioned above, SWNTs synthesized
in flames are often highly entangled, grow in a disorderly manner, and coexist with
catalytic particles and combustion by-products. In the final part of this contribution,
95
therefore, we focus on the flame products and challenges encountered in the flame
synthesis of SWNTs.

Table 3

Flame configurations and generated SWNTs

Recently, Vander Wal and co-workers [71, 72] showed the effects of flame and
catalytic parameters for controlling the growth of SWNTs, including the effect of
different reactive gas mixtures, catalytic particle composition and its introduction rate
using a premixed co-flow flame in the gas-phase. The introduction of pure CO for
hydrocarbon combustion resulted in the heavy encapsulation of the Fe nanoparticles
within the amorphous carbon, Fig. 10 (a). The larger iron catalytic particles are
observed as dark spots – in a matrix of amorphous carbon. The introduction of pure H2
in the flame system resulted in the formation of metallic particles with much cleaner
surfaces than those formed with pure CO, Fig. 10(b). However, in both cases carbon
SWNTs coexist with catalytic particles containing many times their diameter and of
variable sizes. The coexistence of both structures (filament and spherical-like) of
different sizes suggests that during tube growth the catalytic particle grows as well.
Particle surface growth may be due to a continuous absorption of additional metal
species in the vapor phase as they travel along the flame axis in the direction of the tip.
High CO concentrations lead to an amorphous carbon coating on the particle. The
amorphous carbon coating is likely responsible for the termination of tube growth by
blocking the diffusion of carbon species in the Fe particle. The carbon layer around the
particle limits the carbon diffusion through the catalytic particle which can result in a
complete encapsulation of the cata-lytic particles and deactivates the catalyst [73]. The
high H2 concentrations tend to have the opposite effect of high CO concentration, it
96
removes the surface carbon species from the catalytic surface too rapidly, therefore
slowing and even inhibiting SWNT growth [72]. Reacting mixtures with fixed
amounts of H2 þ He and added CO and C2H2 were also tested for the synthesis of
SWNTs. Experiments conducted with mixtures of CO/H2/He, without the presence of
acetylene, proved to be a superior gas mixture for SWNT formation. A reacting
mixture with molar composition of 0.25C2H2/0.27H2/0.75He resulted in the formation
of Fe nanoparticles with ideal sizes for SWNT inception. However, despite the
optimal size of the Fe particles, the majority of the metallic particles were deactivated
and did not initiate the growth of SWNTs as they became fully encapsulated with
carbon showing that the concentration of the gas mixtures plays a key role in SWNT
synthesis.
The synthesis of SWNTs in
flames is difficult as it requires
gas-phase introduction of the
catalyst to obtain the ultra small
catalytic particles ideal for CNT
inception. The growth of
SWNTs in the post-flame gases,
just above the burner’s nozzle,
can be described by the for-
mation of small discrete par-
ticles with a progression toward
CNT inception and growth in
the upper region of the flame.
Fig. 10. TEM images of condensed material collected This formation pathway is very
thermophoretically from rich premixed co-flow flames similar to that encountered in
with different reacting gas mixtures. Adapted from soot formation in flames. At the
Vander Wal et al. [71, 72] lower part of the flame small
unsaturated hydrocarbons tend
to form the young soot particles. Soot particles can further grow through the
coagulation and surface growth processes, and in the upper part of the flame the
appearance of “mature soot” as particles are agglomerated due to collisions as they
travel in the upper part of the flame [56, 74, 75]. The formation pathway of SWNTs in
flames closely follows that of less organized carbon particles, such as soot. This
formation pathway has also been observed in the flame synthesis of metal oxide
nanoparticles [76]. It is well known today that the precursors of crystallized carbon
clusters are high-molecular PAHs and aromers [77]. These molecules, which are
formed in the main oxidation zone of the flame, are direct precursors for fullerenic and
soot nanomaterials [78]. The system can yield both or one or the other depending on
flame conditions [79]. Vander Wal et al. show that the addition of acetylene in the
reacting mixture poisoned the catalytic particle with amorphous carbon preventing the
activation of the catalytic particle for SWNT growth. In contrast, the addition of CO
significantly contributes to the synthesis of SWNTs. The process of the flame
synthesis for SWNTs is quite complex as the proper flame parameter must be
employed to achieve a suitable size catalytic particle with enough carbon species for
particle activation, inception, and their continued growth on their way through the
flame. An earlier abundance of carbon species will poison the particle and prevent
CNT inception earlier on in the flame volume.
The introduction of low, moderate, and high densities of ferro-cene was used by
Vander Wal [71] to study the synthesis of SWNTs in flames. Similar to soot formation
97
in flames, a high number of soot precursors in the post flame can result in a very sooty
upper part of a flame. Therefore, it is anticipated that the presence of a high or low
density of catalytic precursors needed for SWNT inception can significantly influence
growth. Fig. 11 (a and b) represents products synthesized at various ferrocene rates
ranging from ~8, 20, and 40 mg/min, respectively. Samples collected thermo-
phoretically from inside the flame show that the size of the average Fe particle
increases as the rate of introduced ferrocene into the system is increased, Fig. 11(a-c).
In any synthesis process it is well estab-lished that a key step in the growth of SWNTs
is the ability to first form
the catalytic particle
with optimal size for
CNT inception and gro-
wth. The catalytic par-
ticle diameter must
match that of a fullerene
to form the hemispheri-
cal end-cap, therefore
ultra small particles are
needed for SWNT in-
ception. In the work of
Vander Wal et al. the
ferrocene introduction
Fig. 11. TEM images of condensed material collected rate of ~8 mg/min resul-
thermophoretically from rich premixed co-flow flames with various ted in relatively smaller
rates of catalytic particle inception. Adapted from Vander Wal and Fe nanoparticles which
Lee [71] allowed SWNT incep-
tion but the lengthening was stopped due to an excess of Fe particles, Fig. 11(a). The
supply of ferrocene at higher rates, however, can result in Fe particles with larger sizes
due to an abundance of Fe, Fig. 11(a-c). The studies of Janzen et al. on the formation
of Fe2O3 nanoparticles in a low-pressure H2/O2/Ar flame, support the above results,
showing that iron-pentacarbonyl decomposes rapidly upon exposure to the flame with
an increase of particle size in the flame axial direction, the growth of the particles most
likely occurs through coagulation of the iron resulting from this decomposition [76].
The rate of particle deactivation after its inception for SWNT growth decreased as the
rate of ferrocene is increased, creating Fe excess, and resulting in SWNTs of very
short lengths, Fig. 11(c).
The introduction of low, moderate, and high density carbon feedstock for the
synthesis of SWNTs in flames was studied by Height and co-workers at various
heights above burner (HAB) of a flat flame. The carbon feedstock was controlled by
varying the equivalence ratios for the formation of SWNTs in the flame medium [80].
The experiments were conducted in a low pressure, 6.7 kPa (50 Torr), premixed
acetylene/oxygen/15 Mol% argon flame doped with 6100 ppm iron-pentacarbonyl
vapor as the catalytic precursors [81]. Strategic experiments were conducted to
understand the effect of equivalence fuel ratios on SWNT formation. A nano-structure
transformation was observed with the change of equiv-alence ratio. Remarkable
nanostructure transition is observed to take place in a flame zone above the burner
nozzle (70 mm) with the application of a wide range of equivalence ratios. Within the
range of B (equivalence ratio) from 1.4 to 1.5 the presence of discrete spherical shaped
particles is observed. The increase of equivalence ratio from 1.6 to 1.8 resulted in a
nanotube ‘formation window‘. The continuing increase of the equivalence ratio from
98
1.9 to 2.0 resulted in the formation of only combustion by-products such as soot-like
structures. From all of the tested equivalence ratios a ‘formation window’ for nanotube
synthesis was achieved between 1.5 and 1.9. Although it was found that the best
conditions suitable for nanotube synthesis were achieved in a narrower range of the Ø
at 1.6 and 1.7 (Fig. 12).
Fig. 13 represents a schematic interpre-
tation of various processes present at
various heights above burner (HAB) in the
low-pressure flame medium. From Fig. 13
it is evident that within the nanotube
‘window’ the external morphology of the
collected materials is not identical. The
morphology of the CNT in the upper limit
of the nanotube ‘window’ is affected by the
sooting limit. The amount of filament
structures completely disappears as it is
replaced by fully agglomerated soot
particles in the upper region of the equi-
valent ratio range, Fig. 13. The experiments
also show that an increase in the equiva-
lence ratio significantly enhances the
availability of carbon feedstock. Although,
it is evident that high equivalence ratios
promote the formation of soot-like struc-
tures rather than filamental structures.
The variation of equivalence ratio in
Height’s contribution shows a drastic
change in nanostructural morphology. The
various morphologies of the synthesized
structures are evident in the appearance of
discrete particles, SWNTs and soot-like
structures as the ø is increased.
The nanostructure progression along the
axis-line in the post-flame region of a flame
formed at a fixed equivalence ratio was also
studied by Height and co-workers in the
same contribution. Material was collected
thermophoretically from a single flame (Ø
of 1.6) in the flame axial direction at the
HABs of 20, 30, 40, 50, 60, and 70 mm. It
was found that the flame zone within the
range of 40e70 mm is ideal for CNT
synthesis. The external morphology of the
collected materials is not identical within
the 40-70 mm HAB. The length and the
quantity of the SWNTs increase with flame
height. In the lower zone (40 mm HAB) Fig. 12. Schematic interpretation of various
the SWNTs are few but clean of catalytic processes present at various heights above
particles. As it is approached the upper burner in a flat flame for SWNT synthesis.
zone of the flame in the direction of the 70 Adapted from Height et al. [80]

99
 mm HAB the CNTs get longer and are accompanied by metallic and soot-like
particles. Height et al. suggested that the dominant mechanism for the SWNT trans-
formation is by coalescence of the material condensed in the flame gases as the flame
temperature decreases from a peak temperature of w1800 K at 5 mm HAB followed
by a gradual and linear falloff to w1500 K over the following 75 mm HAB. At the
same time the same effect is responsible for causing the SWNTs along the flame to
become highly tangled and, interconnected in a web-like form, and in some cases to be
accompanied by metallic particles and/or combustion by-products (Fig. 13).
Fig. 13 represents a schematic interpre-
tation of various processes present at various
heights above burner (HAB) in the low-
pressure flame medium. From Fig. 13 it is
evident that within the nanotube ‘window’
the external morphology of the collected
materials is not identical. The morphology
of the CNT in the upper limit of the na-
notube ‘window’ is affected by the sooting
limit. The amount of filament structures
completely disappears as it is replaced by
fully agglomerated soot particles in the
upper region of the equivalent ratio range,
Fig. 13. The experiments also show that an
increase in the equivalence ratio signifi-
cantly enhances the availability of carbon
Fig. 13. Schematic interpretation of various feedstock. Although, it is evident that high
processes present at various HAB in the flame. equivalence ratios promote the formation of
The upper box gives a sense of the main soot-like structures rather than filamental
process ‘zones’ and their relative sequence in
the flame. The vertical gray band at 5 mm
structures.
HAB represents the luminous flame zone. The variation of equivalence ratio in
Adapted from Height et al. [80] Height’s contribution shows a drastic change
in nanostructural morphology. The various
morphologies of the synthesized structures are evident in the appearance of discrete
particles, SWNTs and soot-like structures as the ø is increased.
The nanostructure progression along the axis-line in the post-flame region of a
flame formed at a fixed equivalence ratio was also studied by Height and co-workers
in the same contribution. Material was collected thermophoretically from a single
flame (Ø of 1.6) in the flame axial direction at the HABs of 20, 30, 40, 50, 60, and 70
mm. It was found that the flame zone within the range of 40e70 mm is ideal for CNT
synthesis. The external morphology of the collected materials is not identical within
the 40-70 mm HAB. The length and the quantity of the SWNTs increase with flame
height. In the lower zone (40 mm HAB) the SWNTs are few but clean of catalytic
particles. As it is approached the upper zone of the flame in the direction of the 70 mm
HAB the CNTs get longer and are accompanied by metallic and soot-like particles.
Height et al. suggested that the dominant mechanism for the SWNT trans-formation is
by coalescence of the material condensed in the flame gases as the flame temperature
decreases from a peak temperature of w1800 K at 5 mm HAB followed by a gradual
and linear falloff to w1500 K over the following 75 mm HAB. At the same time the
same effect is responsible for causing the SWNTs along the flame to become highly
tangled and, interconnected in a web-like form, and in some cases to be accompanied
by metallic particles and/or combustion by-products (Fig. 13).
100
 The results of the study of a flat diffusion propane-oxygen flame stabilized on an
opposed-jet burner at atmospheric pressure are presented in [82, 83]. Two opposed
flows formed the flat flame, Fig. 14. The flame was surrounded by an external
nitrogen flow supplied from the burner matrices. A solution of catalyst [Fe(CO)5 or an
alcohol solution of nickel nitrate] was sprayed by an ultrasonic nebulizer and delivered
through a metal nozzle into the flame from the side of the fuel. The conditions of the
experiment are shown in Table 4.

Experiment parameters
Catalizators: Correlation С3Н8:О2 =2.1-1.5
– Solution Fe(CO)5, Flame temperature - Т=900-1150 оС
– Spirit Ni(NO3)2×6H2O, Rate of С3Н8 – 130-1000 cm3/min
– Nichrome wire, composition: Rate of О2 – 130-1000 cm3/min
52% Ni Rate of N2 – 50-200 cm3/min
18 Cr + 26 % Fe + 2.1% Al + 1.3 % Si

Fig. 14. Scheme and photos of torch on counter stream for obtaining of nanotubes

The consumption of the catalyst was determined by its weight before and after the
experiment. The maximum flow rate of the catalytic solution was 0.9 cm3/min, and its
minimum flow rate was 0.3 cm3/min. For the catalysts, Fe(CO)5 and an alcohol
solution (Ni(NO3)26H2O were used, and a Ni-Cr alloy wire of diameter 0.7 mm was
used as a substrate for growth of carbon nanotubes and metal nanoparticles. The
regions of effective synthesis of nanomaterials in a counterflow flame were
determined (Fig. 15).

101
 Table 4

Experimental conditions

Catalyst Flow rate Gas flow rate, cm3 /min Flame

of catalyst, C2H2 O2 N2 C3H8 temperature, ◦C
cm3 /min
Fe(CO)5 0.05 130-210 100-170 50-200 - 1700-1850
Ni(NO3 )2 · 0.05 - 160-210 50-200 70-160 900-1 150
6H2 O

Some of the samples obtained are

presented in the electron-microscopic
photographs. The metal particles,
shown in Fig. 16a, have sizes from 30
to 70 nm. These particles were also ob-
tained in the «fuel zone» of a counter-
flow flame at a distance of 6 mm from
its front in the process of atomization
of the 20 % alcohol solution
(Ni(NO3)26H2O used as the catalyst;
the characteristic temperature of the in-
dicated zone is 400-500 oC. Braids of
carbon nanotubes have been detected in
samples under the same conditions of
combustion and catalyst-supply (Fig.
16b). Fig. 16 presents an electron-mic-
roscopic photograph of a sample ob-
1) formation of metal nanoparticles; tained in a counter-flow flame with the
2) formation of carbon nanotubes;
3) formation of oxides. T, oC; R, cm
use of a Ni-Cr alloy wire of diameter
0.7 mm as a catalyst. The wire was positi-
Fig. 15. Temperature profile of a counterflow flame oned at a distance of 3 mm from the
in a diffusion burner flame front in the «fuel zone» with a cha-
racteristic temperature of 1100-1200 oC.

a b

a-The catalyst is a 20 % alcohol solution Ni(NO3)26H2O (aerosol); b-Electron-microscopic

photograph of a sample. The catalyst is a Ni–Cr alloy wire

Fig. 16. Electron-microscopic photographs of carbon-nanotube braid

102
 It is evident from Fig. 16 that the samples contain soot agglomerates, among which
metal particles are encountered. It was found that under certain experimental
conditions, well-ordered bundles of carbon nanotubes 20-30 nm in diameter formed.

6.5. Properties of carbon nanotubes

Carbon nanotubes exhibit remarkable thermal, mechanical and electrical properties

which are associated with their unique struc-ture, high aspect ratio, and quantum size
effects. In recent years, a number of theoretical and experimental works have been devoted
to study the properties of CNTs, Table 5. The superior properties of nanotubes classify
them as the material of the 21st century. It appears that everyday the development of new
nano- and micro-devices based on CNTs is identified and proposed.
One on the main features responsible for the unique properties of CNTs is the
perfection of their structure e nanotubes have almost zero body defects, although,
CNT defects depend highly on the method employed for synthesis. For instance, high
strength steel typically fails at approximately 1 % of its theoretical breaking strength
mostly due to structural defects. In contrast, CNTs attain values very close to their
theoretical predictions principally due to the few structural defects present.

Table 5

Properties of CNTs in comparison with other common materials [51]

Material Young’s modulus Tensile strength Electrical Thermal conductivity

(GPa) (GPa) resistivity (U m) (W/m K)
SWNTa 1054 150 10-6 1750-5800
SWNTa bundle 1054 75 0.1 x 10-3 35
MWNTa 1200 (270-950 GPa 150 (11-63 GPa 10-6 > 3000a
the outermost outer layer)
layer)
Diamonda 1000 1.2 1.0 x 1012 to 1.0 1000-2600
x 1014
Steela (plain 200 0.4 12.0 x 10-8 to 43
carbon low 170 x 10-8
alloy)
Coppera 110 0.413b 1.70 x 10-8 to 390
2.65 x 10-8
0.2067c
a
Room temperature conditions
b
Hard drawn
c
Annealed

The unique properties of CNTs open the door for an endless variety of CNT
applications and drive the search for improved methods to produce CNTs. It is worth
noting that the property values in Table 1 are the results of both theoretical and
experimental studies. Experimental studies on the measure-ments of CNT properties
have been limited by the low quality of available samples. For instance poor structural
qualities of CNTs can strongly enhance phonon scattering and greatly reduce thermal
conductivity [84]. It is important to note that the properties of the CNTs depend on
their physical form. For instance, it has been shown by Fisher et al. that the
longitudinal electrical conductivity of a single SWNT bundle is w50e150 times greater
than the conductivity of a mat sample [27]. Earlier on, measurements of properties
were obtained using CNTs synthesized in arc discharge and pulsed laser vaporization.
103
When CNTs are produced in these high energetic media (~ 3000 K) they exhibit
superior internal and external morphologies, well-crystallized and straight CNTs could
be obtained significantly decreasing their structural defects. Additionally, recent
research has shown that CNTs also exhibit superhydrophobic behavior over liquids
containing a range of pH values [85-87]. Table 1 lists properties of CNTs in
comparison with other common materials [16, 26, 39, 40, 88-92].
In [93], the synthesis of carbon
nanotubes and nanofibres on a
Ni(NO3)2 film deposited on a metal
substrate with the use of a counter-
flow ethylene flame as a heat
source was investigated. Puri et al.
have developed, on the basis of
numerous experimental data on
synthesis of carbon nanotubes in a
flame, a model of formation of
these nanotubes with participation
of nanoparticles of the catalysts
(Fe, Ni, Co) [87, 94-95], Fig. 17, a
schematically described of the
CNT formation process and in Fig.
4, b are shown schematic diagram
of the CNT growth model by Puri
and et al [94]. In this article [94]
was underlined that the chemistry
includes PAH species but soot was
not modeled, since it has negligible
presence in the flame.
In accordance with this model,
Fig. 17. (a) Schematic diagram of the CNT formation
nucleation of carbon nanotubes
process; (b) Schematic diagram of the CNT growth begin when the density of carbon
model [94] atoms on the surface of the catalyst
particles increases due to the
addition of the indicated atoms from the outer shell of these particles. However, once
the nucleation and growth of carbon nanotubes begins, carbon atoms are transported
through the nanoparticles mainly due to their diffusion. This diffusion decreases with
time because of the decrease in the concentration of the carbon atoms.

6.6. Application of carbon nanotubes

Numerous potential applications of these novel carbon nanomaterials include

nanoconductors and nanoswitches, nano-mechanical devices, composites with unique
mechanical and electromagnetic properties that can be used as field emission tips [65,
96-98], storage of hydrogen and natural gas [99, 100], electromechanical sensors
[101], as electrodes in electrochemical capacitors [102, 103], important components in
batteries [104] and fuel cell devices [105, 106], etc. CNTs have thermal, electrical, and
mechanical proper-ties (Young’s modulus and tensile strength), that are
extraordinarily high compared to those materials that were thought to have reached the
mechanical, electric, and thermal property limits.
104
 The high mechanical strength of carbon nanotubes makes them almost ideal
candidates for use as force sensors in scanning probe microscopy (SPM). They
provide higher durability and the ability to image surfaces with a high lateral
resolution, the latter being a typical limitation of conventional force sensors (based on
ceramic tips). The idea was first proposed and tested by Dai et al. [107] using c-
MWNTs. It was extended to SWNTs by Hafner et al. [108], since small-diameter
SWNTs were believed to give higher resolution than MWNTs due to the extremely
short radius of curvature of the tube end. However, commercial nanotube-based tips
(such as those made by Piezomax, Middleton, WI, USA) use MWNTs for processing
convenience. It is also likely that the flexural modulus of a SWNT is too low, resulting
in artifacts that affect the lateral resolution when scanning a rough surface. On the
other hand, the flexural modulus of a c-MWNT is believed to increase with the
number of walls, although the radius of curvature of the tip increases at the same time.
Whether based on SWNT or MWNT, such SPM tips also offer the potential to be
functionalized, leading to the prospect of selective imaging based on chemical
discrimination in chemical force microscopy (CFM). Chemical function imaging using
functionalized nanotubes represents a huge step forward in CFM because the tip can
be functionalized very specifically (ideally only at the very tip of the nanotube, where
the reactivity is the highest), increasing the spatial resolution. The interaction between
the chemical species present at the end of the nanotube tip and the surface containing
chemical functions can be recorded with great sensitivity, allowing the chemical
mapping of molecules [109].
Current nanotube-based SPM tips are quite expensive; typically ≈ 450 $/tip
(Nanoscience Co., 2005). This high cost is due to processing difficulties (it is
necessary to grow or mount a single MWNT in the appropriate direction at the tip of a
regular SPM probe, Fig. 18), and the need to individually control the tip quality. The
market for nanotube SPM tips has been estimated at ≈ 20 M$/year.

Fig. 18. Scanning electron microscopy image of a carbon nanotube (MWNT) mounted onto a regular
ceramic tip as a probe for atomic force microscopy (modified from [110])

Carbon-based materials make good supports in heterogeneous catalytic processes

due to their ability to be tailored to a specific need: indeed, activated carbons are
already currently employed as catalyst supports due to their high surface areas, their
stability at high temperatures (under nonoxidizing atmospheres), and the possibility of
controlling both their porous structure and the chemical nature of their surfaces [111].

105
CONTROL QUESTIONS

1. Can you tell about history of carbon nanotubes?

2. What kind of methods of synthesis carbon nanotubes do you know?
3. Can you describe the method of synthesis carbon nanotubes in the arc vaporization of graphite?
4. Which type of methods of investigation carbon nanotubes do you know?
5. What do you think of synthesis of carbon nanotubes in flame?
6. How do you what type of methods of obtaining of carbon nanotubes is better?
7. What kind of properties has carbon nanotube?
8. Can you describe of physical properties of carbon nanotubes?

REFERENCES

1. Wang, X.; Li, Qunqing; Xie, Jing; Jin, Zhong; Wang, Jinyong; Li, Yan; Jiang, Kaili; Fan, Shoushan
(2009). "Fabrication of Ultralong and Electrically Uniform Single-Walled Carbon Nanotubes on
Clean Substrates". Nano Letters. V. 9, № 9. P. 3137-3141.
2. Gullapalli, S.; Wong, M.S. (2011). "Nanotechnology: A Guide to Nano-Objects". Chemical
Engineering Progress 107 (5): 28–32.
3. Fenelonov V.B. Poristyy uglerod. – Novosibirsk: Izd-vo kataliza SO RAN, 1995. – 518 s, [in
Russian]
4. T.V. Hughes, C.R. Chambers: Manufacture of Carbon Filaments, US Patent 405480 (1889)
5. P. Schützenberger, L. Schützenberger: Sur quelques faits relatifs à l’histoire du carbone, C. R. Acad.
Sci. Paris 111, 774–778 (1890), [in French]
6. Rakov E.G. Nanotrubki i fullereny: uchebnoye posobiye. – M.: Logos, 2006. – 376 s, [in Russian]
7. L.V. Radushkevich, V.M. Lukyanovich: O structure ugleroda, obrazujucegosja pri termiceskom
razlozenii okisi ugleroda na zeleznom kontakte, Zurn. Fis. Chim. 26, 88–95 (1952), [in Russian]
8. M. Endo. Grow carbon fibers in the vapor phase. Chem Technol 1988:568–76.
9. S. Iijima: Helical microtubules of graphite carbon, Nature 354, 56–58 (1991)
10. S. Iijima, T. Ichihashi: Single-shell carbon nanotubes of 1nm diameter, Nature 363, 603–605 (1993)
11. Ebbesen T.W., Ajayan P.M. Large-scale synthesis of carbon nanotubes // Nature. – 1992. – Vol. 358.
– Р. 220-223.
12. Rode А. V., Gamaly Е. G., Luther-Davies В. Formation of cluster-assembled саrbon nano-foam bу
high-repetition-rate laser ablation // Appl. Phys. А. – 2000. – Vol. 70. – P. 135-142.
13. Endo М., Takeuchi К., Kobori К., et. al. Pyrolytic сагbon NT from vapor-grown саrbon fibers //
Саrbon. – 1995. – Vol. 33, № 7. – Р. 873-880.
14. Zeng Н., Zhu L., Нао G., Sheng R. Synthesis of various forms of carbon NT bу АС ARC discharge
// Carbon. – 1998. – V. 36. – № 3. – P. 259-266.
15. Guo Т., Nikolaev Р., Rinzler А. G., et. al. Self-Assembly of tubular fullerenes // J. Phys. Chern. –
1995. – V. 99. – P. 10694-10701.
16. Thess A., Lee R., Nikolaev P., et. al. Crystalline ropes of metallic carbon nanotubes // Science. 1996.
V. 273. P. 483-487
17. Hsin Y L, Hwang K C, Chen F R, Kai J J, Production and in-situ metal filling of carbon nanotubes in
water. Adv Mater 2001;13:830-5.
18. Wang SD, Chang MH, Lan K, Wu CC, Cheng JJ, Chang HK. Synthesis of carbon nanotubes by arc
discharge in sodium chloride solution, Carbon 2005; 43: 1792-1795
19. Beck M T, Dinya Z, Keki S, Papp L. formation of C60 polycyclic aromatic hydrocarbons upon
electric discharge in liquid toluene Tetrahedron. Carbon 1993; 49:285-90
20. Sano N, Nakano J, Kanki T. Synthesis of single-walled carbon nanotubes with Nan horns by arc
discharge in liquid nitrogen. Carbon 2004; 42: 686-688.
21. Ishigami M, Cumings J, Zettl A, Chen S. “A simple method for continues the production of carbon
nanotubes”. Chem. phys let. 2000; 319: 457-469
22. Antisari MV, Marazzi R, Krsmanovic R. Synthesis of multiwall carbon nanotubes by electric arc
discharge in liquid environments. Carbon 2003; 41: 2393-2401.
23. Lange H, Sioda M, Hoszko A, Zhu Y, Kroto HW, Walton D. nano carbon production by arc
discharge in water. Carbon 2003; 41: 1617-23.
24. Imasaka K, Kanataka Y, Ohshiro Y, Suehiro J, Hara M. production of carbon Nano onion and
nanotubes using an intermittent arc discharge in water, Thin solid films, 2005; 506-507: 250-254
25. Zho H W, Li. X S, Jiang B, Xu CL, Zho YF, Wu DH, et al. Formation of carbon nanotubes in water
by the electric-arc technique. Chem. Phys. Let, 2002; 366:664-9
26. J. Hone, M. Whitney, C. Piskoti, A. Zettl. Thermal conductivity of single-walled carbon nanotubes //
Phys Rev B. 1999. V. 59. P. 2514-2516

106
27. Y. Ando, X. Zhao, T. Sugai, M.Kumar. Growing carbon nanotubes // Mater Today 2004. V. 7. P. 22-
29
28. H.W. Kroto, J.R. Heath, S.C. O’Brien, R.F. Curl, R.E. Smalley. C60: buckminsterfullerene // Nature
1985. V. 318 P. 162-3
29. T. Guo, M.D. Diener, Y. Chai, M.J. Alford, R.E. Haufler, S.M. McClure, et al. Uranium stabilization
of C28: a tetravalent fullerene // Science 1992. V. 257. P. 1661-4
30. S. Bandow, S. Asaka, Y. Saito, A.M. Rao, L. Grigorian, E. Richter, et al. Effect of the growth
temperature on the diameter distribution and chirality of single-wall carbon nanotubes // Phys Rev
Lett. 1998. V. 80. P. 3779-82
31. R. Sen, Y. Ohtsuka, T. Ishigaki, D. Kasuya, S. Suzuki, H. Kataura, et al. Time period for the growth
of single-wall carbon nanotubes in the laser ablation process: evidence from gas dynamic studies and
time resolved imaging // Chem Phys Lett. 2000. V. 332. P. 467-73
32. H. Kataura, Y. Kumazawa, Y. Maniwa, Y. Ohtsuka, R. Sen, Suzuki S, et al. Carbon. 2000. V. 38. P.
1691-1697
33. H. Kataura, A. Kimura, Y. Ohtsuka, S. Suzuki, Y. Maniwa, T. Hanyu, et al. Formation of thin
single-wall carbon nanotubes by laser vaporization of Rh/ Pdegraphite composite rod // Jpn J Appl
Phys. 1998. V. 37. P. 616-618
34. Y. Avigal, R. Kalish. Growth of aligned carbon nanotubes by biasing during growth // Appl Phys
Lett 2001. V. 78. P. 2291-2293
35. S. Bandow, A.M. Rao, K.A. Williams, A. Thess, R.E. Smalley, P.C. Eklund Purification of single-
wall carbon nanotubes by microfiltration // J Phys Chem B. 1997. V. 101. P. 8839-8842
36. I.W. Chiang, B.E. Brinson, A.Y. Huang, P.A. Willis, M.J. Bronikowski, J.L. Margrave, et al.
Purification and characterization of single-wall carbon nanotubes (SWNTs) obtained from the gas-
phase decomposition of CO (HiPco Process). J Phys Chem B. 2001. V. 105. P. 8297-8301
37. H. Ishii, H. Kataura, H. Shiozawa, H. Yoshioka, H. Otsubo, Y. Takayama, et al. Direct observation
of TomonagaeLuttinger-liquid state in carbon nanotubes at low temperatures // Nature. 2003. V. 426.
P. 540-544
38. K.B.K. Teo, C. Singh, M. Chhowalla, W.I.Milne. Catalytic synthesis of carbon nanotubes and
nanofibers // Encyclopedia of nanoscience and nanotechnology. 2003. P. 1-22
39. J.E. Fischer, H. Dai, A. Thess, R. Lee, N.M. Hanjani, D.L. Dehaas, et al. Metallic resistivity in
crystalline ropes of single-wall carbon nanotubes // Phys Rev B. 1997. V. 55. P. 4921-4924
40. M. Meyyappan. Carbon nanotubes science and applications // 1st ed. CRC-Press. 2004
41. M. Maghrebi, A.A. Khodadadi, Y. Mortazavi, S. Mhaisalkar. Detailed profiling of CNTs arrays
along the growth window in a floating catalyst reactor // Appl Surf Sci. 2009. V. 255. P. 7243-7250
42. C.N.R. Rao, R. Sen, B.C. Satishkumar, A. Govindaraj. Large aligned-nanotube bundles from
ferrocene pyrolysis. Chem Commun. 1998. V. 15. P. 1525-1526
43. S. Fang, M. Zhang, A.A. Zakhidov, R.H. Baughman. Structure and process-dependent properties of
solid-state spun carbon nanotube yarns // J. Phys. Condens. Matter. 2010. V. 22. P. 334221
44. K. Kuwana, K. Saito. Modeling ferrocene reactions and iron nanoparticle formation: application to
CVD and flame synthesis of carbon nanotubes // Proc Combust Inst 2007. V. 31. P. 2625-2531
45. K. Kuwana, K. Saito. Modeling CVD synthesis of carbon nanotubes: nanoparticle formation from
ferrocene // Carbon. 2005. V. 43. P. 2088-2095
46. A.C. Lysaght, W.K.S.Chiu. Modeling of the carbon nanotube chemical vapor deposition process
using methane and acetylene precursor gases // Nanotechnology. 2008. V. 19. P. 165607
47. M. Grujicic, G. Cao, B. Gersten. Reactor length-scale modeling of chemical vapor deposition of
carbon nanotubes // J Mater Sci. 2003. V. 38. P. 1819
48. M.J. Bronikowski, P.A. Willis, D.T. Colbert, K.A. Smith, R.E. Smalley. Gas-phase production of
carbon single-walled nanotubes from carbon monoxide via the HiPco process: a parametric study // J
Vac Sci Technol A. 2001. V. 19. P. 1800-1805
49. H. Ago, T. Komatsu, S. Ohshima, Y. Kuriki, M. Yumura. Dispersion of metal nanoparticles for
aligned carbon nanotube arrays // Appl Phys Lett. 2000. V. 77. P. 79-81
50. Q. Wu, Z. Hu, X.Z. Wang, Y. Yang, Y.Chen. Porous alumina template in preparation of one-
dimensional novel Nanomaterials // Chin J Inorg Chem. 2002. V. 18. P. 647-653
51. W. Merchan-Merchan, A.V. Saveliev, L. Kennedy, W.C. Jimenez. Combustion synthesis of carbon
nanotubes and related nanostructures // Progress in Energy and Combustion Science 2010. V. 36. P.
696-727
52. A. Jorio, G. Dresselhaus, M.S. Dresselhaus. Carbon nanotubes. Advanced Topics in the Synthesis,
Structure, Properties and Applications // Springer-Verlag Berlin Heidelberg. 2008. P. 720
53. H. Jiang, F.H. Li, E.I. Kauppinen: Springer, Netherlands, 2004.
54. M. Haider, S.Uhlemann, E. Schwan, H. Rose, B. Kabius, K. Urban: Electron microscopy image
enhanced // Nature. 1998. V. 392. P. 768
55. J.M. Singer, J. Grumer. Seventh symposium (international) on combustion. Pittsburgh: The
Combustion Institute. 1959. P. 559
107
56. K. Saito, A.S. Gordon, W.F. Stickle. A study of the early history of soot formation in various
hydrocarbon diffusion flames // Combust Sci Technol. 1991. V. 80. P. 103-119
57. S. Löffler, Ph. Löffler, P. Weilmuenster, K.H. Homann. Growth of large ionic polycyclic aromatic
hydrocarbons in sooting flames // In: Bockhorn H, editor. Soot formation in combustion. Springer
Verlag; 1992. P. 66-82.
58. J.B. Howard, K.D. Chowdhury, J.B. VanderSande. Carbon shells in flames // Nature. 1994. V. 370.
P. 603
59. L. Yuan, K. Saito, C. Pan, F.A. Williams, A.S. Gordon. Nanotubes from methane flames // Chem
Phys Lett. 2001. V. 340. P. 237-241
60. L. Yuan, K. Saito, W. Hu, Z. Chen. Ethylene flame synthesis of well-aligned multi-walled carbon
nanotubes // Chem Phys Lett. 2001. V. 346. P. 23-25
61. F. Xu, X. Liu, S.D. Tse. Synthesis of carbon nanotubes on metal alloy substrates with voltage bias in
methane inverse diffusion flames // Carbon. 2006. V. 44. P. 570-577
62. R.L. Vander Wal. Flame synthesis of substrate-supported metal-catalyzed carbon nanotubes // Chem
Phys Lett. 2000. V. 324. P. 217-223
63. R.L. Vander Wal, L.J. Hall, G.M. Berger. Optimization of flame synthesis for carbon nanotubes
using supported catalyst // J Phys Chem B. 2002. V. 106. P. 13122-13132
64. R.L. Vander Wal. Flame synthesis of Ni-catalyzed nanofibers // Carbon. 2002. V. 40. P. 2101-2107
65. W.I. Milne, T.B.K. Teo, G.A.J. Amaratunga, P. Legagneuz, L. Gangloff, J.P. Schnell, et al. Carbon
nanotubes as field emissions sources // J Mater Chem. 2004. V. 14. P. 933-943
66. W. Merchan-Merchan, A.V. Saveliev, L.A. Kennedy. High-rate flame synthesis of vertically aligned
carbon nanotubes using electric field control // Carbon. 2004. V. 42. P. 599-608
67. W. Merchan-Merchan, A.V. Saveliev, L.A. Kennedy. Flame nanotube synthesis in moderate electric
fields: from alignment and growth rate effects to structural variations and branching phenomena //
Carbon. 2006. V. 44. P. 3308-3314
68. W. Wasel, K. Kuwana, P.T.A. Reilly, K. Saito. Experimental characterization of the role of
hydrogen in CVD synthesis of MWCNTs // Carbon. V. 45. P. 833-838
69. M. Silvestrini, W. Merchan-Merchan, A.V. Saveliev, L.A. Kennedy. Fullerene formation in
atmospheric pressure opposed flow oxy-flames. Proc Combust Inst. 2005. V. 30. P. 2545-2552
70. T.X. Li, H.G. Zhang, F.J. Wang, Z. Chen, K. Saito. Synthesis of carbon nanotubes on Ni-alloy and
Si-substrates using counterflow methaneeair diffusion flames // Proc Combust Inst. 2007. V. 31.
P. 1849-1856
71. R.L. Vander Wal, J.H. Lee. Ferrocene as a precursor reagent for metal-catalyzed carbon nanotubes:
competing effects // Combust Flame. 2002. V. 130. P. 27-36.
72. R.L.Vander Wal. Fe-catalyzed single-walled carbon nanotube synthesis within a flame environment
// Combust Flame. 2002. V. 130. P. 37-47.
73. J.W. Snoeck, G.F. Froment, M. Fowles. Filamentous carbon formation and gasification:
thermodynamics, driving force, nucleation and steady state growth // J Catal. 1997. V. 169. P. 240-
249
74. H. Wang, M. Frenklach. A detailed kinetic modeling study of aromatics formation in laminar
premixed acetylene and ethylene flames // Combust Flame. 1997. V. 110. P. 173-221
75. M. Frenklach, H. Wang. Detailed modeling of soot particle nucleation and growth. In: The twenty-
third symposium (international) on combustion // Pittsburgh, PA: The Combustion Institute. 1990.
P. 1559-1571
76. C. Janzen, P. Roth. Formation and characteristics of Fe2O3 nanoparticles in doped low pressure
H2/O2/Ar flames. Combust Flame. 2001. V. 125. P. 1150-1161
77. K.H. Homann. Fulleren-und russbildung-wege zu grossen teilchen in flammen // Angew Chem 1998.
V. 110. P. 2572-2590
78. H. Jander, H.G. Wagner. Formation of flame ions, clusters, nanotubes, and soot in hydrocarbon
flames // Combust Explosion Shock Waves. 2006. V. 42. P. 696-701
79. A. Chambers, N.M. Rodriguez, R.T.K. Baker. Influence of copper on the structural characteristics of
carbon nanofibers produced from the cobalt-catalyzed decomposition of ethylene // J Mater Res.
1996. V. 11. P. 430-438
80. M.J. Height, J.B. Howard, J.W. Tester, J.B. Vander Sande. Flame synthesis of single-walled carbon
nanotubes // Carbon. 2004. V. 42. P. 2295-2307
81. M.J. Height, J.B. Howard, J.W. Tester. Synthesis of single-walled carbon nanotubes in a premixed
acetylene/oxygen/argon flame // In: Proceedings of the third joint meeting of the U.S. section of the
combustion institute. 2003
82. Z.A. Mansurov. Soot and nanomaterials synthesis in the flame // Journal of Materials Science and
Chemical Engineering. 2014. 2. P.1-6
83. Z.A. Mansurov Producing Nanomaterials in Combustion. Combustion, Explosion, and Shock
Waves. 2012. V. 48, № 5. P. 561-569

108
84. C. Park, P.E. Anderson, A. Chambers, C.D. Tan, R. Hidalgo, N.M. Rodriguez. Further studies of the
interaction of hydrogen with graphite nanofibers // J Phys Chem B. V. 103. P. 10572-10581
85. X. Wang, Z. Zhong, J. Xu. Noncontact thermal characterization of multiwall carbon nanotubes //
J Appl Phys. 2005. V. 97. P. 302-305
86. L. Zhang, D.E. Resasco. Single-walled carbon nanotube pillars: a superhydrophobic surface //
Langmuir. 2009. V. 25. P. 4792-4798
87. S. Naha, S. Sen, I.K. Puri. Flame synthesis of superhydrophobic amorphous carbon surfaces //
Carbon. 2007. V. 45. P. 1702-1706
88. H. Schulz, M. Leonhardt, H-J. Scheibe, B. Schultrich. Ultra hydrophobic wetting behavior of
amorphous carbon films // Surf Coat Technol. 2005. V. 200. P. 1123-1126
89. M.F. Yu, O. Lourie, M.J. Dyer, K. Moloni, T.F. Kelly, R.S. Ruoff. Strength and breaking
mechanism of multiwalled carbon nanotubes under tensile load // Science. 2000. V. 287. P. 637-640
90. L. Forro, C. Schnongenberger. Carbon nanotube synthesis, structure, and applications // 1st ed. New
York: Springer-Verlag. 2001
91. M. Yu, M.J. Dyer, G.D. Skidmore, H.W. Rohrs, X. Lu, D.K. Ausman, et al. Threedimensional
manipulation of carbon nanotubes under a scanning electron microscope // Nanotechnology. 1999.
V. 10. P. 244-252
92. F. Ding, A. Rosen, K. Bolton. The role of the catalytic particle temperature gradient for SWNT
growth from small particles // Chem Phys Lett. 2004. V. 393. P. 309-313
93. G. W. Lee, J. Jurng, and J. Hwang, Formation of Ni-catalyzed multiwalled carbon nanotubes and
nanofibers on a substrate using an ethylene inverse diffusion flame, J. Combust. Flame, 139, 167–
175 (2004).
94. S. Naha, S. Sen, A. K. De, I. K. Puri, A detailed model for the flame synthesis of carbon nanotubes
and nanofibers, Proc. Combust. Inst., 31, Issue 2, 1821–1829 (2007).
95. S. Naha and I. K. Puri, A model for catalytic growth of carbon nanotubes, J. Phys. D: Appl. Phys.,
41, No. 065304, 6 (2008).
96. W.B. Choi, D.S. Chung, J.H. Kang, H.Y. Kim, Y.W. Jin, I.T. Han, et al. Fully sealed, high-
brightness carbon-nanotube field-emission display // Appl Phys Lett. 1999. V. 75. P. 3129-3131
97. Y.H. Lee, Y.T. Jang, D.H. Kim, J.H. Ahn, B.K. Ju. Realization of gated field emitters for
electrophotonic applications using carbon nanotube line emitters directly grown into submicrometer
holes // Adv Mater. 2001. V. 13. P. 479-482
98. W.B. Choi, Y.W. Jin, H.Y. Kim, S.J. Lee, M.J. Yun, J.H. Kang, et al. Electrophoresis deposition of
carbon nanotubes for triode-type field emission display // Appl Phys Lett. 2001. V. 78. P. 1547-1549
99. M.S. Dresselhaus, G. Dresselhaus, P.C. Eklund. Science of fullerenes and carbon nanotubes. San
Diego: Academic Press. 1996
100. T.W. Ebbesen. Carbon nanotubes: preparation and properties // 1st ed. Boca Raton: CRC-Press.
1997
101. T.W. Tombler, C.W. Zhou, L. Alexseyev, J. Kong, H.J. Dai, L. Lei, et al. Reversible
electromechanical characteristics of carbon nanotubes under local-probe manipulation // Nature.
2000. V. 405. P. 769-772
102. E. Frackowiak, K. Metenier, V. Bertagna, F. Beguin. Supercapacitor electrodes from multiwalled
carbon nanotubes // Appl Phys Lett. 2000. V. 77. P. 2421-2423
103. M. Hughes, G.Z. Chen, M.S.P. Shaffer, D.J. Fray, A.H. Windle. Electrochemical capacitance of a
nanoporous composite of carbon nanotubes and polypyrrole // Chem Mater. 2002. V. 14. P. 1610-
1613
104. E. Frackowiak, F. Beguin. Electrochemical storage of energy in carbon nanotubes and
nanostructured carbons. Carbon. 2002. V. 40. P. 1775-1787
105. A.C. Dillon, K.M. Jones, T.A. Bekkedahl, C.H. Kiang, D.S. Bethune, M.J. Heben. Storage of
hydrogen in single-walled carbon nanotubes // Nature. 1997. V. 386. P. 377-379
106. C. Park, P.E. Anderson, A. Chambers, C.D. Tan, R. Hidalgo, N.M. Rodriguez. Further studies of the
interaction of hydrogen with graphite nanofibers // J Phys Chem B. V. 103. P. 10572-81
107. H. Dai, A.G. Rinzler, P. Nikolaev, A. Thess, D.T. Colbert, R.E. Smalley: Single-wall nanotubes
produced by metal-catalysed disproportionation of carbon monoxide // Chem. Phys. Lett. 1996. V.
26. P. 471–475
108. J.H. Hafner, C.L. Cheung, A.T. Wooley, C.M. Lieber: Structural and functional imaging with carbon
nanotube AFM probes // Progr. Biophys. Mol. Biol. 2001. V. 77. P. 73-110
109. S.S. Wong, E. Joselevich, A.T. Woodley, C.L. Cheung, C.M. Lieber: Covalently functionalized
nanotubes as nanometre-size probes in chemistry and biology // Nature. 1998. V. 394. P. 52-55
110. R. Stevens, C. Nguyen, A. Cassel, L. Delzeit, M. Meyyapan, J. Han: Improved fabrication approach
for carbon nanotube probe devices // Appl. Phys. Lett. 2000. V. 77. P. 3453-3455
111. F. Zheng, L. Liang, Y. Gao, J.H. Sukamto, L. Aardahl: Carbon nanotubes synthesis using
mesoporous silica templates // Nano Lett. 2002. V. 2. P. 729-732

109
 Chapter 7

SYNTHESIS OF HYDROPHOBIC SOOT IN FLAME

Wettability of a solid surface describes the ability of a liquid to maintain contact

with the surface, which is important in the bonding or adherence of two materials [1].
Dependent on both roughness and chemical heterogeneity, wettability is a very
important characteristic in nature as well as in our daily life. Hydrophobicity and
hydrophilicity are two principal wettablility conditions. The surfaces of materials are
hydrophilic if the water contact angle (WCA, θ) is in the range of 0° ≤ θ < 90° and
they are hydrophobic if the WCA is 90° < θ ≤ 180°. Hydrophobicity is observed in
nonpolar substances, which tend to aggregate in aqueous solutions and exclude water
molecules.
Superhydrophobic materials have recently attracted attention from both academic
and industrial circles because of their importance in fundamental research and potential
industrial applications such as bio-surfaces, anti-biofouling, transparent and
antireflective superhydrophobic coatings, structural color, fluidic drag reduction,
enhancing water supporting force, controlled transportation of fluids,
superhydrophobic valves, battery and fuel cell applications, prevention of water
corrosion and oil-water separation corrosion protective coatings, preventing adhesion
of water and snow to windows or antennas, self-cleaning, enhancing buoyancy, and
coating films for electronic devices [2, 3].

7.1. The history of hydrophobic surface

The phenomenon of superhydrophobicity was first studied by Johnson and

Dettre [4] in 1964 using rough hydrophobic surfaces. They developed a theoretical
model based on experiments with glass beads coated with paraffin and
polytetrafluoroethylene telomer, respectively. Barthlott and Ehler [5] studied the self-
cleaning property of superhydrophobic micro-nanostructured surfaces in 1977, and
they described such self-cleaning and superhydrophobic properties for the first time as
the “lotus effect”. The superhydrophobic state of a surface is defined by the contact
angle between a water droplet and the surface of a material that has a WCA above 150
°. Superhydrophobic surfaces display a self-cleaning effect (water repellent) widely
known as the “lotus effect”. Superhydrophobicity can be achieved either by selecting
low surface energy materials or by introducing roughness [5, 6].
In order to create the lotus effect, it is important to understand how nature
generates this property. E. Stratakis [7] and his co-workers have researched by
scanning electron microscope (SEM) images of the lotus leaf surface and have
revealed large structural diversity, with the common characteristic of roughness over
two different length scales. As shown in Fig. 1(a) the lotus leaf surface and (b), is
covered with micrometer-sized papillae decorated with nanometer branchlike
protrusions.
Activated carbons, graphite, carbon fibers (CFs), fullerenes, carbon nanofibers
(CNFs) (carbon nanotubes [CNTs], graphite nanofibers [GNFs]), micro-/meso-
110
porous carbon, and, more recently, graphene carbon-based materials have emerged
as extremely promising materials for various types of applications owing to their
outstanding electronic and optical properties [8-10]. Recently, carbon-based
superhydrophobic surfaces have been fabricated because of their promising
applications such as conductive-transparent films, oil-water separation, and electro-
magnetic-interference-shielding [11-13]. In general, small water droplets on CF
fabrics will be up to 120°, but not
much greater for a flat surface.
To enhance the hydropho-
bicity of carbon-based materials,
two approaches have been de-
veloped recently [12]. First, re-
search on superhydrophobic car-
bon-based materials has focused
on increasing surface roughness
by introducing geometric surface
area.
Dip-coatings methods are
also often employed to obtain
superhydrophobic surfaces. As Fig. 1. (a) SEM image of the surface of a lotus leaf and (b)
demonstrated by Nguyen et al. a higher magnification with hierarchical structures clearly
[14], iterative dip coating of a resolved. (c) A water drop on the surface of the lotus leaf
foam in a graphene solution attains a nearly spherical shape [7]
allows desired amounts of grap-
hene sheets to be deposited on the pore-wall surface of melamine foams, rendering the
foams superhydrophobic and superoleophilic, with a WCA of 162 ° and a chloroform
absorption capacity of 165 times their own weight (1.86 ton m-3) [14, 15]. Second,
various functional groups or additional organic/inorganic materials can be intro- duced
to remarkably vary surface wettability [16]. For example, Zeng et al. [17] and
Charinpanitkul et al. [15] reported a simple method to produce CNT-based films with
exceptional superhydrophobicity and impact icephobicity based on deposition of
acetone-treated single-walled CNTs (SWCNTs) onto glass substrates. The acetone-
functionalized films showed a strong ability to mitigate ice accretion from
supercooled water droplets (-8 °C), and the droplets were found to bounce off the films
tilted at 30 °. The untreated nanotubes films did not display similar behavior, and the
supercooled water droplets remained attached to the films’ surfaces. Such studies could
serve as the foundation of highly versatile technologies for both water and ice
mitigation.
Most current research on carbon-based materials has focused on their electrical,
physico-chemical, and mechanical properties [18]. In particular, surface properties
such as robustness against environmental contamination are critical design
considerations if intrinsic properties are to be maintained. Herein, we start by
discussing the properties of carbon-based materials with superhydrophobic surfaces
provided by emerging methods to create superhydrophobic surfaces with enhanced
bulk properties. The recent advances in this field are summarized, including the
wetting behavior of water on carbon materials such as CFs, CNTs, graphene, and
amorphous carbons, and the formation of hierarchical structures and low surface
energy chemical composition, with emphasis on fundamental understanding of related
processes. Potential applications in energy, environmental remediation, and thermal
management are also discussed [19].
111
 .
7.2. Methods of obtaining hydrophobic surface

A great variety of methods to fabricate superhydrophobic surfaces on different

substrates have been introduced in recent years. A typical approach to create a
superhydrophobic surface is either to pattern a hydrophilic material and coat the
surface afterwards with a thin layer of hydrophobic material, or to pattern some
inherently hydrophobic material. Low surface energy materials such as siloxanes or
fluoropolymers are typically used to obtain the hydrophobic chemistry on artificial
superhydrophobic surfaces. However, the extremely low energy surface chemistry is
not a necessary condition for superhydrophobicity, because natural superhydrophobic
surfaces, which can have extremely high contact angles (CAs) and low sliging angles
(SAs), possess only moderately low energy chemistry on their waxy surface.
Almost any solid material, e.g. hard solids such as metals and metal oxides, elastic
and flexible polymers, and fiber-based materials such as paper and textiles, can be
used as a substrate for a superhydrophobic surface. Among others, plasma, laser, and
chemical etching can be used to roughen solid substrates in order to fabricate a
superhydrophobic surface. Photolithography is a common way to create well-defined
silicon micropillar arrays for superhydrophobic surfaces.
Other frequently used methods to fabricate superhydrophobic surfaces include the
use of different templates, electrospinning of nanofibers, evaporation of polymer
solutions, sol-gel coatings, and various spraying techniques, e.g. the use of
nanoparticle suspensions [20-22].

7.2.1. Roughening of hydrophilic materials

Fabrication of superhydrophobic surfaces from hydrophilic materials, such as

metals and metal oxides, involves two steps: (1) roughening of the solid and (2)
introduction of hydrophobic surface chemistry.

(a) plasmaetched TiO2 surface (inset: side view) [23] and (b) laser-etched silicon surface [24], and
(c-e) tilted views of silicon micropillar arrays fabricated by photolithography

Fig. 2. SEM micrographs of superhydrophobic surfaces fabricated by roughening and subsequent

chemical modification of inherently hydrophilic materials [25]

112
 Zhang et al. [23] roughened titanium dioxide (TiO2) film using CF4 plasma and
hydrophobized the roughened surface with an octadodecylphosphonic acid (ODP)
monolayer (Fig. 2, a). Water CA on the resulted superhydrophobic surface was as high
as 168 º and CAH was measured to be 0 º. Prior to the ODP modification, water CA
on the roughened TiO2 film was close to 0º due to the high roughness and intrinsic
hydrophilic nature of the surface. Gao et al. [24] used a facile laser-etching method to
fabricate anisotropic microstructures and rugged nanoprotrusions on a silicon wafer
(Fig. 2, b). After modification with fluoroalkylsilane, the surface showed perfect
isotropic superhydrophobicity (CA ~180º) without any apparent CAH, water adhesion,
or drag resistance. Öner and McCarthy [25] prepared well-defined silicon micropillars
with various dimensions by the means of photolithography (Fig. 2, c-e) and modified
the surfaces with hydrophobic compounds in order to execute fundamental
superhydrophobicity studies. Micropillars with closely controlled dimensions are very
useful, for example, in combining the theoretical and practical understanding of
superhydrophobicity.

7.2.2. Fabrication of rough surfaces from hydrophobic materials

When a superhydrophobic surface is made by roughening intrinsically

hydrophobic material, further hydrophobization steps are not needed. Jin et al. [26]
prepared the high-adhesive superhydrophobic PS nanotube film shown in Fig. 3, using
a porous alumina membrane as a template. Feng et al. [27] fabricated the
superhydrophobic anisotropic CNT film shown in Fig. 3, b by local growth of packed
aligned CNTs.

(a) Side view (inset: top view) of PS nanotube film mimicked from gecko feet [26];
(b) biomimicked CNT film [27]

Fig. 3. SEM micrographs of biomimicked and natural superhydrophobic surfaces

The use of polymeric templates, e.g. polydimethylsiloxane (PDMS) or poly(vinyl

alcohol) (PVA), is an effective way to mimic natural surfaces with high level of
details, as was shown by Sun et al. [28] and Feng et al. [29], who replicated
superhydrophobic surfaces of lotus leaf and rose petal, respectively. Water CA on the
PDMS replica of lotus surface was 160º and SA for a droplet with a radius of 1 mm
was less than 2º, while the PS replica of rose petal did show similar high-adhesive
superhydrophobicity (CA = 155 º, droplets were pinned to the surface) that was
observed on the natural petal surface.
Zhu et al. [30] prepared a lotus-leaf-like conductive polyaniline (PANI)/PS
composite film by the electrospinning method (Fig. 4, a). The superhydrophobic film
113
showed CA of 166.5 º for 4 μl water droplets, and the SA was measured to be < 5º.
Erbil et al. [31] prepared a porous superhydrophobic polypropylene (PP) coating (Fig.
4, b), on which water CA was 160º, by evaporating a mixture of PP, xylene and
methyl ethyl ketone. A similar approach was later adapted by Lu et al. [32] for
preparation of superhydrophobic low-density polyethylene (LDPE) coating, on which
water CA as high as 173 º and SA of only 1.9 º were measured. Shirtcliffe et al. [33]
prepared superhydrophobic and highly porous organosilica sol-gel foam (Fig. 4, c), on
which the advancing water CA and CAH were 155 º and 6 º, respectively. Kulinich
and Farzaneh [34] sprayed zirconium dioxide (ZrO2) nanoparticlecontaining
fluoropolymer solution on aluminum plate to produce a high-adhesive superhyd-
rophobic surface (Fig. 4, d), where water CA was 152 º and CAH was > 70 º.

(a) electrospun lotus-like PANI/PS fiber mat [30], (b) porous PP film prepared by evaporating the
polymer solution [31], (c) organosilica sol-gel foam [33], and (d) spray-coated fluoropolymer
nanoparticle suspension [34]

Fig. 4. Superhydrophobic surfaces fabricated from hydrophobic materials

Many of the techniques that are used to fabricate superhydrophobic surfaces or

coatings are multi-step and time consuming, and may require special equipment or
conditions, e.g. low pressure vacuum chambers. However, there are also very simple
approaches to create superhydrophobic surfaces. For example, Onda et al. [35]
fabricated the surface used in the Kao experiment by letting melted alkyl ketene dimer
(AKD) solidify at room temperature in the medium of dry nitrogen gas. Water CA on
the resulted fractal AKD surface was as high as 174 º. Zhang et al. [36] achieved a
superhydrophobic surface simply by extending a Teflon (poly(tetrafluoroethylene),
PTFE) film. The CA increment from 118 º to 165 º was due to the changed
microstructure on the surface and the increased fraction of air between the fibrous
PTFE crystals. Xu et al. [37] demonstrated a simple laminating templating method,
which appears to be suitable for large-scale fabrication of robust superhydrophobic
surfaces on polymers. They laminated different types of woven meshes on LDPE
under heat and pressure. After cooling, the meshes were peeled off leaving behind

114
highly textured polymer surfaces (Fig. 5), on which water CAs ~160 º were measured,
and CAH and SA for 10 -l droplets were in the range of 5-8 º and 3-5 º, respectively. In
fact, a superhydrophobic coating can be fabricated, as simply as, by collecting a layer
of soot from a candle flame on, for example, a metal or glass plate.

Fig. 5. SEM images of superhydrophobic LDPE surfaces fabricated by laminating templating method
using different types of woven mesh templates. Images b, d, and f are magnifications from images a, c,
and e, respectively [37]

7.3. Superamphiphobic surfaces

Because of the high surface tension of water, fabrication of superhydrophobic

surfaces is relatively easy. Natural superhydrophobic surfaces have shown that
extremely low surface energy materials are not needed to obtain high CA, low SA, and
low CAH for water. Neither are the requirements for the physical structure of the
surface especially strict, but superhydrophobicity can be obtained, for example, on
simple micropillar surfaces. When the target is to achieve repellency against liquids of
significantly lower surface tension than water, requirements for both the chemistry and
physical structure of the surface are much stricter. In recent years, interest towards
superamphiphobic surfaces, which repel both water and oils, has grown increasingly.
The number of potential applications for superamphiphobic surfaces is greater
compared to superhydrophobic surfaces which can repel solely water. However,
because of the strict requirements for the chemistry and physical structure,
superamphiphobic surfaces are rare.
Superamphiphobicity can be obtained for example with fluorinated T-letter shaped
or mushroom-like micropillar surfaces, where re-entrant curvature plays a critical role
in the nonwetting properties (Fig. 6) as was shown by Tuteja et al. [38, 39]. However,
it is not necessary to use carefully designed re-entrant curved structures to obtain
superamphiphobicity. For example, Deng et al. [40] applied a simple method to
fabricate a robust superamphiphobic coating that could repel a variety of low surface
tension liquids (Fig. 13). They collected candle soot on a glass slide, and used the soot
layer as a template to produce a porous silica network. After coating the silica layer
with a semi-fluorinated silane, the surface showed excellent superamphiphobic
properties, e.g. CA and SA for water (surface tension of 72.1 mN/m) were 165 ° and 1
°, respectively, and the corresponding values for hexadecane (surface tension of 27.5
mN/m) were 156 ° and 5 °, respectively.

115
 Recently, Wong et al. [41] introduced a new approach to create slippery surfaces
with pressure stable omniphobicity and self-healing properties. The idea is that a
porous substrate, similar to many superhydrophobic surfaces, is infused with a
lubricating fluid. While the substrate resembles superhydrophobic surfaces, the
strategy to create slippery surfaces using lubricating fluids is very different from
superhydrophobicity, because the lubricating fluid (e.g. perfluorinated liquid), which
is locked in place by a nano/microporous substrate, creates a smooth and slippery
surface for both high- and low surface tension liquids.

a-b) SEM micrographs of the micropillar surfaces with re-entrant topography; c) a cartoon
highlighting the formation of a liquid-solid-air composite interface on the re-entrant curved surface

Fig. 6. Superamphiphobic surfaces with re-entrant curvature [38]

The as-prepared slippery liquid-infused porous surfaces (SLIPS) had minimal

adhesion to variety of liquids, e.g. liquids with low surface tension such as hexane
(surface tension of 18.6 mN/m). Droplets of various liquids could easily slide on the
surface without leaving any stain: the measured CAH values were < 2.5 ° and SAs
were ~5 ° at the maximum. Once again, inspiration for the surface design was found
from nature, from the Nepenthes pitcher plants [42].

7.4. Synthesis of hydrophobic soot in flame

Most of methods for preporation of superhydrophobic surfaces required special or

expensive materials, special equipment; the complex process was time consuming.
Nevertheless, a convenient and cheap way to fabricate superhydrophobic surfaces is
provided by exposing the substrates to a flame. During the exposure, flame soot was
deposited on the substrates leading to formation of a superhydrophobic surface.
Superhydrophobic surfaces showing high water contact angle were achieved by
burning candle, ethylene-air, and rapeseed oil. However, the structure of the flame
soot deposited on glass substrate is fragile because the particle-particle interactions are
only physical and weak. When water rolls off the flame soot layer, the drop carries
soot particles with it until nearly all of the soot deposit is removed and the drop
undergoes a wetting transition. Therefore, it is important to develop a new method to
avoid this problem.
Low surface-energy materials like amorphous carbon (a-C) films are frequently
used to modify surfaces in order to control their wettability. The nanobeads are
116
morphologically similar to the carbon nanopearls synthesized by Levesque and co-
workers [43] by acetylene dissociation at 700 °C on nickel catalyst nanoclusters. Puri
et al. [44] have determined new ways of synthesis of carbon nanotubes in the fuel-rich
diffusion flames, exposed to an electric field during 2-10 min, on superhydrophobic
surfaces representing nanodimensional round amorphous carbon particles deposited on
a silicon substrate [45].

а) b)
Fig. 7. Image and schematic representation of the burning experimental setup

A silicon (Si) disk is placed 2 cm above the

burner, as shown in Fig. 7, and exposed to a
flame for 4, 6 and 10 minutes. A nonpremixed
flame is established using propane and oxygen
flow rates between 51-150 cm3/ min and 260-
310 cm3/min, respectively, as described in
Table 1 [46]. This results in the deposition of a
carbon film, as shown in Fig. 8. The gas phase
flame temperature just adjacent to the Si disk
is measured to be 900 ºC with a chromel-
alumel thermocouple of 0.2 mm diameter. The
deposition occurs in three zones that are shown Fig. 8. A Si disc with carbon nanobead
in Fig. 2, i.e., (1) a central gray area that is (2) coating after 4 min exposure to the flame
surrounded by a brown zone, which (3) in turn showing three deposition zones
is enclosed within a black sooty outer zone. (1, 2 and 3)
(Fig. 8 does not show these zones in the form
of concentric circles due to minor flickering movements of the flame).
Figure 9 represents images of liquid droplets placed on the central super-
hydrophobic region of the surface (Zone 1 in Fig. 8) and on a more hydrophilic bare Si
wafer surface. A water droplet placed on the hydrophobic surface is able to roll freely
upon it. Regardless of the deposition time, θ lies in the 152.4-157.1º range for such a
surface and the coating is very stable. In the opposite the contact angle for the wettable
bare Si wafer is equal to θ = 50º.
The deposited carbonaceous material has been examined using Raman
spectroscopy. The Raman spectra recorded for different zones (identified in Fig. 8) are
presented in Fig. 10. They all indicate the presence of various forms of carbon. Raman
peak near 1350 cm-1 (marked as D) is due to amorphous-C, while the peak at 1590 cm-
1
(marked as G) is due to graphite. The most original aspect of the present study is the
evidence of a new peak recorded at 1470 cm-1 only for the two zones, 1 and 2. This
peak has been attributed to the fullerene of carbon. Fig. 11 shows the scanning
117
electron micrographs of two samples at the hydrophobic deposit (Zone 1). Both
samples consist of similarly agglomerated nanobeads that are produced as a result of
gas phase fuel pyrolysis in the oxygen-deficient flame core. Heavier hydrocarbons in
this central region of the flame are transported into the stagnation layer adjacent to the
relatively colder substrate where they condense and form the nanostructures.

Table 1

Experimental conditions for the synthesis of carbon deposits

Propane flowrate Oxygen flowrate Substrate height above Exposure time,

Substrate
(cm3/min) (cm3/min) burner, (cm) (min)
51 260 2 4
51 260 2 6
51 260 2 10
51 260 2 4
51 260 2 6
Si
51 260 2 10
70 285 3 6
70 285 3 10
150 310 2 10
150 310 3 10
Ni 150 310 2 10

We next investigated the influence of an electric field of 1 kV strength that is

applied between the substrate and the burner rim for 10 min. Substrates thus exposed
are compared with others exposed to the same flame in the absence of such a field. For
all these cases, the flame is established with propane and oxygen flowrates of 51 and
260 cm3/min, respectively. For this flame, Zones 1 and 2 contain carbon nanobead
chains that are 15 - 30 nm long in the absence of an electric field. This chain length
increases to 40-50 nm when the field is applied. In all cases, the outermost Zone 3
contains coagulated soot aggregates that have characteristic sizes of 30-50 nm.

а b
Fig. 9. (a) A drop of liquid placed on the central superhydrophobic region (zone 1) of the surface shown
in Fig. 2, and (b) a more hydrophilic bare Si wafer surface [47]

When an electric field (of 1 kV) is applied to either the Si or Ni substrate, the
flame changes visually and becomes brighter. It contains a periodic spark that jumps
between the burner rim and the substrate. Then the soot formation in the flame is
enhanced. The increase in flame soot decreases the electric resistance of the flame
which results in the increase of the electric current through the flame. Large quantities
of soot are deposited on the substrate in the form of dendrites, increasing the thickness
118
of the hydrophobic deposit for this case. While a normal flame, i.e., one established in
the absence of an electric field, spreads over the surface of the substrate, in the
presence of an electric field, it propagates as a narrower cylinder of 1.5 cm diameter,
even as it touches the substrate. When the substrate has a negative polarity, the flame
adheres to it. A corresponding positive polarity produces a small gap between the
flame front and substrate, in which case the Zone 1 superhydrophobic deposit exhibits
a larger wetting angle of 160–165 º.

a – Raman spectra of the central gray area (Zone 1); b – Raman spectra of the intermediate brown
region (Zone 2); c – Raman spectra of the outermost black sooty area (Zone 3)

Fig. 10. Raman spectra of soot deposited on Si substrate. The ratio between the intensities of the G and
the D peaks provides information about the lateral dimensions of the flakes [47]

We have also investigated the interactions of three surfactants with the resulting
hydrophobic soot surface on a silicon substrate, namely (1) anionic sodium dodecyl
sulfate (С12H25OSO3Na), (2) cationic N-cetylpyridium bromide (C16H33C5H5N Br),
and (3) the nonionic оxyethylenated alkylphenol (C16H33C6H4O(CH2CH2O)10H).

a b

Fig. 11. Scanning electron micrographs of the chains formed with (a) and without (b)
the application of a 1 kV electric field during the carbon deposition of carbon in Zone 1 (Fig. 8)
119
 In the second step, we investigate the influence of surfactants on the hydrophobicity of
the deposit in Zone 1. The modifying effect of the anionic sodium dodecyl sulfate
(S12H25OSO3Na), cationic acetylpiride bromide (C16H33C5H5N Br) and nonionic
alkylphenol oxyethylenated (C16H33C6H4O (CH2CH2O)10H) surfactants was considered.
The wetting angles obtained as the surfactant concentration is varied are presented in
Fig. 12. This figure shows that the required hydrophobicity or hydrophilicity of the sub-
strates can be modified by varying the surfactant concentration in water.
We also carried out the study of
the conditions of surfactants interac-
tion with the obtained hydrophobic
soot surface on substrates. The mo-
difying effect of anionic – sodium
dodecyl sulfate (S12H25O SO3Na),
cationic – acetylpiride bromide
(C16H33C5H5N Br) and a nonionic
surfactant – alkylphenol Oxyethyle-
nated (C16H33C6H4O (CH2 C H2O)10
H) was analysed. On the basis of the
experimental data, the diagram de-
pendence of wetting angle versus the
logarithm of surfactant concentration
was made by the plot of the contact
Fig. 12. Variation of the contact angle versus the angle versus the logarithm of surfac-
logarithm of concentration [48] tant concentration (Fig. 12).

7.5. Theoretical Background of hydrophobic surface

7.5.1. Surface roughness

Superhydrophobicity, as shown by lotus leaves, is an area that has received much

interest in the past few years in relation to self-cleaning surfaces and other processes
as well as in the design and production of artificial biomimetic surfaces [49-52]. Fig.
13 shows a structural diagram of the lotus effect [49]. The structures of a lotus leaf,
i.e., branch-like nanostructures on top of micropapillae, have been carefully observed
by scientists. These microscale-structures can induce superhydrophobic surfaces with
large contact angles and low sliding angles. It is believed that this unique self-cleaning
property is based on the surface lotus effect illustrated in Fig. 13. The roughness is
caused by the microscale papillae and nanoscale tomenta (Fig. 13, b-d). To construct
superhydrophobic carbon-based material surfaces, it is essential to understand the
wetting behavior of water on the surface of carbons.
Superhydrophobic surfaces are generally expressed by the contact angle between a
water droplet and the solid surface. Wenzel’s early works and later Cassie-Baxter’s
work defined the importance of surface roughness and heterogeneity, which are now
recognized as key parameters in the wettability of hydrophobic surfaces [50, 51].
Wang et al. [52] and Yu et al. [54] showed that superhydrophobic lotus leaves have a
specuticular wax with two scales of roughness (a: papillose epidermal cells of about
around 10 μm; b: tubular wax crystals of about around 10 nm on the papillose
epidermal cells).Contact angle measurements, as described by Young in 1805, remain
120
the most accurate method for determining the interaction energy between a liquid and
solid in a condensed state at the minimum equilibrium distance of solid and liquid.
When a droplet of water rests on a surface, the contact angle can be measured at the
edge of the droplet (Fig. 13). Fig. 14 shows the liquid droplet wetting behavior of the
corresponding theory: (a) Young’s state, (b) Wenzel’s state, (c) Cassie-Baxter’s state,
and (d) an example of re-entrant morphology [3, 55]. More than 200 years ago,
Young, identified in a recent biography as “the last man who knew everything” [56],
described the forces acting on a liquid droplet spreading on a flat surface (Fig. 13, a).

Fig. 13. Lotus effect: (a) lotus leaf and beads; (b) scanning electron microscopy image of micropapillae
present on the surface of a lotus leaf; (c) image of a water droplet on a lotus leaf (a); (d) structural
diagram of micro- and nanostructure of a single micropapilla [43]

On flat surfaces, Young’s equation defines the contact angle (θ) depending on the
solid-vapor, solid-liquid, and liquid-vapor surface tensions, as given in Eq. (1), where
γSV is the interfacial tension between solid and vapor, γSV is the interfacial tension
between solid and liquid, and γLV relates to the interfacial tension between liquid and
vapor. Depending on the value of θ as measured by water, if θ is less than 90 °, the
surface is conventionally described as hydrophilic; if θ varies between 90 ° and 150 °,
the surface is hydrophobic; and if θ is greater than 150 °, the surface is conventionally
described as superhydrophobic [57, 58].

(1)

Depending on the level of surface roughness, two different models can be

distinguished. The first, shown in Fig. 14, b, is Wenzel’s state. The droplet maintains
contact with the surface and fills the asperities, and the surface area associated with
the contact angle is increased by the roughness factor r.
r = roughness factor = actual surface area / planar area (2)

Wenzel’s state is described by Eq. (3).

(3)

121
 In Eq. (3), r is equal to the ratio between the actual surface area of the rough
surface and the projected (apparent) area, where the non-dimensional surface
roughness factor r > 1. r emphasizes the effect of the surface chemistry determined by
the term cos q, the non-dimensional surface roughness. When q is less than 90 °, the
increase in the roughness factor r reduces qw but if q is higher than 90 °, the increase
in roughness leads to the increase in θw .

Fig. 14. Liquid droplet spreading on (a) flat surface (b) and rough surface (b-d). Depending on the
roughness of the surface, the droplet is either in the so-called (a) Young, (b) Wenzel, or (c) Cassie-Baxter
state; (d) example of re-entrant morphology (color figure available online).

If the droplet is suspended on the surface asperities and the liquid does not
penetrate the protrusions of the surface features, then this case refers to Cassie-
Baxter’s model. In this model, the apparent contact angle is the result of all
contributions of different phases [59]. In Eq. (4), f1 and f2 are the fraction of the
projecting solid and vapor on the surface, respectively, and f1 + f2 = 1. Eq. (4) only
applies to cases where the liquid only contacts the top of the surface. However, a more
complex version of Eq. (4) is required when partial penetration of the grooves occurs
because the pores are filled with air. The hydrophobicity is enhanced by the increases
of the vapor fraction. Because the pores are filled with air, which is hydrophobic, the
contact angle always increases, relative to the behavior seen on a flat substrate having
an identical chemical composition [59].

(4)

From these two theories, it can be found that surface topography can enhance
surface wettability on solid surfaces, whether they are hydrophobic or hydrophilic.
This finding guides us to tune the surface wettability by controlling surface
geometrical structures independently of the chemical composition [59-61]. On the
other hand, the surface morphology was found to be an important parameter and more
precisely “reentrant” structures are required to obtain such a superhydrophobic surface
(Fig. 14, d). These morphologies induce a negative Laplace pressure difference
because they introduce a transition in the liquid-vapor interface from concave to
122
convex, inducing a higher energy barrier between Wenzel’s states and Cassie-Baxter’
states [60, 61].

7.5.2. Modification of roughness of low surface energy materials

Based on the general contact angle theory of Wenzel, two main approaches have
been developed to generate superhydrophobic surfaces. One is increasing the surface
area on a microscopic scale of low surface energy materials; the other is to fabricate a
suitable surface roughness with certain materials and subsequently modify the as-
prepared surface with low surface energy materials [62]. The latter method is no longer
limited to low surface energy materials, and it can be extended to the fabrication of
hydrophobic surfaces for many systems. Various low surface energy materials have
been developed to modify microscopic scale surfaces to form superhydrophobic
surfaces. Fluorinated compounds, silane, and long alkyl chain fatty acids are typical
low surface energy compounds, and have the ability to endow various substrates
(SiO2, TiO2, Al2O3, metals, polymers etc.) with high hydrophobicity [62-64].
The surface energy is generally defined as the work required to build a unit of area
of a given surface, using the sessile drop technique [65, 66]. Numerous such theories
have been developed by various researchers. These methods differ in several ways, in
terms of derivation and convention for example. Most importantly, they differ in the
number of components or parameters. Simpler methods containing fewer com-
ponents simplify the system by converging the surface energy values into one number,
whereas strict methods using more components are derived to distinguish various
components of the surface energy.
In addition, the total surface energy of solids and liquids depends on the different
types of molecular interactions, such as the London dispersive and the polar or
acid/base interactions, and is considered to be the sum of these independent
components [66].
In the early 1960-ies, Fowkes [67, 68] introduced the concept of the surface free
energy of a solid. The total surface free energy can be divided into the London
dispersive and specific (or polar) components [67, 68].

(5)

where γS is the total surface free energy, the subscript S represents a solid state, and
γSL and γSSP are the London dispersive and specific (Debye, Keesom of van der Waals,
H-bonding, π-bonding, and other small polar effects) components of the surface free
energy of the constitutive elements. The London dispersive and specific components
are determined by measuring the contact angles of two testing liquids with known
London dispersive and specific components. The surface free energy of a solid can
be determined on the basis of contact angle measurements using the geometric mean,
according to Fowkes’ proposition based on a solid (subscript S)-liquid (L) droplet-air
system, as described by the following equation [67]:

(6)

where θ is the contact angle of a liquid droplet in the solid state. As reported in our
previous work, the specific component γSSP is highly dependent on the surface
123
functional groups, and the dispersive component γSL is largely dependent on the total
electron density in the carbon [68].

7.6. Application of superhydrophobic surfaces

Superhydrophobic surfaces and coatings can potentially be utilized in variety of

applications including self-cleaning, stain-resistant, anti-icing, and anti-fogging surfaces;
wind screens; waterproof breathable clothing and coatings; anti-biofouling and anti-
corrosion coatings; fogharvesting; oil separation from water; microfluidics; drag reduction;
and liquid transportation. In addition to the special wettability, many other useful
properties have been typically incorporated into superhydrophobic coatings, for example
transparency, structural color, gas permeability, and flexibility [20-22].
Even though many promising applications for superhydrophobicity have been
identified, and there are several facile methods to fabricate superhydrophobic surfaces
and coatings, the large-scale utilization of superhydrophobic surfaces is limited. This
is mainly due to the poor wear resistance, which is a common problem with
superhydrophobic surfaces [20]. The fragile micrometer and submicrometer scale
surface structures are easily damaged in daily use, for example, due to abrasion.
Artificial superhydrophobic surfaces cannot renew and heal the damages similarly to
natural surfaces. Fabrication of robust surfaces, which can maintain their
functionalities also in wearing conditions, is one of the major issues to be resolved
with superhydrophobic and other functional surfaces.
However, major progress in fabrication of robust superhydrophobic surfaces has
been achieved in recent years. For example, the superamphiphobic coating introduced
by Deng et al. [40] could maintain its liquid-repellent properties even when the
topmost surface layer was worn out or peeled off using adhesive tape. Zhao et al. [42]
fabricated a grooved surface with re-entrant curvature to obtain directional
superamphiphobicity and mechanical robustness, and Zhou et al. [69] applied a highly
durable superhydrophobic silicone rubber/nanoparticle composite coating on fabrics,
which was capable of withstanding 500 laundering cycles without any significant
change in its wetting properties. Recently, also self-healing functions have been
successfully incorporated into superhydrophobic surfaces [70-72]. Apparently, the
first self-healing superhydrophobic coating was introduced by Li et al. [70]. Their
porous micro-/nanostructured fluoro-modified polymer coating could preserve healing
agent units of reacted fluoroalkylsilane, and the preserved healing agents could
migrate to the coating surface to heal its chemistry after the original fluoroalkylsilane
layer was decomposed by an oxygen plasma treatment. However, so far there are no
artificial superhydrophobic surfaces which are capable of healing physical damages
similar to SLIPS [41] or natural superhydrophobic surfaces.
Although many natural superhydrophobic surfaces possess a dual-scale roughness
that contributes to easy mobility of water droplets and self-cleaning properties, by now
it is clear that hierarchical surface roughness is not a necessary condition for
superhydrophobicity with low droplet adhesion and self-cleaning properties, but
similar superhydrophobic properties can be achieved with a single-scale submic-
rometric surface roughness. It seems reasonable that one important reason why nature
has chosen dual-scale hierarchical roughness for many of its superhydrophobic sur-
faces, e.g. lotus leaf, is the enhanced wear resistance of the fragile surfaces [24, 74].

124
CONTROL QUESTIONS

1. Can you tell about history of superhydrophobic surface?

2. What kind of methods of obtaining superhydrophobic surface do you know?
3. Can you describe the method of fabrication of rough surfaces from hydrophobic materials?
4. Which type of methods for identification of fullerenes do you know?
5. What do you think of synthesis of hydrophobic soot in flame?

REFERENCES

1. Qiao R, Zhang R, Zhu W, Gong P. Lab simulation of profile mod- ification and enhanced oil recovery
with a quaternary ammonium cationic polymer. J Ind Eng Chem, 18, 111 (2012). http://dx.doi.
org/10.1016/j.jiec.2011.11.092.
2. Fei T, Chen H, Lin J. Transparent superhydrophobic films pos- sessing high thermal stability and
improved moisture resistance from the deposition of MTMS-based aerogels. Colloids Surf
Physicochem Eng Aspects, 443, 255 (2014).
3. Wolfs M, Darmanin T, Guittard F. Superhydrophobic fibrous polymers. Polym Rev, 53 460
(2013)
4. Johnson RE, Dettre RH. Contact angle hysteresis. In: Fowkes FM, Zisman WA, eds. Contact
Angle, Wettability, and Adhesion (Advances in Chemistry Series Vol. 43), American Chemical
Society, Washington, DC, 112 (1964).
5. Barthlott W, Ehler N. Raster-Elektronenmikroskopie der Epidermis-Oberfla chen von
Spermatophyten (Tropische und subtropische Pflanzenwelt Vol. 19), Akademie der Wiss. u.d.
Literatur, Mainz (1977).
6. Yong J, Yang Q, Chen F, Zhang D, Du G, Bian H, Si J, Yun F, Hou X. Superhydrophobic PDMS
surfaces with three-dimensional (3D) pattern-dependent controllable adhesion. Appl Surf Sci, 288,
579 (2014).
7. E. Stratakis, V. Zorba, M. Barberoglou, E. Spanakis, S. Rhizopoulou, P. Tzanetakis, S. Anastasiadis,
C.Fotakis. Laser structuring of
water-repellent biomimetic surfaces // Newsroom. 10.1117/2.1200901.1441
8. Hassan AF, Youssef AM, Priecel P. Removal of deltamethrin insecticide over highly porous
activated carbon prepared from pistachio nutshells. Carbon Lett, 14, 234 (2013).
9. Li B, Zhao Z, Gao F, Wang X, Qiu J. Mesoporous microspheres composed of carbon-coated TiO2
nanocrystals with exposed {0 0 1} facets for improved visible light photocatalytic activity. Appl
Catal B, 147, 958 (2014).
10. Jain A, Jayaraman S, Balasubramanian R, Srinivasan MP. Hy- drothermal pre-treatment for
mesoporous carbon synthesis: enhancement of chemical activation. J Mater Chem A, 2, 520
(2014).
11. Chen Z, Dong L, Yang D, Lu H. Superhydrophobic graphene- based materials: surface
construction and functional applica- tions. Adv Mater, 25, 5352 (2013).
12. Bayer IS, Steele A, Loth E. Superhydrophobic and electrocon-ductive carbon nanotube-
fluorinated acrylic copolymer nanocomposites from emulsions. Chem Eng J, 221, 522 (2013).
13. Bai BC, Cho S, Yu HR, Yi KB, Kim KD, Lee YS. Effects of ami- nated carbon molecular sieves on
breakthrough curve behavior in CO2/CH4 separation. J Ind Eng Chem, 19, 776 (2013).
14. Nguyen DD, Tai NH, Lee SB, Kuo WS. Superhydrophobic and superoleophilic properties of
graphene-based sponges fabricated using a facile dip coating method. Energy Environ Sci, 5, 7908
(2012).
15. Charinpanitkul T, Suthabanditpong W, Watanabe H, Shirai T, Faungnawakij K, Viriya-empikul N,
Fuji M. Improved hydrophi- licity of zinc oxide-incorporated layer-by-layer polyelectrolyte film
fabricated by dip coating method. J Ind Eng Chem, 18, 1441 (2012).
16. Chauhan NPS. Structural and thermal characterization of macro- branched functional terpolymer
containing 8-hydroxyquinoline moieties with enhancing biocidal properties. J Ind Eng Chem, 19,
1014 (2013).
17. Zheng L, Li Z, Bourdo S, Khedir KR, Asar MP, Ryerson CC, Biris AS. Exceptional
superhydrophobicity and low velocity impact icephobicity of acetone-functionalized carbon
nanotube films. Langmuir, 27, 9936 (2011).
18. Heo GY, Yoo YJ, Park SJ. Effect of carbonization temperature on electrical conductivity of carbon
papers prepared from petro- leum pitch-coated glass fibers. J Ind Eng Chem, 19, 1040 (2013).
19. Lee SY, Yop Rhee K, Nahm SH, Park SJ. Effect of p-type multi- walled carbon nanotubes for
improving hydrogen storage behaviors. J Solid State Chem, 210, 256 (2014).

125
20. Roach, P.; Shirtcliffe, N. J.; Newton, M. I., Progess in superhydrophobic surface development. Soft
Matter 2008, 4, (2), 224.
21. Ma, M.; Hill, R. M., Superhydrophobic surfaces. Curr. Opin. Colloid Interface Sci. 2006, 11, (4),
193-202.
22. Carré, A.; Mittal, K. L. (eds.), Superhydrophobic Surfaces. VSP/Brill: Leiden, 2009; p 495.
23. Zhang, X.; Jin, M.; Liu, Z.; Nishimoto, S.; Saito, H.; Murakami, T.; Fujishima, A., Preparation and
Photocatalytic Wettability Conversion of TiO2-Based Superhydrophobic Surfaces. Langmuir 2006,
22, 9477-9479.
24. Gao, X.; Yao, X.; Jiang, L., Effects of Rugged Nanoprotrusions on the Surface Hydrophobicity and
Water Adhesion of Anisotropic Micropatterns. Langmuir 2007, 23, 4886-4891.
25. Öner, D.; McCarthy, T. J., Ultrahydrophobic Surfaces. Effects of Topography Length Scales on
Wettability. Langmuir 2000, 16, 7777-7782.
26. Jin, M.; Feng, X.; Feng, L.; Sun, T.; Zhai, J.; Li, T.; Jiang, L., Superhydrophobic Aligned
Polystyrene Nanotube Films with High Adhesive Force. Adv. Mater. 2005, 17, 1977-1981.
27. Feng, L.; Li, S.; Li, Y.; Li, H.; Zhang, L.; Zhai, J.; Song, Y.; Liu, B.; Jiang, L.; Zhu, D., Super-
Hydrophobic Surfaces: From Natural to Artificial. Adv. Mater. 2002, 14, 1857-1860.
28. Sun, M.; Luo, C.; Xu, L.; Ji, H.; Ouyang, Q.; Yu, D.; Chen, Y., Artificial Lotus Leaf by
Nanocasting. Langmuir 2005, 21, 8978-8981.
29. Feng, L.; Zhang, Y.; Xi, J.; Zhu, Y.; Wang, N.; Xia, F.; Jiang, L., Petal Effect: A Superhydrophobic
State with High Adhesive Force. Langmuir 2008, 24, 4114-4119.
30. Zhu, Y.; Zhang, J.; Zheng, Y.; Huang, Z.; Feng, L.; Jiang, L., Stable, Superhydrophobic, and
Conductive Polyaniline/Polystyrene Films for Corrosive Environments. Adv. Funct. Mater. 2006,
16, (4), 568-574.
31. Erbil, H. Y.; Demirel, A. L.; Avci, Y.; Mert, O., Transformation of a simple plastic into a
superhydrophobic surface. Science 2003, 299, (5611), 1377-80.
32. Lu, X.; Zhang, C.; Han, Y., Low-Density Polyethylene Superhydrophobic Surface by Control of Its
Crystallization Behavior. Macromol. Rapid Commun. 2004, 25, (18), 1606-1610.
33. Shirtcliffe, N. J.; McHale, G.; Newton, M. I.; Perry, C. C., Intrinsically Superhydrophobic
Organosilica Sol-Gel Foams. Langmuir 2003, 19, 5626-5631.
34. Kulinich, S. A.; Farzaneh, M., Effect of contact angle hysteresis on water droplet evaporation from
super-hydrophobic surfaces. Appl. Surf. Sci. 2009, 255, (7), 4056-4060.
35. Onda, T.; Shibuichi, S.; Satoh, N.; Tsujii, K., Super-Water-Repellent Fractal Surfaces. Langmuir
1996, 12, 2125-2127.
36. Zhang, J.; Li, J.; Han, Y., Superhydrophobic PTFE Surfaces by Extension. Macromol. Rapid
Commun. 2004, 25, (11), 1105-1108.
37. Xu, Q. F.; Mondal, B.; Lyons, A. M., Fabricating superhydrophobic polymer surfaces with excellent
abrasion resistance by a simple lamination templating method. ACS Appl. Mater. Interfaces 2011, 3,
(9), 3508-14.
38. Tuteja, A.; Choi, W.; Ma, M.; Mabry, J. M.; Mazzella, S. A.; Rutledge, G. C.; McKinley, G. H.;
Cohen, R. E., Designing superoleophobic surfaces. Science 2007, 318, (5856),1618-22.
39. Tuteja, A.; Choi, W.; Mabry, J. M.; McKinley, G. H.; Cohen, R. E., Robust omniphobic surfaces.
Proc. Natl. Acad. Sci. U.S.A. 2008, 105, (47), 18200-5.
40. Deng, X.; Mammen, L.; Butt, H. J.; Vollmer, D., Candle soot as a template for a transparent robust
superamphiphobic coating. Science 2012, 335, (6064), 67-70.
41. Wong, T. S.; Kang, S. H.; Tang, S. K.; Smythe, E. J.; Hatton, B. D.; Grinthal, A.; Aizenberg, J.,
Bioinspired self-repairing slippery surfaces with pressure-stable omniphobicity. Nature 2011, 477,
(7365), 443-7.
42. Bohn, H. F.; Federle, W., Insect aquaplaning: Nepenthes pitcher plants capture prey with the
peristome, a fully wettable water-lubricated anisotropic surface. Proc. Natl. Acad. Sci. U.S.A. 2004,
101, (39), 14138-43.
43. Levesque A., Binh V.T., Semet V., Guillot D., Fillit R.Y., Brookes M.D., T.P. Nguyen.
Monodisperse carbon nanopearls in a foam-like arrangement: a new carbon nano-compound for cold
cathodes // Thin Solid Films. 2004. 464-465: 308 - 314.
44. Naha S, Sen S, Puri I.K. Flame synthesis of superhydrophobic amorphous carbon surface. Carbon
2007; 45:1696-1716.
45. Z.A. Mansurov. Journal of Materials Science and Chemical Engineering, 2014, 2, PP. 1-6.
46. Z.A. Mansurov. Some applications of nanocarbon materials for novel devices / R. Gross et. al (eds.),
Nanoscale Devices - Fundamentals and Applications, 355-368. 2006 Springer.
47. Z.A. Mansurov, M. Nazhipkyzy, B.T. Lesbaev, I.K. Puri. Synthesis of superhydrophobic carbon
surface during combustion propane. Oil and Gas, (2010); 5; p. 27-33 (In Russian).
48. Z.A. Mansurov, M. Nazhipkyzy, B.T. Lesbayev, N.G. Prikhodko, M. Auyelkhankyzy, I.K. Puri.
Synthesis of superhydrophobic carbon surface during combustion propane// Eurasian chemico-
technological journal. 2012. V. 14: 1 P. 19-23
126
49. Mao C, Liang C, Luo W, Bao J, Shen J, Hou X, Zhao W. Prepara- tion of lotus-leaf-like polystyrene
micro- and nanostructure films and its blood compatibility. J Mater Chem, 19, 9025 (2009).
50. Barthlott W, Neinhuis C. Purity of the sacred lotus, or escape from contamination in biological
surfaces. Planta, 202, 1 (1997).
51. Neinhuis C, Barthlott W. Characterization and distribution of water-repellent, self-cleaning plant
surfaces. Ann Bot, 79, 667 (1197)
52. Wang J, Chen H, Sui T, Li A, Chen D. Investigation on hydro- phobicity of lotus leaf: experiment
and theory. Plant Sci, 176, 687 (2009).
53. Luo C, Zuo X, Wang L, Wang E, Song S, Wang J, Wang J, Fan C, Cao Y. Flexible carbon nanotube:
polymer composite films with high conductivity and superhydrophobicity made by solution
process. Nano Lett, 8, 4454 (2008
54. Yu Y, Zhao ZH, Zheng QS. Mechanical and superhydrophobic stabilities of two-scale surfacial
structure of lotus leaves. Langmuir, 23, 8212 (2007).
55. Celia E, Darmanin T, Taffin de Givenchy E, Amigoni S, Guit- tard F. Recent advances in designing
superhydrophobic surfaces. J Colloid Interface Sci, 402, 1 (2013).
56. Robinson A. The Last Man Who Knew Everything: Thomas Young, the Anonymous Polymath
Who Proved Newton Wrong, Explained How We See, Cured the Sick, and Deciphered the Rosetta
Stone, Among Other Feats of Genius, Pi Press, New York, NY (2006).
57. Gupta N, Kavya MV, Singh YRG, Jyothi J, Barshilia HC. Superhydrophobicity on transparent
fluorinated ethylene propylene films with nano-protrusion morphology by Ar + O2 plasma etching:
study of the degradation in hydrophobicity after exposure to the environment. J Appl Phys, 114
58. Yu E, Lee HJ, Ko TJ, Kim SJ, Lee KR, Oh KH, Moon MW. Hierarchical structures of AlOOH
nanoflakes nested on Si nanopil- lars with anti-reflectance and superhydrophobicity. Nanoscale, 5,
10014 (2013).
59. Wang S, Song Y, Jiang L. Photoresponsive surfaces with controllable wettability. J Photochem
Photobiol C, 8, 18 (2007).
60. Liu JL, Feng XQ, Wang G, Yu SW. Mechanisms of superhydrophobicity on hydrophilic substrates.
J Phys: Condens Matter, 19, 356002 (2007).
61. Herminghaus S. Roughness-induced non-wetting. Europhys Lett, 52, 165 (2000).
62. Zhang X, Shi F, Niu J, Jiang Y, Wang Z. Superhydrophobic surfaces: from structural control to
functional application. J Mater Chem, 18, 621 (2008).
63. Wang FJ, Li CQ, Tan ZS, Li W, Ou JF, Xue MS. PVDF surfaces with stable
superhydrophobicity. Surf Coat Technol, 222, 55 (2013).
64. Liu H, Zhai J, Jiang L. Wetting and anti-wetting on aligned carbon nanotube films. Soft Matter, 2,
811 (2006).
65. Park SJ, Brendle M. London dispersive component of the surface free energy and surface enthalpy.
J Colloid Interface Sci, 188, 336 (1997).
66. Park SJ, Seo MK. Solid-liquid interface. Interface Sci Technol, 18, 147 (2011).
67. F.M Fowkes. Determination of interfacial tensions, contact angles, and dispersion forces in
surfaces by assuming additivity of intermolecular interactions in surfaces. J Phys Chem, 66, 382
(1962).
68. F.M. Fowkes Additivity of intermolecular forces at interfaces. I. Determination of the contribution
to surface and interfacial tensions of dispersion forces in various liquids. J Phys Chem, 67, 2538
(1963).
69. Zhou, H.; Wang, H.; Niu, H.; Gestos, A.; Wang, X.; Lin, T., Fluoroalkyl silane modified silicone
rubber/nanoparticle composite: a super durable, robust superhydrophobic fabric coating. Adv. Mater.
2012, 24, (18), 2409-12.
70. Li, Y.; Li, L.; Sun, J., Bioinspired self-healing superhydrophobic coatings. Angew. Chem. Int. Ed.
2010, 49, (35), 6129-33.
71. Wang, H.; Xue, Y.; Ding, J.; Feng, L.; Wang, X.; Lin, T., Durable, self-healing superhydrophobic
and superoleophobic surfaces from fluorinated-decyl polyhedral oligomeric silsesquioxane and
hydrolyzed fluorinated alkyl silane. Angew. Chem. Int. Ed. 2011, 50, (48), 11433-6.
72. Zhou, H.; Wang, H.; Niu, H.; Gestos, A.; Lin, T., Robust, Self-Healing Superamphiphobic Fabrics
Prepared by Two-Step Coating of Fluoro-Containing Polymer, Fluoroalkyl Silane, and Modified
Silica Nanoparticles. Adv. Funct. Mater. 2012, 23, (13), 1664-1670.
73. Bittoun, E.; Marmur, A., The role of multiscale roughness in the lotus effect: is it essential for super-
hydrophobicity // Langmuir 2012, 28, (39), 13933-42.
74. Verho, T.; Bower, C.; Andrew, P.; Franssila, S.; Ikkala, O.; Ras, R. H. A., Mechanically durable
superhydrophobic surfaces. Adv. Mater. 2011, 23, (5), 673-8.

127
 Chapter 8

SYNTHESIS OF GRAPHENE LAYERS IN FLAME

Graphene is a single layer two-dimensional carbon structure, the surface of which
is laid out in regular hexagons with sides of 0.142 nm and carbon atoms at the
vertices. In its properties, graphene is a semimetal with a small overlap of the
conduction and valence bands. Currently, much attention is paid to graphene because
of its unusual and potentially useful properties such as high electrical and thermal
conductivity, the dependence of electronic properties on the presence of attached
radicals of different nature on the surface of graphene, an adjustable band gap,
quantum Hall effect, extremely high mobility of charge carriers (at room temperature
it reaches 104 cm2/(V·s), which is significantly higher than mobility in silicon), high
elasticity and good electromechanical characteristics. These properties allow using it
in future as the basis for new nanomaterials with improved mechanical, electrical and
thermal physical characteristics, as well as an element of nanoelectronic devices.
However, widespread use of graphene will be possible after detailed studies of its
characteristics, as well as by the development of reliable and relatively inexpensive
methods for synthesis of these structures.

8.1. The history of graphene

The history of graphene goes back to the ancient times. It is in 1859 that Benjamin
Brodie observed a new form of carbon, obtained due to the effect of a strong acid on
graphite (Fig. 1, a) [1]. Brodie called the obtained suspension "carbon acid", and the
new form of carbon with the molecular weight of 33 - "grafon".
At present, we know that he obtained and observed suspension of graphene oxide
crystals [2]. Since then there have appeared only few papers describing the layered
structure of graphite oxide, but it is worthy of note that the existence of individual
atomic planes of graphene oxides was proved [3].
In 1948, G. Ruess and F. Vogt using the transmission electron microscope (TEM)
observed the flakes of graphite oxide with the thickness of several nanometers
obtained after evaporation of the suspension droplet deposited onto the TEM grid [4].
In 1962, Ulrich Hofmann and Hanns-Piter Boehm studied the flakes of reduced
graphite oxide searching for the thinnest layers and determined that some of them are
monolayer ones (Figure 1.1, b) [5]. This work can be considered as the first
observation of graphene, as it ia Boehm and his colleagues who in 1986 introduced the
term "graphene" as a combination of the word "graphite" with the suffix indicating the
polycyclic aromatic hydrocarbons [6].
Subsequently graphene was observed during epitaxial growth on metallic
substrates [7-11] non-conductive carbides [12-15] and graphite [16]. Also, hyperfine
graphite films were obtained by micromechanical exfoliation of graphene with
application of a scotch tape [17], followed by the research on the atomic force
microscope [18], SEM [19], and scanning tunneling microscope [20]. The electrical
properties of thin graphite films were studied revealing the effect of the electric field –
the change in electrical resistance was up to 8% [21-23].

128
 However, all the studies of the produced thin graphite films were directed mainly
at their visualization without any indication of the remarkable properties of graphene.
The researches of electronic properties of thin films by Geim and Novoselov have
enabled the development of a new stage in the life of graphene [24, 25]. They showed
the unique electronic properties of graphene, which gave a new impetus for
researchers to further their studies.

Fig.1. The suspension of yellow layers of graphene oxides floating in water obtained by Benjamin Brodie
in 1859 (a) [1]; layers of hyperfine scales of reduced graphite oxide, obtained by Ulrich Hofmann and
Hanns-Piter Boehm (b) [5]

8.2. Methods of obtaining graphene

8.2.1. Micromechanical exfoliation of graphite

The method of micromechanical exfoliation of graphite is, at first sight, the most
available and simple. However, it is only at first sight, as it is not so easy to obtain a
few layers of graphene on scotch tape. This method requires some experience and
qualification of the researcher. Despite the fact that this method has been used earlier
[17], it has gained the most popularity due to the work of Geim and Novoselov [24-
26]. Graphene sheets are separated from crystalline graphite either due to friction of
small graphite crystals on each other, or with scotch -tape, the subseguent solution of
which results in production of individual graphene layers. Thus, as a result of a
micromechanical effect, a substantial amount of fragments representing graphene
samples with a different number of layers (from one to one hundred) are formed.
This approach also allows separating single-layer graphene sheets having an
ordered structure of 10 um in width and about 100 um in length [27]. The share of
single-layer samples is relatively small, so that the main difficulty of the considered
method of synthesis is associated with the detection of such single-layer samples.

8.2.2. Liquid phase exfoliation of graphite

This method is based on the use of the layered crystal structure of graphite due to
which penetration of molecules or atoms of different nature into the space between the
layers is possible [28]. This increases the distance between the layers and
129
conseguently reduces the energy of interaction between them. As a result, it is possible
to split the graphite layers under the conditions of mechanical action on them.
For this method of graphite separation into graphene sheets we use organic
surface-active agents (surfactants), for example, N-polymethyl-pyrrolidone (NMP),
which is characterized by high energy of interaction with graphene layers. Along with
NMP, such compounds as N, N-dimethylacetamide (DMA), g-butyrolactone (GBL)
and 1.3-dimethyl-2-imidazo lidinon (DMEU) have been successfully used as
surfactants. Prolonged sonication and centrifugation of particulate graphite in a liquid
surfactant leads to formation of suspension containing both multi-layered and
unilamellar graphene sheets [29]. Study of the samples obtained by transmission
electron microscope shows the presence of flat and curved unilamellar graphene sheets
in bilayered and multilayered graphene samples. The transverse dimensions of these
objects do not exceed a few micrometers. The proportion of single-graphene sheets in
the slurry is about 28%, corresponding to a mass fraction of ~ 12 % and total yield of
monolayered graphene sheets (ratio of monolayered sheets to the total weight of initial
graphite powder) ~ 1 %. [30].
The difficulty of obtaining graphene by liquid-phase exfoliation of graphite can be
seen in [31]. The authors obtained high-quality graphene sheets with the content of the
slurry of about 90 % of single-layer graphene sheets of various shapes and sizes. An
average transverse dimension of graphene is ~ 250 nm, and the average height above
the surface of the substrate is 1 nm. However, to obtain this result, the authors used a
rather labor-consuming process. Initially, fine powder graphite was subjected to brief
(1 min) heat treatment at 1000 °C. Then, the mixture containing 10 mg of powder and
microcrystalline NaCl was subjected to further grinding to form a gray powder. To
isolate microcrystalline graphite from the obtained mixture the powder was poured
with water and filtered.
The filter cake was treated during 24 hours with fuming sulfuric acid containing
about 20% of free S03. After removing the acid by filtration the sample was placed
into 40% aqueous hydroxide solution of tetrabutyl aluminum (TBA) (130 ml)
containing 10 ml of N-dimethylformamide (DMF), and exposed for 5 minutes to
sonication. The resulting suspension was left to stand for three days at room
temperature so as to allow TBA molecules to fully penetrate into the interlayer space
of graphite crystals. Then 5 ml of this suspension was mixed with 15 mg of 1.2-
distearoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy (polyethyleneglycol) -
5000] (DSPE-mPEG) and within an hour subjected to RCD, this resulting resulted in
formation of a homogeneous suspension. Centrifugation of the suspension for 3
minutes resulted in formation of the material containing unilamellar graphene sheets.

8.2.3. Oxidation of graphite

The method of separation of graphite layers using chemical oxidants is considered

to be one of the effective methods [32]. When exposing graphite to the effect of strong
gaseous oxidants such as oxygen and halogens, there takes place oxidation of the inner
layers of graphite. This leads to the increase in the interlayer distance in the initial
graphite crystal and the decrease of the interaction energy between the layers. As a
result, this facilitates separation of graphite layers in the liquid phase that allows to
synthesize graphene oxide samples with transverse dimensions of the order of
hundreds of micrometers [33-37]. Subsequent reduction of graphene from the oxide is

130
also carried out using chemical effect of such reducing agents as hydrazine, hydrogen
and others on it.
The method of bulk graphite oxidation accompanied by the increase in the
interlayer distance in the crystal is known from the beginning of XX century [38-45].
The oxidation degree and chemical composition of the formed oxides are determined
by the process conditions of the initial material (graphite) and reagent grade [46-50]. It
is stated that generally the surface of oxidized graphite layers contains hydroxyl and
epoxy groups, while the edges maybe ended with carbonyl and carboxyl groups. The
presence of water molecules greatly effects the oxidized graphite structure, increasing
the interlayer distance in the crystal from the standard value of 0.34 nm to 1.2 nm [51].
The hydrophilicity of graphite oxide allows separation by prolonged stirring or ultra
sonic effect on the sonication to an aqueous suspension containing particulate oxidized
graphite [52, 53].
The most common method of graphite oxidation is the Hummers method [54] based
on the use of an anhydrous mixture of concentrated sulfuric acid, sodium nitrate and
potassium permanganate. Oxidation of graphite by this method is as follows [54]: 100 g of
fine graphite powder and 50 g of sodium nitrate are introduced into 2.3 liters of technical
sulfuric acid, and thoroughly mixed at the temperature of 0 °C. Then, while stirring, 300 g
of potassium permanganate are slowly added to the suspension making sure that the
temperature does not rise above 20 °C. Thereafter, in the suspension during 30 min at 35 °
C, the oxidation reaction is carried out with increasing volume of the mixture and
bubbling. The reaction completed, the suspension acquires a dough-like structure and
grayish brown color. The resulting paste is poured by 4.6 liters of water while stirring and
that results in a rapid gas evolution and the increase of temperature to 98 °C. At this
temperature, the obtaining suspension was allowed to stand for 15 min. After this time, the
slurry is poured by 14 liters of warm water containing 3% hydrogen peroxide, that
contributing to reduction of the residual potassium permanganate and manganese oxide to
a colorless soluble manganese sulfate. As a result of the treatment with hydrogen peroxide,
the slurry acquires a bright yellow color. Filtration of the suspension leads to formation of
a yellow-brown mass the repeated washing of which in a large amount of warm water can
practically remove impurities and allows obtaining a suspension of pure graphite oxide in
water. In order to dewater the material, centrifugation followed by vacuum treatment at 40
° C in the presence of phosphorus pentoxide P2O5 is used. The chemical composition of
the obtained graphite oxide was determined by analyzing the combustion products. The
measurements show that the material contains 47 % of carbon, 28 % of oxygen and 23 %
of water. There are minor impurities of ash (about 2 %).
Graphene oxide layers contained in the aqueous suspension have a significant
negative charge due to attachment of cations present in water. This results in
electrostatic repulsion of graphene layers in water, which facilitates their separation
and increases stability of the suspension. On the basis of this phenomenon, the authors
of [55] demonstrated a relatively simple approach to the problem of isolation of
oxidized graphene samples. Fine graphite oxide obtained by the modified Hummers
method [42-45, 56] was poured by water of the increased purification degree to form a
brown suspension which was then subjected to dialysis in order to completely remove
residues of salts and acids. Concentration of the slurry was 0.05% by weight.
Separation of graphite oxide to graphene oxide (GO) occurred as a result of ultra
sound treatment of the suspension for 30 min. In order to remove particles of graphite
oxide which did not undergo separation, the slurry was subjected for 30 minutes to
centrifugation with the frequency of 3000 rpm. For the purpose of subsequent
reduction of graphene from graphene oxide, 5 ml of suspension was mixed in a vial
131
with 5 ml of water, 5 ml of 35 % aqueous hydrazine solution and 35 ml of 28 %
aqueous ammonia solution. The weight ratio of hydrazine to graphene oxide was 7:10.
After thorough mixing for several minutes the ampoule was placed in a water bath for
one hour at ~ 95 °C. The obtained suspension containing graphene sheets was used for
further studies and preparation of the film. Upon completion of the reduction
procedure, the remains of hydrazine were removed by dialysis.
There is another approach to reduction of graphene from GO [57]: reduction of
graphene in three stages: Stage 1 – Treatment of GO samples with sodium
borohydride for 1 hour at 80°C resulting in the removal of most of the oxygen-
containing groups; Step 2 – sulfonation of the sample with aryldiazonium salt of
sulfanyl acid for 2 hours in an ice bath, stage 3 – treatment with hydrazine for 24
hours at 100°C resulting in the removal of the remaining oxygen-containing groups. In
contrast to [36, 52], where treatment of GO with hydrazine in water leads to
irreversible coagulation and deposition of graphene sheets, in this experiment [57], an
aqueous suspension containing sulfonated graphene remained stable after treatment
with hydrazine for 24 hours. This is due to the presence of negatively charged groups
S03- on the surface of graphene sheets leading to electrostatic repulsion of the sheets in
the reduction process.

8.2.4. Chemical vapor deposition synthesis of graphene

The method of chemical vapor deposition (CVD) is based on thermocatalytic

decomposition of gaseous hydrocarbons on the surface of certain metals with
formation of various nanocarbon structures. This method is widely used for synthesis
of graphene. The sequence of the CVD synthesis of graphene can be seen in the work
[58], where a nickel film with the thickness less than 300 nm was used as a substrate
which plays the role of a catalyst. The film was deposited on a substrate Si02/Si by
electron-beam deposition. Synthesis of graphene was carried out in a quartz tube
through which at 1000 °C the mixture CH4:H2:Ar with the ratio of 50: 65: 200 was
pumped. Upon completion of the synthesis reaction, the samples were cooled in an
argon flow at the rate of ~ 10 °C sec-1 to room temperature.
The experiment shows that such rapid cooling is an important factor preventing
agglomeration of graphene sheets into multilayer structures and promoting separation
of these sheets from the substrate for future use. It is shown that the average number of
graphene layers and the degree of coverage of the substrate are determined by the
thickness of the nickel film and duration of the growth process. Thus, the film
synthesized during 7 minutes on a nickel substrate with the thickness of 300 nm,
contained mainly bilayer graphene samples [58]. To separate samples of graphene
from a nickel substrate, an aqueous solution of ferric chloride FeCl3 (1 M) was used,
which made it possible to carry out the process of etching the substrate without
additional gas evolution or precipitation. After a few minutes of etching, a graphene
film easily separates from the substrate and rises to the surface of solution. After
transferring to the substrate, the residues of nickel film are removed using hydrofluoric
acid. Along with the method of transfer of graphene sheets onto the substrate based on
the use of acid etching, in [58] the authors present a method involving the use of a soft
substrate such as a stamp of polydimethylsiloxane (PDMS). Initially, the PDMS stamp
is pressed against the graphene film grown on a nickel substrate, after the removal of
which by etching using FeCl3, graphene film is attached to PDMS substrate.

132
 The improvement of CVD method allowed to obtain graphene sheets ranging in
size from 1 cm in transverse dimension [59] to 75 cm in diagonal [60]. This result
exceeding tens or hundreds times the earlier achieved indexes allows to see the
prospects of using graphene in nanoelectronics more optimistically.
In paper [59] graphene was synthesized on copper foil 25 um thick at the
temperature of 1000° C in the flow of hydrogen and methane by CVD method. The
obtained graphene film with the area of 1 x 1 cm was investigated by optical,
transmission and scanning electron microscopes, as well as by the Raman scattering. It
is stated that the film has, as a rule, a continuous structure and containes basically a
single-layer graphene (up to 95 % of the total area), sometimes interspersed with two-
layered graphene (3-4 % of the area) and with a large number of layers (not more than
1 %). The studies have shown that the growth of graphene on copper substrate is
completed after 10 minutes. This implies that on the substrate surface coated with a
single layer of graphene film the further growth graphene hardly occurs. This indicates
the catalytic mechanism of growth for realization of which the presence of copper
surface plays an important role [59].
A graphene sheet up to 75 cm in diagonal was obtained in a cylindrical reactor with
the diameter of 20 cm [60]. Rolled into a cylinder a sheet of copper foil of 75 cm in
diagonal was used as a substrate. To ensure uniform temperature distribution across the
substrate surface, a quartz tube with the diameter of 18.75 cm was inserted into the
cylinder formed by the foil. At the first stage of the process the sample for 30 minutes at
the temperature of 1000 °C was blown with hydrogen at the pressure of 90 mTorr at 8 cm3
min-1 (reduced to normal conditions). This led to the increase in grain size in the structure
of copper foil from a few micrometers to about 100 um. Synthesis of graphene was carried
out at the same temperature for 30 min while passing through the reactor a mixture of CH4
and H2 in the ratio of 3:1 at the pressure of 460 mTorr at 32 cm3 min-1. Upon completion
of this procedure, the sample is rapidly (at 10 oC s-1) cooled in the flow of hydrogen at the
pressure of 90 mTorr to room temperature. Separation of the graphene sheet from the
copper substrate was carried out in several stages.
First, the graphene sheet grown on the copper foil by rolling between two rolls
with a slight compressive pressure (~ 2 kg cm-2) attaches to the polymer film.Then the
copper foil was subjected to electrochemical etching in an aqueous solution (0.1 mol)
of ammonium persulfate (NH4)2S2О8, leading to formation of transparent conductive
film of graphene on the polymer basis. After removal of residual copper, with the help
of deionized water the graphene film may be transferred to the polymer substrate to
any other surface. For this purpose, the film together with the was rolled at an elevated
temperature (90-120 °C) between the rolls at 150-200 mm min-1.
The result is a flexible transparent film, the electrical properties of which are
determined by the properties of graphene. Repeating this procedure allows obtaining
laminated graphene sheets with an increased electrical conductivity. Polyethylene
terephthalate (PET) is used as a substrate used for such films. Studies of the Raman
spectra have shown that the films have predominantly a single-layer structure, but
there are some sites of bi-layer and even multi-layer graphene [60].

8.2.5. Obtaining of graphene in an electric arc

The arc discharge is one of the most effective ways of converting crystalline
graphite into surface carbon structure - fullerenes [61] and carbon nanotubes [62, 63].
In [64], arc discharge in hydrogen medium was used for synthesis of graphene.
133
 For production of graphene, graphene oxide samples obtained by Hummers
method served as initial material [41]. The samples were processed in an electric arc
DC in the mixture of argon and hydrogen at atmospheric pressure and the ratio of H2:
Ar = 1: 9. A graphite rod with the diameter of 10 mm and 150 mm in length was used
as a cathode and a rotating graphite cylinder having 8 holes with the diameter of 20
mm and depth of 10 mm, which were filled with powder served as the anode [64]. The
distance between electrodes was kept at the level of 2 mm, and the discharge current
varied between 100 and 150 A. As a result of this treatment there took place reduction
of graphene from graphene oxide. 10 mg of the produced material containing graphite
was poured by 10 ml of N-methylpyrrolidone, and the resulting suspension was
subjected for 2 hours to sonication, and then centrifuged. This led to removal of large
graphite particles and aggregates. The suspension formed by centrifugation was
subjected to filtration and vacuum dried for 24 hours at 60 °C.
Studies of the samples using an atomic force microscope showed a high quality of
the structure of graphene samples produced in electric arc discharge of hydrogen.
According to these studies, the characteristic height of graphene sheets arranged on the
substrate is 0.9-1.1 nm, this indicating their single layered structure. The arc method
allows to synthesize up to 18% (by weight) of the single-layer graphene sheets from
the initial graphite. The obtained samples are characterized by a low oxygen content.
As follows from the results of elemental chemical analysis, the oxygen content in the
synthesized samples of graphene is 6-7 %. As a comparison it can be noted that this
index for graphene sheets produced in the arc discharge of argon is 8.5 %, for the
sheet obtained by the thermal separation of graphite – 10 %, for graphene structures
obtained by the thermal separation of graphite followed by treatment with hydrogen -
6.5 % [65], for graphene samples chemically reduced with hydrazine, - 10 % [66].
Graphene samples obtained in [67, 68] from graphene oxide due to thermal reduction
are characterized by the content of oxygen in the range 5-10 %.

8.2.6. Thermal decomposition of silicon carbide

The method of thermal decomposition of silicon carbide allows synthesizing

single-layer graphene films on the surface of crystal SiC [69-72]. The quality of
graphene depends on perfection of the initial crystal structure, and its cross-sectional
sizes usually do not exceed a few hundred nanometers and are comparable to the size
of the original crystal [73-76]. Furthermore, as the results of observation with an
electron microscope show the surface of the grown film was not uniform in thickness:
the sites with the size of several hundred nanometers coated with single-layer
graphene films are interspersed with areas filled with graphene structures of two or
more layers, as well as the sites not containing graphene.
The authors of [72] as a result of thorough treatment of initial crystal surface
achieved significant improvement in the degree of uniformity of graphene films
produced by thermal decomposition of silicon carbide. Plates of 6H-SiC (0001) with a
high degree of orientation were used as the initial material. In some experiments, a
substrate doped with nitrogen with the concentration of (1-2) x1018 cm-3 was used.
Before the procedure of heat treatment leading to formation of a graphene film, the
samples were etched in hydrogen for 15 min at atmospheric pressure and T = 1550 °C
to remove residual material from the grinding surface.
Synthesis of graphene was carried out for 15 minutes at the temperature of 1500-
2000 °C in a vertically oriented quartz tube with double-walls and water cooling
134
through which argon was slowly pumped. The rise of the temperature to the indicated
values was performed at 2-3 °C-1 and with the same rate the chamber was cooled after
the end of synthesis. Synthesis was carried out either in vacuum or at argon pressure in
the range of 10-900 mbar. The measurements show that, even after exposure to air for
one hour the contact of oxygen in the synthesized graphene films did not exceed 1%.
However, more prolonged exposure to air is accompanied by formation on the
surface of graphene of adsorbed layers of hydrocarbons and water which may be
removed by vacuum heat treatment at 600 °C [72]. In contrast to vacuum synthesis
conditions that do not provide high-quality of graphene, heat treatment of SiC in the
atmosphere of argon (900 mbar) allowed obtaining a film with improved surface
morphology. Higher quality of graphene samples and a larger transverse dimension of
the samples synthesized in the atmosphere of argon, compared with the vacuum
conditions are conditioned by the possibility of carrying out synthesis under argon at
substantially higher temperature (1500 °C instead of 1200 °C under vacuum).
The use of elevated temperatures promotes synthesis of graphene with higher
surface uniformity, as the temperature rise is accompanied by an exponential increase
in the rate of diffusion of defects and eventually opens the possibility for "self-
healing" of defects. Observations made via a low energy electron microscope with a
spatial resolution revealed that the initial stages of 6H-SiC (0001) plate are covered
with homogeneous surface monolayer stripes of graphene.
The typical size of these stripes is much higher than the corresponding value for
samples produced as a result of vacuum annealing. It seems that the size of the stripes
is limited only by the size of steps on SiC substrate. Along with the single-layer
graphene structures, on the edges of the steps graphene sheets consisting of two or
even three layers are also observed. It is concluded that formation of new graphene
layers starts from the edge regions on the substrate surface[72].

8.2.7. Epitaxial growth of graphene on the metallic surface

Along with SiC mono crystals, metal surfaces with a well-ordered crystal structure
are succesfully used as a substrate for epitaxial growth of graphene [77-83]. In [77],
served as a substrate where ruthenium Ru(0001) surface synthesis of graphene was
carried out under the conditions of high vacuum and controlled in situ using the
method of diffraction of slow electrons. This allowed to observe the process of
formation of single-layer graphene sheets and register the time of the second layer
formation. At the heart of our approach to the layer growth of large-area graphene on
the surface of Ru(0001) is the increasing temperature dependence of carbon solubility
in transition metals. At high temperatures (approximately 1150 °C), when solubility of
carbon is high enough, the metal sample is saturated with carbon along the volume.
Slow cooling of the sample to about 825 °C leads to a sixfold reduction of solubility
[83], resulting in liberation of excess carbon on the metal surface. The surface is
covered with extensive (size greater than 100 microns) islands of the film which are
visible in the SEM images. When the cross-sectional sizes of single-layer graphene
sheets reach 100 um on the surface of these structures there take place an intensive
formation of the second layer simultaneously both single layer and bi-layer graphene
sheets are observed on the surface. Observations by electron microscopy indicate that
the film is a single layer epitaxial graphene sheets having a moire structure [78]. It
should be noted that a similar moire structure is also characteristic of graphene
samples obtained by epitaxial growth on the surface of iridium Ir (111) [81, 82]. The
mentioned size of graphene sheets (over 100 micrometers) is significantly higher than

135
the values previously achieved during epitaxial growth on the surface of 4H-
SiC(0001) (less than 1 um) [76], Ru(0001) (approximately 1 um) [79] and Ir(111)
(approximately 1 um) [82].

8.3. Methods of investigation of graphene

8.3.1. Raman spectroscopy of graphene

Raman spectroscopy (RS) is a rapid method for identification of graphene and can
reliably distinguish single-layer graphene from graphite and from a multilayer
graphene. It is stated [84, 85, 86] that for graphene, as well as for graphite, in the
Raman spectrum two major peaks are observed (Fig. 2): G-line, which characterizes
fluctuations of sp2 carbon bonds (~ 1580 cm-1) (graphite-like zone), and 2D-line
(~ 2700 cm-1) which is the overtone of D-line (defect area) (~ 1350 cm-1). The ratio
between the intensity of the G peak (IG) and 2D peak (I2D) (IG/I2D) gives an estimate of
the number of layers. For single-layer graphene, the ratio is less than a unit.
In the spectrum of single-
layer graphene, a sharp symmet-
rical peak of 2D-line is observed
(Fig. 2), while for graphite - an
unresolved broad peak shifted to
the region of long waves with res-
pect to the single-layer graphene.
For investigations, Raman spec-
trometers with different wa-
velengths of laser radiation:
λ = 457, 473, 514.5, 532 and 633
nm are used. There is no fun-
damental difference in the Raman
spectra in the study of graphene
with different wavelengths of la-
Fig. 2. Typical Raman spectra of graphite and single-layer ser, but there is a shift of the peaks
graphene [84] to the larger side with the decrease
wavelength (Fig. 2) [84, 85].
With the increase in the number of layers there takes place the decrease in the
intensity of peak 2D and with the number of layers more than five it does not differ
from graphite (Fig. 3). Thus, Raman spectroscopy can clearly detect the presence of
single-layer graphene up to the five-layer (Fig. 3) [86].
Analysis of the Raman spectra shows that the change in the thickness of a
graphene layer renders an effect mainly on the shape, position and intensity of 2D-
peak (Fig. 3 and 4). The shape of the peak in the spectrum varies in accordance with
the mechanism of double resonance and the course of dispersion dependences of the
energy of electrons and phonons for graphene with the different number of layers form
of 2D-peak is very specific for each number of layers so that it is used for estimation
of this parameter [84-86].

136
 Fig. 3. The change of 2D-lines of Raman spectra of graphene with the increase in the number of layers
for λ = 457 nm [85], 514.5 and 633 nm [84]

Fig. 4. The change in the intensities of G-peak and 2D-peak of Raman spectra of graphene with the
increase in the number of layers for λ = 532 nm [86]

8.3.2. Visual observation of graphene

One layer graphene absorbs no more than 2.3 % of the incident light making it
optically transparent and excludes direct visual observation [87]. However, the use of
dyes allows to observe it in a polarization microscope.
A detailed study of the optical characteristics of a single-layer and multi-layer
graphene carried out in [88] allows to reliably determine the number of layers by the
optical properties of graphene samples. To improve visual observation, a graphene
layer is placed on the glass activated by fluorescent dye [89]. The increase in the
number of layers of graphene results in production of blue colored samples.
The work [86] presents the results of research on visualization and estimation of
the number of graphene layers by the contrast of the reflection spectrum of the
background substrate SiO2/Si (the thickness of SiO2 layer is from 90 to 300 nm) and
graphene. Contrasts of spectra are obtained by calculation:

137
 (λ) = (λ) − R(λ))/R (λ)

where R0 (λ) is the reflection spectrum from the substrate, and R (λ) - from graphene.

b
Fig. 5. The contrast spectra of graphene layers of different thickness (a) and the optical image of all
samples (b) (the arrows indicate the direction of the increasing thickness of graphene) [86]

The results are shown in Fig. 5. The contrast of the spectrum for a single-layer
graphene has a peak centered at 550 nm, that corresponds to the green-and-orange
range and makes it visible (Fig. 5, a). With the increase in the number of graphene
layers to 10 the position of the center of the peak does not change, and it is at 550 nm.
In this case, the contrast for a single graphene layer is about 0.09 ± 0.005 and
increases with the number of layers: for two layers – 0.175 ± 0.005, for three layers –
0.255 ± 0.01 and for four layers – 0.33 ± 0.015. The increase in number of layers
result in the increase in the contrast and the peak of the spectrum shifts to higher
wavelengths (samples a and b, Fig. 5, a). With the further increase in the number of
graphene layers (samples c-f, Fig. 5, b), the contrast becomes negative (Fig. 5,, a),
because the intensity of the reflection spectrum from graphene is higher than the
intensity of the reflection spectrum from the substrate SiO2/Si [86]. The thickness of
the graphene layers can be determined by the formula dl = NΔd, where N is the number
of layers and the thickness of one layer is equal to Δd = 0.335 nm [90, 91].

138
Fig. 6. Comparison of the calculated values of the contrasts of spectra nG (blue curve with triangles), and
nz (red line with circles) and experimental values of contrasts of spectra (thick black lines) with the
increasing of graphene layers from 1 to 10 layers layers of different thickness (a) [86]

Figure 6 shows the results of comparison of the calculated and experimental data
on the contrast for the spectra of graphite with indices of refraction nG = (2.6-1.3i) and
a single-layer graphene with indices of refraction nZ = (2.0-1.1i) [86]. Using this plot
one can determine the number of layers and consequently the thickness of the
graphene layers by the obtained values of contrast spectra. Also, the contrast of the
spectra can be calculated by the formula [86]:

С  0.0046  0.0925 N  0.00255 N 2

where N (≤ 10) is the number of layers of graphene.

8.3.3. Electron microscopic study of graphene

The scanning tunneling microscope allows us to observe all six carbon atoms in a
single-layer graphene that can reliably distinguish it from the multi-layered one.
However, to obtain such images, it is necessary to make a rather complicated
experiment requiring the work in ultrahigh vacuum.
The atomic force microscope is successfully used to measure the shape and
number of layers (from 3 to 10) in multilayer graphenes. As for as a single layer
graphene is concered, in some experiments it is usually seen as a sheet with the
thickness of 0.6 nm or slightly greater (up to 1 nm). High values of graphene
monolayer thickness is explained by the incomplete fit of the substrate layer to the
substrate or by the presence of adsorbed molecules of the solvent (particularly, water
molecules), or other hardly controlled impurities on the surface of graphene [92, 93].
The needle of an atomic force microscope describes the complex as a single object.
Topographic images of graphene layers varying in color, when observed under an
optical microscope, are shown in Fig. 7 [85]. The values of the thickness of graphene
layers were obtained using the function of the altitude distribution. Shown in Fig. 7 a
139
set of graphene layer thicknesses corresponds to the sequential addition of individual
graphene layers.

Fig. 7. Topographical image of graphene flakes of varying thickness and statistical

distribution of thickness over the surface [85]

It should be noted that these results are heavily dependent on sample preparation
and application to the substrate as well as the choice of the substrate [85].

8.4. Synthesis of graphene layers in flame

8.4.1. Synthesis of graphene layers in a hydrocarbon flame

Since the discovery of the first method of isolating graphene based on mechanical
cleavage of graphite layers [3, 24-27], the efforts of many research laboratories
focused on developing new and more effective ways to solve the low volume
production. Many methods, such as liquid-phase oxidation of graphite, chemical vapor
deposition, epitaxial growth on the metal surface, thermal decomposition of silicon
carbide, creation of graphene in an electric arc have all been used to obtain graphene
[27]. However, these methods are very complicated and time-consuming and
predominantly result in low yields of graphene. The search for new, less labor
intensive and more efficient methods of graphene synthesis is an urgent task.
One such possibility may be production of graphene layers in the flame. The process
of flame formation of nanostructures is a fast reaction proceeding usually 10-5 to 10-3
seconds. Flame production is a continuous method, simple, inexpensive, and has the
ability to develop into a mass production technique. Comparative studies with a number of
authors [4-5] confirm that the preliminary process of obtaining graphene in the flame can
successfully compete with its synthesis by chemical vapor deposition (CVD).
This paper presents the results on synthesis of graphene films in a flame: the
parameters of combustion, the type of catalytic substrate and its orientation in the
flame, providing the smallest number of layers with the greatest structural regularity.

140
 8.4.2. Synthesis of graphene layers in the dual flame

In paper [94] the authors present the results on synthesis of graphene layers in a
dual flame at atmospheric pressure a schematic representation of which is shown in
Fig. 8.

Fig. 8. Schematic representation of the dual-flame for synthesis of graphene

layers on a nickel substrate [94]

A nickel plate which was used as a catalist was placed vertically in an alcohol
flame (burner 1), and it had a protective function preventing oxidation of the substrate
by atmospheric oxygen. The alcohol flame was a source of carbon atoms and the
flame from the second burner was a source of heat for carbonization of carbon and
growth of graphene layers on the substrate.

Fig. 9. Graphene layers with their identification by means of SEM (a, b) and TEM (d) [94]

The process took place at the temperature above 800 °C for 10-40 seconds, then
after turning off the burners, the substrate was covered and cooled rapidly to room
temperature. Upon cooling, solubility of carbon decreases and there takes place
141
formation of surface layers of graphene. Varying the temperature conditions of the
experiment, graphene layers with different amounts of impurities were produced [94].
After purification and separation of graphene layers from the substrate they were
examined using an electron microscope. The results are shown in Fig. 9 [94].
The presence of graphene layers (Figure 4.2, a and b) was confirmed by electron
diffraction pattern with characteristic hexagonal packing of carbon atoms (Fig. 9, c).
With the help of TEM of high resolution it was determined that the number of
graphene layers being formed was in the range of 2-5 (Fig. 9, d). It is found that pre-
annealing at 1000 °C for two hours and the thickness of the nickel substrate are key
conditions for formation of high quality graphene layers. Annealing results in the
removal of the oxide layer and the surface contaminants on the nickel substrate, this
leading to synthesis of a single layer graphene, Raman spectrum of which is shown in
Fig. 10(a) (upper spectrum) [94].

Fig. 10. Raman spectra of graphene layers obtained in the flame by CVD method (a),
and with different fuels of the protective flame (b) [94]

The studies on synthesis of graphene layers, where xylene, hexane, acetone,

ethanol, methanol were used as fuel for protective flame, showed that the fuel with a
high carbon content results in formation of graphene layers with a large amount of an
amorphous structure (Fig. 10) [94]. The best quality of graphene layers was observed,
when using ethanol and methanol (Fig. 10) [94].

8.4.3. Synthesis of graphene layers in the diffusion back flame

Of considerable interest are the results on production of graphene layers in the

diffusion back flame on nickel and copper substrates with the thickness of 25
micrometers at atmospheric conditions [95]. For synthesis, a multi-burner flame with a
separate combustion process was used, which is schematically presented in Fig. 11
[95]. As a fuel, hydrogen is used with introduction of CH4, which is a source of carbon
for synthesis of graphene layers.The fuel came out on the edges of the burner and the
oxidizer was in the center. The reverse flame is characterized by lower soot formation
and the presence of unburned fuel undergoing pyrolysis on the substrate.
The experiments were performed in the following order: the substrate was
preliminarily subjected to roasting in rich hydrogen flame for 10 minutes, then the
graphene layers were synthesized in the process of methane feed (at an equivalent
142
ratio of ~ 3) for 5-10 minutes at 950 °C; after putting out the supply of oxygen and
extinction of the flame there occurs cooling of the substrate to room temperature under
the effect of inert gases and fuel [95].
After purification and separation of
the graphene layers from the substrate
with SiO2/Si, they were examined on
the Raman spectrometer (λ = 514.5
nm). Raman spectrum of graphene
layers synthesized on a copper
substrate is shown in Fig. 12 [95].
Graphene layers were transferred
to a quartz substrate (Fig. 12, a) and
studied on the atomic force micros-
cope. The presence of flakes of grap-
hene with the thickness of about 4 nm,
which corresponds to 8-10 layers of
graphene was stated (Fig. 12, a, b). Ra-
man spectrum of graphene layers has
three characteristic peaks D, G and 2D
(Fig. 12, d). IG/I2D peak intensity ratio
is in the range from 1.3 to 1.7 that
characterizes the presence of about 3- Fig. 11. Schematic representation of synthesis of
10 layers of graphene (Fig. 12, c) [95]. graphene layers using a burner with the inverse
The ratio of the peak intensities IG/I2D diffusion flame [95]
in this case is 0.35, which is compa-
rable with the dimensions of graphene layers obtained by other methods [96].

Fig. 12. Graphene layers synthesized in flames on a copper substrate (a) – layers of graphene deposited
on a quartz substrate 1 x 1 cm; (b) – Pictures of graphene layers on SiO2/Si substrate, obtained in an
optical microscope and the height of the graphene layer profile obtained with an atomic force microscope,
(c) – two-dimensional image of the change in the ratio of Raman spectra peaks IG/I2D on the surface with
the size more than 12 x 12 μm; (d) – Raman spectrum of graphene layers on the SiO2/Si substrate with the
thinkness of, an oxide film of 300 nm [95]
143
 The effect of the flame temperature on synthesis of graphene layers on a copper
substrate in the form of changes in the ratio of peak intensities I2D/IG is shown in Fig.
13 [95].
It is shown that temperature is a
determining factor in synthesis of
graphene layers. The increase in
temperature result in the increase in
the intensity of the change in the in-
tensities of 2D-lines of Raman spect-
ra 2D-peak, which characterizes the
decrease in the amount of synthe-
sized layers on a copper plate with
the increase in their quality [95]. The
decrease in temperature result in the
decrease in the intensity of 2D-peak
and the increase in the intensity of D-
peak, that characterizes the increase
Fig. 13. The effect of temperature on the growth of of the defect of graphene layers
graphene layers on a copper substrate by changing
the ratio IG/I2D and a typical Raman spectrum of a being synthesized. Typical Raman
graphitized structure observed at low temperatures spectrum of graphene layers pro-
(shown in the inset) [95] duced at low temperature is shown
in the inset of Fig. 13 [95]. The ef-
fect of the ratio of CH4: H2 on the growth of graphene layers on a copper substrate is
stated: at values from 1:5 to 1:20, there occurs the growth of graphene layers, and at
the value of 1:40 the growth of graphene layers is not observed.
The research results on synthesis of graphene layers on a nickel substrate have shown
that the ratio of the intensities of I2D/IG of typical Raman spectrum shown in Fig. 14
corresponds to 5-10 layers of graphene [95]. TEM image of graphene layers synthesized
on a nickel substrate and the diffraction pattern are shown in Fig. 14 (b) [95].
The defects of graphene layers formed on the nickel substrate (ID/IG=0.1, Fig. 14,
a) were stated to be than those of smaller graphene layers synthesized on a copper
substrate (ID/IG = 0.35, Fig. 12, d), which is due to different mechanisms of growth of
graphene layers on nickel and copper substrates [95]. Different solubility of carbon in
copper and nickel determines a different mechanism of the growth of graphene layers.
An extremely low level of carbon dissolution in copper forms the surface mechanism
of growth of graphene layers on the substrate. Surface growth of graphene on a copper
substrate creates the preconditions for controlling the number of layers on the surface
of the copper substrate. However, it creates certain requirements to the parameters of
the flame. As the synthesis of graphene on a copper substrate proceeds on the surface,
excess soot formation leads to a rapid coating of the surface with soot and ceases
formation of graphene layers.
On a nickel substrate at the temperature above 800 °C, there takes place formation
of a metastable solid phase and, upon cooling, carbon diffuses from nickel onto the
surface to form graphene. This somewhat limits the process of controlling the number
of layers on a nickel substrate. The growth of graphene layers on a nickel substrate can
last longer than that on the copper substrate.
Fig. 15 shows the comparative picture of the growth of graphene layers with the
ratio of CH4: H2 = 1:10 on Cu and Ni substrates by peak 2D of Raman spectrum [95].
For Cu, methane was introduced during 10 minutes at the temperature of 950°C, and
for Ni, methane was injected within 5 minutes at the temperature of 850 °C.
144
 Fig. 14. Raman spectrum of graphene layers synthesized in flames on a nickel substrate (a) TEM image
(b) and the diffraction pattern showing the crystal structure of graphene (b, inset) [95]

It is shown that the values of 2D peak of graphene layers for a copper substrate
does not exceed 2700 cm-1, which characterizes the limit and the formation of several
layers of graphene (see Fig. 15, a). The values of 2D peak of graphene layers on a
nickel substrate reaches 2727 cm-1, indicating the formation of more than 10 layers
(Fig. 15, b) [95].

Fig. 15. Raman peak 2D mapping of 12 um x 12 um with a ratio CH4: H2 = 1:10: a – Raman mapping for
Cu, illustrating the self-limiting growth of several graphene layers b – Raman mapping for Ni, indicating
the growth of 10 layers of graphene [95]

8.4.4. Synthesis of graphene layers in premixed flames

Investigations on formation of layered graphene films were carried out in the

flame comprising a premixed propane-oxygen mixture at atmospheric pressure under
the following conditions: flow rate of propane – 219 cm3/min, oxygen consumption –
381 cm3/min, corresponding to the ratio of C/O=0.86. The studies were also carried
out with the addition of argon in the propane-oxygen mixture in an amount of 300-650

145
cm3/min. To support the growth of graphene, the substrates consisting of either copper
or nickel plates, placed in the central part of the flame were used, Fig. 16 [97].
The residence time of the substrate in the flame was varied: 5, 10,
20, 30, 40, 60 seconds, 5 and 10 minutes, and the angle of the
substrate relative to the vertical axis of the flame was altered: α = 0,
30, 45, 60, 85°. The flame temperature in the experiments was in the
range 900-950 °С. Measurements of the temperature profile in the
central part of the flame were conducted using a chromel-alumel
thermocouple in steps, each measurement point being at the distance
of 5 mm, and a reference point was taken at the base of burner. The
results of measurements of the temperature profile at the height in the
middle of the propane-oxygen flame by adding argon are shown in
Table 4.
At the second stage, the process of formation of layered
graphene films in the flame of premixed propane-oxygen and
butane-oxygen mixtures at atmospheric pressure depending on the
ratio C/О with addition of argon was studied. Argon was fed in the
amount of 150, 250, 350, 450 and 500 cm3/min. The residence time
of the plate in the flame was 5 minutes. Since the flame
temperature is largely dependent on the amount of argon and the
Fig. 16. Photo of ratio C/O, and with placing of the substrate in the flame its
experimental temperature decreases by 30-50 °C. Due to the heat dissipation by
setup [97] the plate and the holder, the initial flame temperature was within
950-970 oC. The flame temperature in the process of synthesis of
graphene on a catalytic substrate was maintained within 900-920 °С.
Table 4

The temperatures at the height of the central part of the propane-oxygen-argon flame

The value of C/O 0.75 0.85 0.95 1.05

Consumption of О2 for С3Н8, cm3/min 438.2 386.6 345.9 313.0
Consumption of О2 for С4Н10, cm3/min 584.3 515.5 461.3 417.3

The study on formation of layered graphene films was carried out at a constant
rate of propane or butane consumption in the amount of 219.1 cm3/min. In order to
obtain the desired ratio C/O, oxygen consumption was varied the values of which are
given in Table 5.
In the experimental study of the influence of the ratio C/O on synthesis of
graphene layers in the combustion of premixed propane-oxigen or butane-oxigen
mixture, the comsumption of argon was constant and was equal to 250 cm3/min.
The flame temperature largely depends on the amount of argon in the given point
of the flame, which explains the variation of the temperature traversing the height of
the flame. Samples of the obtained soot on the substrate were investigated by Raman
spectroscopy (NTEGRA spectra at an excitation wavelength λ= 473 nm).
Analysis of the samples showed that formation of layered graphene films is
preferred, when using a nickel substrate. A copper substrate, which has been used
elsewhere to support the growth of single layers of carbon, is exposed to strong
oxidation conditions in the flame to form an oxide coverage of the surface, which
flakes on cooling.
146
 Table 5

The values of oxygen consumption depending on the type

of fuel to obtain the designed ratio С/О

The flame height, mm Voltage of the thermocouple, mV Temperature, оС

5 36 867
10 37 892
15 36 867
20 39 943
25 41 993
30 40 968

It is stated that the formation of graphene layers on the substrate occurs when it is
placed vertically in the flame, as shown in Fig. 16. The formation of graphene occurs
on both sides of the substrate, Fig. 17 (areas of graphene formation are shown by
arrows).
When tilting the substrate at an angle relative to the vertical axis on the side closer
to the stabilizer, the graphene formation region increases. It is determined that the
zone of formation of graphene layers precedes the border of soot particles, as is seen
in Fig. 17 (gray area). When tilting the substrate at an angle relative to the vertical axis
on the side closer to the stabilizer, the graphene formation region is further increased
(Fig. 17, b).
In the typical spectrum of grap-
hene there are three peaks: the first
peak corresponds to the D-band at
1351 cm-1, the second peak is the G-
band at 1580 cm-1 and the third peak is
the 2D band at 2700 cm-1 [94, 95]. The
ratio of the intensity of the G peak (IG)
and 2D peak (I2D) provides an estimate
of the number of graphene layers
(IG/I2D) [95]. For single-layer grap-
hene, this ratio is less than unity.
The studies reveal that the increase
a b
in the residence time of the substrate in a – straight side of the substrate,
the flame from 5 to 60 seconds corre- b – the reverse side of the substrate
lates to the formation of graphene
layers exhibiting the decrease in the Fig. 17. Photo of nickel substrate with graphene
degree of disorder of their graphitized
structure, characterized by the appearance of 2D peak at 2700 cm-1 and the decrease in
the intensity of D-peak at 1351 cm-1, Fig. 18.
The increase of the residence time of the substrate in the flame to five minutes
results in a stable formation of graphene layers with a decreasing degree of disorder.
The further increase in the residence time does not significantly change the resulting
graphene structures on the substrate.

147
 а b

с
Fig. 18. Raman spectra of samples revealing a graphitized carbon structure on a nickel substrate at
different exposure times (C/O=0.86): a – 5 seconds, b - 30 seconds, c – 60 seconds

Further studies were conducted on a nickel substrate during the exposure time of 5
minutes. Estimation of the influence of the substrate tilt angle relative to the vertical
axis of the flame on formation of graphene layers elucidated the fact that formation of
graphene layers occurs best at the inclination of less than 90о. When the substrate is at
an angle of 90о (perpendicular to the axis of the flame), the formation of an amorphous
carbon black structure results. The changing ratio of IG/I2D, characterized by the
number of graphene layers on the substrate, depends on the angle of the substrate with
respect to the vertical axis of the flame and is shown in Table 6.
It was determined that the minimum number of graphene layers (two or three
layers, IG/I2D=1.1-1.3) occurred in the range of the angle of inclination of the substrate
between 0° and 30°, Fig. 19.
Placing the substrate at an angle of 30о to the vertical axis of the flame yielded two
layers of graphene (IG/I2D = 1.1), as depicted in the Raman spectrum shown in Figure
4.12, b. The further increase in the angle of inclination of the substrate relative to the
vertical axis of the flame (over 30 °) leads to the increase in the minimum number of
graphene layers formed on a substrate, i.e. from 5 to 10 layers (IG/I2D = 1.7-2.4), (Fig.
19, c).

148
 а b

a - three layers (α = 0 , IG/I2D = 1.3), b - two layers (α = 30о, IG/I2D = 1.1),

о

c – five layers (α = 45о, IG/I2D = 1.7)

Fig. 19. Raman spectra of graphene on a nickel substrate (С/О = 0.86, t = 917 оС, τ=5 min)

Table 6

Ratio values of IG/I2D depending on the angle of the nickel substrate relative to the vertical
axis of the flame

The angle of inclination, degrees The minimum value, IG/I2D The maximum value, IG/I2D

0 1.3 2.0
30 1.1 1.9
45 1.7 2.4
60 2.0 2.3
85 2.0 2.2

The studies on formation of graphene layers in the propane-oxygen flame without

addition of argon on a nickel substrate during the exposure time of 5 minutes and the
inclination of 30 о showed that the substrate contained minimum of 5 to 10 layers
(IG/I2D = 1.6-2.5), as revealed in the Raman spectra in Fig. 20.

149
 а b

а – five layers (IG/I2D = 1.6), b – ten layers (IG/I2D = 2.5)

Fig. 20. Raman spectra of graphene layers synthesized on a nickel substrate in propane-oxygen flame
without addition of argon (С/О=0.86, t = 917оС, τ = 5 min)

It is also evident that the zone of formation of graphene layers precedes the border
of formation of soot particles. To confirm this, a series of experiments were conducted
where the flame on a nickel substrate did not reach the area of that soot formation, but
was part of it. After five minutes, the substrate was removed from the flame (Fig. 21).
Preceding the area of particulate structures is a zone of graphenes, shown by the
arrow in Fig. 21. It is confirmed that only the formation of particulate structures on the
surface of the substrate contributes to the formation of graphene layers. Fig. 22 shows
the Raman spectra of the resulting structures as they approach the particulate layer.
The data obtained from Raman spectroscopy of graphene formation on a nickel
substrate allowed to elucidate the size of graphene layers formed both in the vertical
and horizontal directions. It was discovered, that formation of a continuous layer of
graphenes, as a rule, takes place in the vertical direction. The maximum size of the
graphene layer achieved in the vertical direction in these studies was 2 mm, and the
horizontal - one to five micrometers (Fig. 22). However, there is scope to transform
these early results into rapid graphene formation over wider areas using a flame
combustion approach.

a б

Fig. 21. Photo of a nickel substrate with synthesized carbon structure: a – the inner side of the cylinder of
the nickel substrate, b – the outer side of the cylinder of nickel substrate

150
 а
b

a – two layers (IG/I2D = 1.2); b – three layers (IG/I2D = 1.3); c – five layers (IG/I2D = 1.5)

Fig. 22. Raman spectra of samples with a graphitized carbon structure on a nickel substrate

Argon consumption in all the experiments was constant and equal to 250 cm3/min.
The results of the obtained minimum ratio IG/I2D, characterizing the number of
graphene layers on the substrate depending on the type of initial fuel and the ratio С/О
are given in Table 7.

Table 7

The minimum ratio of IG/I2D depending on the ratio of C/O and type of fuel

The value of the ratio, С/О 0.75 0.85 0.95 1.05

Propane, IG/I2D 2.2 1.8 1.5 1.9
Butane, IG/I2D 1.5 2.0 1.3 1.6

It is preferably in the flame of propane and butane under these conditions that
formation of graphene layers in an amount of 5-10 layers was stated to take place. The
results of Raman spectroscopic studies of soot samples are shown in Fig. 23 and 24.

151
 а b

c d

а – 0.75 (IG/I2D = 2.2); b – 0.85 (IG/I2D = 1.8); c – 0.95 (IG/I2D = 1.5); d – 1.05 (IG/I2D = 1.9)

Fig. 23. Raman spectra of the samples with a graphitized carbon black structure obtained on a nickel
substrate in propane-oxygen-argon flames at different C/O

It is found that the graphene layers are formed independently of the type of fuel in
the studied range of ratios C/O with a pronounced peak of graphite G. The tendency
for increasing the degree of disorder of the graphene layers with increasing ratio of
C/O, which is characterized by the increase in the intensity of D-peak is observed. For
propane-oxygen mixture, the degree of disorder is observed at a higher ratio of C/O
than for butane-oxygen mixture. However, with the butane-oxygen mixture at the ratio
of C/O = 0.95, the minimum number of graphene layers was obtained (three layers,
IG/I2D = 1.3).
The effect of argon flow on synthesis of graphene layers with its supply to the
propane or butane-oxygen flames was studied. It is found that at the ratio C/O = 0.75
for propane-oxygen flame, the increase in the argon content from 150 cm3/min to 500
cm3/min does not activate the process of formation of graphene layers, but leads to
formation of soot of predominantly amorphous structure, which is shown in the
Raman spectra presented in Fig. 25.

152
 а b

c d
а – 0.75 (IG/I2D = 1.5); b – 0.85 (IG/I2D = 2.0); c – 0.95 (IG/I2D = 1.3); d – 1.05 (IG/I2D = 1.6)

Fig. 24. Raman spectra of the samples with a graphitized carbon black structure obtained on a nickel
substrate in the butane-oxygen-argon flames at different С/О

a b

с
а – 150 cm3/min (IG/I2D = 2.2); b – 350 cm3/min (IG/I2D = 2.4); c – 500 cm3/min (IG/I2D = 2.2)

Fig. 25. Raman spectra of the samples with a graphitized soot structure obtained on a nickel substrate in
propane-argon-oxygen flame at C/O = 0.75 at various argon flow
153
 A similar situation for propane-oxygen flame is observed at a higher ratio of C/O.
Raman spectra of graphene layers at C/O = 1.05 with a different consumption of argon
are shown in Fig. 26.
In the propane-oxygen flame at the ratio of C/O = 1.05 with the feed of argon 150
cm3/min on a nickel substrate we observed the formation of a minimum number of
graphene layers (Fig. 26, and three layers, IG/I2D = 1.3).

а b

a – 150 cm3/min (IG/I2D = 1.3); b – 250 cm3/min (IG/I2D = 1.9)

Fig. 26. Raman spectra of graphene layers produced on a nickel substrate in a propane-oxygen-argon
flame with C / O = 1.05 at different flow rates of argon

For butane-oxygen flame at C/O = 0.75 the increase in the feed of argon does not
lead to formation of an amorphous soot structure, as observed in propane-oxygen
flame (Fig. 25). Raman spectra of the obtained graphene layers are shown in Fig. 27.
The increase in argon consumption at С/О = 0.75 in butan-oxygen flame result in
formation of minimum number of graphene layers (Fig. 27, b: three layers, IG/I2D =
1.3).

а b

Fig. 27. Raman spectra of the samples with a graphitized carbon black structure obtained on a nickel
substrate in the butane-oxygen flame with C/O = 0.75 at different flow rates of argon:
a – 250 cm3/min (IG/I2D = 1.5); b – 500 cm3/min (IG/I2D = 1.3)

At a greater ratio of С/О > 0.75 for butane-oxygen flame the increase of argon
consumption leads to a greater disorder of graphene layers being formed as observed
154
in propane-oxygen flame. Raman spectra of the obtained soot structures are shown in
Fig. 28.

а b
3 3
a – 250 cm /min (IG/I2D = 1.9); b – 500 cm /min (IG/I2D = 1.9)

Fig. 28. Raman spectra of the samples with a graphitized carbon black structure obtained on a nickel
substrate in the butane-oxygen flame with C/O = 0.95 at different flow rates of argon

However, in butane-oxygen flame this process is observed at a higher flow rate of

argon (500 cm3/min) than for the oxygen-propane flame (250 cm3/min). Earlier, H.
Bockhorn [98] proposed a general reaction scheme for soot formation in a
homogeneous mixture of premixed flames, where intermediate products are polycyclic
aromatic hydrocarbons (PAHs). Then this scheme was developed and modified by
Z.A. Mansurov in [99], where he noted that PAHs can be considered as the basis of
graphene formation in flames. Fig. 29 shows the modified scheme for soot and
graphene formation process in flames.

Fig. 29. Modified scheme for soot and graphene formation process in flames
155
 8.4.5. Synthesis of graphene at low pressure in the flame

A promising method of producing graphene layers are the method of synthesis of

graphene in flames. The studies [100, 101] confirm that the process of obtaining
graphene in flame can compete with a number of existing methods. The process of
formation of graphene in flame is continuous, rapid and inexpensive. The studies
[102] on synthesis of graphene at atmospheric conditions in the flame had shown that
basically on a nickel substrate from 3 to 10 layers of graphene are formed. The studies
on the synthesis of graphene at low pressures in the flame have not been conducted.
The results on synthesis of graphene layers in premixed hydrocarbon flames on a
nickel catalyst substrate at low pressure are reported.
Synthesis of graphene in the flame was conducted at a pressure of 40-100 Torr by
combustion of premixed butane-oxygen mixture with addition of benzene on a nickel
substrate. For this purpose, the burner was placed in a quartz reactor (Fig. 30, a),
which created initial low pressure of 5 Torr. The experiments were performed under
the following conditions: butane flow rate of 450 сm3/min, oxygen flow rate of 740
сm3/min and benzene flow rate of 70 – 120 сm3/min corresponding to the С/О ratio =
0.8-0.9. A nickel plate was used as a catalyst substrate. The flame temperature in the
experiments was in the range of 900-950 оС. The residence time of the plate in the
flame was 3 min. Fig. 30 shows the photograph of a quartz reactor (a) and photo of the
flame with the insert of a nickel substrate (b).
The flame synthesis was ter-
minated by removing the
substrate from the flame. Then
the system was switched off,
the pressure was increased to
the atmospheric pressure, and
the substrate was removed for
further investigation. The pre-
sence of graphene layers on
the catalytic substrate was
evaluated by the presence of
three characteristic peaks
have: a first peak D at 1351
a b сm-1, the second peak G at
a – general view, b – with the substrate placed in the 1580 сm-1 and the third peak
flame 2D at 2700 сm-1 in the Raman
spectra obtained with Raman
Fig. 30. The photograph of the burner used for low-pressure
synthesis of graphene layers in the flame
spectrometer NTEGRA.
Spectra are at the wavelength
λ = 473 nm. The ratio between
the intensity of G peak (IG) and 2D peak (I2D) gives an estimate of the number of
layers (IG/I2D) [101]. For a monolayer of graphene this ratio is less than unity. Raman
spectroscopy allowed us to determine the presence of layered graphene films as well
as to evaluate the quality of graphene layers.
Investigations on synthesis of graphene layers in hydrocarbon flame at low
pressure have shown that the formation of graphene layers is occurred in sooting zone
as well as at atmospheric pressure, as was shown in [102]. Fig. 31 shows the Raman
spectra (a) that characterizing the carbon structures are formed on nickel plate in areas
0, 1, 2, 3 (b). In zones 0, 1 and 2 (Fig. 31, and b) the amorphous carbon structure is
156
observed, but in zone 3 the graphene layers are synthesized. At that, an area of
graphene formation at low pressure is more expanded than at atmospheric pressure.
Above zone 3 there is soot structure is formed (Fig. 31b).

Fig. 31. The Raman-Spectra of carbon structures in accordance with (a) zones; and photo of nickel
substrate (b) with an indication of formed carbon structure zones
(P = 90 Torr, C/O = 0.8, T = 900 °C, t = 30 sec)

Analysis of the results of synthesis of graphene layers in the flame at low pressure
and with addition of benzene showed that in the pressure range of 40-100 Torr one- to
three-layer graphene sheets on nickel substrate are formed, which is characterized by
the intensity ratio IG/I2D ≤ 1.3. The range of IG/I2D ratios in Raman spectra of graphene
samples obtained at various pressures is shown in Table 8.

Table 8

IG/I2D ratios in Raman spectra of graphene samples obtained

at different pressures of С4Н10/O2/C6H6 flame

Pressure value, Torr 45 55 60 65 90

Range of ratios, IG/I2D 0.99-1.27 0.58-1.25 1.1-1.25 1.2-1.35 1.27-1.33
Number of layers 1-2 1-2 2-3 2-3 2-3

It is found that increasing pressure promotes the growth of the larger number of
graphene layers on a nickel substrate, which is associated with the increase in soot
formation. At a pressure of 45-55 Torr preferential formation of a single layer
graphene is observed, its characteristic Raman spectra are shown in Fig. 32.

157
 a b

а – 45 Тоrr (IG/I2D = 0.99), b – 55 Тоrr (IG/I2D = 0.58)

Fig. 32. Raman spectra of single-layer graphene samples obtained with the addition of benzene to
C4H10/O2 flame on a nickel substrate at a pressure [103, 104]

Graphene sheets with 1-3 layers are mainly formed in the range of 40-100 Torr on
a nickel substrate in butane-oxygen flame with addition of benzene. At the pressure of
45-55 Torr preferential formation of a single layer graphene is observed.

8.5. Properties of graphene

Graphene is the only form of carbon (and generally all solid materials) in which
each single atom is in exposure for chemical reaction from two sides (due to the 2D
structure). It is known that carbon atoms at the edge of graphene sheets have special
chemical reactivity, and graphene has the highest ratio of edgy carbons (in comparison
with similar materials such as carbon nanotubes). In addition, various types of defects
within the sheet, which are very common, increase the chemical reactivity [50]. The
onset temperature of reaction between the basal plane of single-layer graphene and
oxygen gas is below 260 °C (530 K) [51] and graphene burns at very low temperature
(e.g., 350 °C (620 K)) [52]. In fact, graphene is chemically the most reactive form of
carbon owing to the lateral availability of carbon atoms. Graphene is commonly
modified with oxygen- and nitrogen-containing functional groups and analyzed by
infrared spectroscopy and X-ray photoelectron spectroscopy. However, determination
of graphene structures with oxygen-[53] and nitrogen-[54] containing functional
groups is a difficult task unless the structures are well controlled.
In 2013, Stanford University physicists reported that sheets of graphene one atom
thick are a hundred times more chemically reactive than thicker sheets [55].

8.5.1. Electronic properties

One of the most useful properties of graphene is that it is a zero-overlap semimetal

(with both holes and electrons as charge carriers) with very high electrical
conductivity. Carbon atoms have a total of 6 electrons; 2 in the inner shell and 4 in the
outer shell. The 4 outer shell electrons in an individual carbon atom are available for
158
chemical binding, but in graphene, each atom is bound to 3 other carbon atoms on the
two dimensional plane, leaving 1 electron freely available in the third dimension for
electronic conduction. These highly-mobile electrons are called pi (π) electrons and
are located above and below the graphene sheet. These pi orbitals overlap and help to
enhance the carbon to carbon bonds in graphene. Fundamentally, the electronic
properties of graphene are dictated by the bonding and anti-bonding (the valance and
conduction bands) of these pi orbitals [111].
Combined research over the last 50 years has proved that at the Dirac point in
graphene, electrons and holes have zero effective mass. This occurs because the
energy-movement relation (the spectrum for excitations) is linear for low energies near
the 6 individual corners of the Brillouin zone. These electrons and holes are known as
Dirac fermions, or Graphinos, and the 6 corners of the Brillouin zone are known as the
Dirac points. Due to the zero density of states at the Dirac points, electronic
conductivity is actually quite low. However, the Fermi level can be changed by doping
(with electrons or holes) to create a material that is potentially better at conducting
electricity than, for example, copper at room temperature.
Tests have shown that the electronic mobility of graphene is very high, with
previously reported results above 15,000 cm2·V−1·s−1 and theoretically potential limits
of 200,000 cm2·V−1·s−1 (limited by the scattering of graphene's acoustic photons). It is
said that graphene electrons act very much like photons in their mobility due to their
lack of mass. These charge carriers are able to travel sub-micrometer distances without
scattering; a phenomenon known as ballistic transport. However, the quality of the
graphene and the substrate that is used will be the limiting factors. With silicon
dioxide as the substrate, for example, mobility is potentially limited to 40,000
cm2·V−1·s−1.

8.5.2. Mechanical Strength

Another of graphene’s prominent properties is its inherent strength. Due to the

strength of its 0.142 nm-long carbon bonds, graphene is the strongest material ever
discovered, with an ultimate tensile strength of 130,000,000,000 Pascals (or 130
gigapascals), compared to 400,000,000 for A36 structural steel, or 375,700,000 for
Aramid (Kevlar). Not only is graphene extraordinarily strong, it is also very light at
0.77 milligrams per square metre (for comparison purposes, 1 square metre of paper is
roughly 1000 times heavier). It is often said that a single sheet of graphene (being only
1 atom thick), sufficient in size enough to cover a whole football field, would weigh
under 1 single gram.
What makes it particularly special is that graphene also has elastic properties being
able to retain its initial size after strain. In 2007, Atomic force microscopic (AFM)
tests were carried out on graphene sheets that were suspended over silicone dioxide
cavities. These tests showed that graphene sheets (with thicknesses of between 2 and 8
nm) had spring constants in the region of 1-5 N/m and a Young’s modulus (different
to that of three-dimensional graphite) of 0.5 TPa. Again, these superlative figures are
based on theoretical prospects of using graphene that is unflawed containing no
imperfections whatsoever and currently very expensive and difficult to artificially
produce, though production techniques are steadily improving, ultimately reducing
costs and complexity [111].

159
 8.5.3. Optical Properties

The ability of graphene to absorb a rather large 2.3 % amount of white light is also
a unique and interesting property, especially considering that it is only 1 atom thick.
This is due to its aforementioned electronic properties; the electrons acting like
massless charge carriers with very high mobility. A few years ago, it was proved that
the amount of white light absorbed is based on the Fine Structure Constant, rather than
being dictated by material specifics. Addition of another layer of graphene increases
the amount of white light absorbed by approximately the same value (2.3%).
Graphene’s opacity of πα ≈ 2.3% equates to a universal dynamic conductivity value of
G=e2/4ℏ (±2-3%) over the visible frequency range.
Due to these impressive characteristics, it has been observed that, once optical
intensity reaches a certain threshold (known as the saturation fluence), saturable
absorption takes place (very high intensity light causes a reduction in absorption). This
is an important characteristic with regards to the mode-locking of fibre lasers. Due to
graphene’s property of wavelength-insensitive ultrafast saturable absorption, full-band
mode locking has been achieved using an erbium-doped dissipative soliton fibre laser
capable of obtaining wavelength tuning as large as 30 nm.
In terms of how far along we are to understanding the true properties of graphene,
this is just the top of the iceberg. Before graphene is heavily integrated into the areas
in which we believe it will excel at, we need to spend a lot more time understanding
just what makes it such an amazing material. Unfortunately, while we have a lot of
imagination in coming up with new ideas for potential applications and uses for
graphene, it takes time to fully appreciate how and what graphene really is in order to
develop these ideas into reality. This is not necessarily a bad thing, however, as it
gives us opportunities to stumble over other previously under-researched or
overlooked super-materials, such as the family of 2D crystalline structures that
graphene has born [111].

8.5.4. Superhydrophobicity of graphene [112]

Graphene is a single-atom-thick sheet composed of sp2- hybridized carbon atoms

and it exhibits many intriguing properties [48,127-130]. A perfect graphene
nanosheet is hydrophobic, but surface treatment or introduction of various functional
groups or additional components will remarkably vary its surface properties.
Graphene has aroused strong interest in the context of exploring novel functional
superhydrophobic surfaces [30]. Several experimental and modeling studies have
focused on exploiting micro-scale surface roughness to engineer superhydrophobic
graphene. Several research groups have fabricated superhydrophobic graphene
surfaces using an irregular stack of graphene oxides (GO) prepared by chemical
oxidation of graphite, and pertinent results are shown in Table 9 [131-144].
Rafiee et al. [131] demonstrated that a surface roughness effect in conjunction
with the surface chemistry of graphene sheets can be used to dramatically alter the
wettability of a substrate. In order to disperse graphene sheets on a substrate, they
performed high-power ultrasonication of graphene sheets in water or acetone. By
controlling the relative proportion of acetone and water in the solvent, the contact
angle of the resulting graphene film can be tailored over a wide range (from
superhydrophobic to superhydrophilic). Such graphene-based coatings with control-

160
lable wetting properties provide a facile and effective means to modify the
wettability of a variety of surfaces.

Table 9
List of graphene-based superhydrophobic surfaces

Method WCA Ref.

Superhydrophobic films Thermal expansion ~160º 131
Conductive, superhydrophobic Graphene/POSS/CNT 155º 132
coatings

Aerogels Fluorinated silane 160 133

Porous graphene PVDF >150 135
PVDF-HFP/graphene microspheres PVDF 151.6±1.4º 136

Graphene/Nafion films Nafion ~161º 137

Biomimetic surfaces- Two-beam laser interference 156.7º 138
superhydrophobicity, and iridescence

Mesoporous graphene for separation CaCO3 microspheres-hard templates,

and absorption surface modification-PDMS, CVD >150º 139

White graphenes- nano-rough films A mixture of boron (B) and boron

trioxide (B2O3) powder-precursor ~152º 140

Amine functionalized GO films Alkylamine of varying chain lengths 162º 141

Superhydrophobic electroconductive
graphene-coated cotton cellulose Dip-pad-dry method 163± 3.4º 142

3D superhydrophobic foams CVD method >150º 143

Superhydrophobic film as corrosion Reduced graphene sheets 150.4° 144

barrier to copper

WCA: water contact angle, POSS: polyhedral oligomeric silsequioxane, PVDF: polyvinylidene fluoride,
HFP: hexafluoropropylene, PDMS: polydimethylsiloxane, CVD: chemical vapor deposition, GO: graphene
oxide.

Lin et al. [133] prepared a novel superhydrophobic graphene aerogel (GA) with
extremely low bulk density and a high WCA. GA offers lower density and simpler
processing than other superhydrophobic surfaces developed using vertically aligned
CNTs or silica aerogels. They demonstrated that the GA is naturally hydrophobic
because of its high surface roughness; following the application of fluorinated
silane, it becomes superhydrophobic with the WCA reaching 160°.
Fluorinated polymers are commonly used as low surface energy materials; for
example, polyvinylidene fluoride (PVDF) is used to produce superhydrophobic
materials [134-136]. Zha et al. [135] demonstrated a combined method of solvent
161
exchanging and freeze-drying to fabricate PVDF porous materials that can effectively
avoid the skin-layer formation of dried materials. It was observed that addition of
graphene (1 wt%) exerted clear influences on the crystallization, morphology,
surface area, and wettability of PVDF. PVDF/graphene (1 wt%) porous materials are
superhydrophobic materials with a large contact angle (>150°).
These properties are due to it being a hybrid porous material composed of strings
of roughened nanospheres (250-300 nm) with hierarchical micro-/nano-scaled
roughness. On the basis of the Cassie-Baxter model, such multi- level surface
roughness account for its superhydrophobicity. Zhang et al. [136] demonstrated the
fabrication of hybrid PVDF-hexafluoropropylene (HFP)/graphene composite micro-
spheres through a spontaneous formation process. Addition of graphene to a PVDF-
HFP/DMF solution has a strong influence on its gel structure in both the wet and dry
states. Morphological characterization by SEM indicates that the dried gel is
composed of PVDF-HFP/graphene (0.25 wt% to PVDF-HFP) microspheres with a
size distribution of 8~10 μm. The contact angles of pure PVDF-HFP and PVDF-
HFP/graphene are 132.7± 2.7 ° and 151.6 ± 1.4 °, respectively. This superhyd-
rophobicity is due to the PVDF-HFP/graphene gel being composed of microspheres
with nanoscaled surface roughness.
Usually, the micro-/macro-scale surface structures of functional films strongly
affect the critical properties, such as wettability, and they should exhibit super water
repellency with a WCA larger than 150 ° for self-cleaning and antimicrobial surfaces.
Choi and Park [137] demonstrated superhydrophobic thin films of graphene-based
materials induced by a hierarchically petal-like structure. The superhydrophobic
graphene/Nafion nano hybrid films were prepared by controlling the structures with
respect to the chemical composition from an interpenetrating networked and
compactly interlocked structure (specific surface area of 9.56 m2/g) to a
hierarchical petal-like, porous structure (specific surface area of ~ 413 m2/g). The
hybrid films revealed a petal-like, porous structure with hierarchical roughness,
where microscale roughness was produced in the lateral direction of the hybrid
sheets while nanoscopic roughness was created on the edges of the hybrid sheets. The
surface morphologies of the hybrids were changed with respect to the amount of
Nafion and their CAs increased from ~97° to ~161° with increasing Nafion content.
Shanmugharaj et al. [141] demonstrated a facile method for the synthesis of
amine functionalized GO films using alkylamines of varying chain lengths.
Alkylamines consisting of hydrophobic long chain alkyl groups and hydrophilic
amine groups were chemically reacted to the GO surface via two types of reactions viz.
1) An amidation reaction between amine groups and carboxylic acid sites of GO, and
2) a nucleophilic substitution reactions between amine and epoxy groups on the
GO surface. Alkylamine-modified GO surfaces showed enhanced roughness, and
this effect was more pronounced with increasing amine chain length. WCA
measurements revealed that the hydrophobic nature of graphene depended on the
chain length of the grafted alkylamines, and this may be corroborated to the decrease
in the surface energy values. Grafting of long chain alkylamines such as
hexadecylamine or octadecylamine on the GO surface followed by thermal reduction
resulted in formation of superhydrophobic surfaces with WCA values of 152 ° and
162 °, corroborating the role of alkylamine chain length in superhydrophobic wetting
control of a thermally annealed GO surface.
Singh et al. [143] used Teflon coated graphene building blocks to create a
superhydrophobic structure with an ordered pore structure of ~200 μm in dimension.
The advancing WCA on the foam surface exceeded 163 °. The graphene foam
162
inherits the pore structure of the Ni foam template used in the CVD growth process.
This method can be used to uniformly tune the pore size and structure of the graphene
foam by selecting the appropriate Ni foam template. The graphene foam can store
elastic strain energy as it is deformed by the impacting drop and then deliver it back
to the drop, which could assist the rebound process.

8.6. Application of graphene

Despite the reigning optimism about graphene-based electronics, “graphenium”

microprocessors are unlikely to appear for the next 20 years. In the meantime, one can
certainly hope for many other graphene-based applications to come of age. In this
respect, clear parallels with nanotubes allow a highly educated guess of what to expect
soon.
The most immediate application for graphene is probably its use in composite
materials. Indeed, it has been demonstrated that graphene powder of uncoagulated
micron-size crystallites can be produced in a way scaleable to mass production [114].
This allows to produce conductive plastics at less than 1 volume percent filling [114],
which in combination with low production costs makes graphene-based composite
materials attractive for a variety of uses. However, it seems doubtful that such
composites can match the mechanical strength of their nanotube counterparts because
of much stronger entanglement in the latter case.
Another enticing possibility is the use of graphene powder in electric batteries that
are already one of the main markets for graphite. An ultimately large surface-to-
volume ratio and high conductivity provided by graphene powder can lead to
improvements in the efficiency of batteries taking over from carbon nanofibres used in
modern batteries. Carbon nanotubes have also been considered for this application but
graphene powder has an important advantage of being cheap to produce [114].
One of the most promising applications for nanotubes is field emitters and,
although there have been no reports yet about such use of graphene, thin graphite
flakes were used in plasma displays (commercial prototypes) long before graphene
was isolated, and many patents were filed on this subject. It is likely that graphene
powder can offer even more superior emitting properties.
Carbon nanotubes were reported to be an excellent material for solid-state gas
sensors but graphene offers clear advantages in this particular direction [116]. Spin-
valve and superconducting field-effect transistors are also obvious research targets,
and recent reports describing a hysteretic magnetoresistance [115] and substantial
bipolar supercurrents [116] prove the major potential of grapheme for these
applications. An extremely weak spin-orbit coupling and the absence of hyperfine
interaction in 12C-graphene make it an excellent if not ideal material for making spin
qubits. This guarantees graphene-based quantum computation to become an active
research area. Finally, it is impossible to forget mentioning hydrogen storage, which
has been an active but controversial subject for nanotubes. It has already been
suggested that graphene is capable of absorbing an ultimately large amount of
hydrogen [117], and experimental efforts in this direction are duly expected.

CONTROL QUESTIONS

1. Can you tell about history of graphene?

2. What kind of methods for producing graphene do you know?

163
3. Can you describe the method of synthesis of graphene by the micromechanical exfoliation of
graphite?
4. What types of methods for identification of graphene do you know?
5. What do you think of synthesis of graphene in flame?
6. Can you describe the difference between synthesis of graphene in flames at atmosphere pressure and
low pressure?
7. What kind of properties does grapheme have?
8. What do you know about application of graphene?

REFERENCES

1. Brodie B.C. Phil. Trans. R. Soc. Lond. 149 249 (1859).

2. Dreyer D.R., Park S., Bielawski C.W., Ruoff R.S. Chem. Soc. Rev. 39 228 (2010).
3. Geim A.K. Uspehi Fiz. Nauk 181 (12) (2011).
4. Ruess G., Vogt F. Monatshefte Chem. 78 222 (1948).
5. Boehm H.P., Clauss A., Fischer G.O., Hofmann U.Z. Anorg. Allgemaine Chem. 316 119 (1962).
6. Boehm H.P., Setton R., Stumpp E. Carbon 24 241 (1986).
7. Grant J.T., Haas T.W. Surf. Sci. 21 76 (1970).
8. Blakely J.M., Kim J.S., Potter H.C. J. Appl. Phys. 41 2693 (1970).
9. Rosei R., De Crescenzi M., Sette F., et al. Phys. Rev. B 28 1161 (1983).
10. McConville C.F., Woodruff D.P., Kevan S.D., et al. Phys. Rev. B 34 2199 (1986).
11. Land T.A., Michely T., Behm R.J., Hemminger J.C., Comsa G. Surf. Sci. 264 261 (1992).
12. van Bommel A.J., Crombeen J.E., van Tooren A. Surf. Sci. 48 463 (1975).
13. Nagashima A., Nuka K., Satoh K., et al. Surf. Sci. 287 609 (1993).
14. Terai M., Hasegawa N., Okusawa M., et al. Appl. Surf. Sci. 130 132 876 (1998).
15. Forbeaux I., Themlin J-M., Debever J-M. Phys. Rev. B 58 16396 (1998).
16. Affoune A.M., Prasad B.L.V, Sato H., et al. Chem. Phys. Lett. 348 17 (2001).
17. Seibert K., Cho G.C., Kutt W, et al. Phys. Rev. B 42 2842 (1990).
18. Ebbesen T.W, Hiura H. Adv. Mater. 7 582 (1995).
19. Lu X., Yu M., Huang H., Ruoff R.S. Nanotechnology 10 269 (1999).
20. Gan Y., Chu W., Qiao L. Surf. Sci. 539 120 (2003).
21. Ohashi Y., Koizumi T., Yoshikawa T., Hironaka T., ShiikiK TANSO 180 235 (1997).
22. Ohashi Y., Hironaka T., Kubo T., Shiiki K. TANSO 195 410 (2000).
23. Dujardin E., Thio T., Lezec H., Ebbesen T.W. Appl. Phys. Lett.79 2474 (2001).
24. Novoselov K.S., Geim A.K., Dubonos S.V., et al. Nature 426 812 (2003).
25. Novoselov K.S., Geim A.K., Morozov S.V.,et al. Science 306 666 (2004).
26. Geim A.K., Novoselov K.S. Nature Mater. 6 183 (2007).
27. Еletsky А.V., Iskandarova I.М., Кnizhnik А.А. Uspehi Fiz. Nauk, 181 (3), 233 (2011).
28. Dresselhaus M.S., Dresselhaus G. Adv. Phys. 30 139 (1981).
29. Hernandez Y. et al. Nature Nanotechnol. 3 563 (2008).
30. Meyer J.C. et al. Solid State Commun. 143 101 (2007).
31. Li X. et al. Nature Nanotechnol. 3 538 (2008).
32. Park S., Ruoff R.S. Nature Nanotechnol. 4 217 (2009).
33. Eda G., Fanchini G., Chhowalla M. Nature Nanotechnol. 3 270 (2008).
34. Dikin D.A. et al. Nature 448 457 (2007).
35. Stankovich S. et al. Nature 442 282 (2006).
36. Li D. et al. Nature Nanotechnol. 3 101 (2008).
37. Li X. et al. Science 319 1229 (2008).
38. Brodie B.C. Ann. Chim. Phys. 59 466 (1860).
39. Schafhaeutl C. Phil. Mag.Ser. 3 16 570 (1840).
40. Staudenmaier L. Ber. Deutsch. Chem. Ges. 31 1481 (1898).
41. HummersW.S.(Jr.), Offeman R.E.J. Am. Chem. Soc. 80 1339 (1958).
42. Kotov N.A., Dekany I., Fendler J.H. Adv. Mater. 8 637 (1996).
43. Cassagneau T., Guerin F., Fendler J.H. Langmuir 16 7318 (2000).
44. Kovtyukhova N.I. et al. Chem. Mater. 11 771 (1999).
45. Hirata M., Gotou T., Ohba M. Carbon 43 503 (2005).
46. He H. et al. J. Phys. Chem. 100 19954 (1996).
47. He H. et al. Chem. Phys. Lett. 287 53 (1998).
48. Lerf A. et al. J. Phys. Chem. B 102 4477 (1998).
49. Cai W. et al. Science 321 1815 (2008).
50. Buchsteiner A., Lerf A., Pieper J.J. Phys. Chem. B 110 22328 (2006).
51. Kim K.S. et al. Nature 457 706 (2009).
164
52. Stankovich S. et al. J. Mater. Chem. 16 155 (2006).
53. Jung I. et al. Nano Lett. 7 3569 (2007).
54. HummersW.S. (Jr.), Offeman R.E. J. Am. Chem. Soc. 80 1339 (1958).
55. Li D. et al. Nature Nanotechnol. 3 101 (2008).
56. Li X. et al. Nature Nanotechnol. 3 538 (2008).
57. Si Y., Samulski E.T. Nano Lett. 8 1679 (2008).
58. Kim K.S. et al. Nature 457 706 (2009).
59. Li X. et al. Science 324 1312 (2009).
60. Bae S. et al. Nature Nanotechnol. 5 574 (2010).
61. Kratschmer W. et al. Nature 347 354 (1990).
62. Iijima S. Nature 354 56 (1991).
63. Iijima S., Ichihashi T. Nature 363 603 (1993).
64. Wu Z.S. et al. ACS Nano 3 411 (2009).
65. Wu Z.S. et al. Carbon 47 493 (2009).
66. Stankovich S. et al. Carbon 45 1558 (2007).
67. Kudin K.N. et al. Nano Lett. 8 36 (2008).
68. Schniepp H.C. et al. J. Phys. Chem. B 110 8535 (2006).
69. Hibino H., Kageshima H., NagaseM.J. Phys. D 43 374005 (2010).
70. Berger C. et al. J. Phys. Chem. B 108 19912 (2004).
71. Berger C. et al. Science 312 1191 (2006).
72. Emtsev K.V. et al. Nature Mater. 8 203 (2009).
73. Zhang Y. et al. Nature 438 201 (2005).
74. Hibino H. et al. Phys. Rev. B 77 075413 (2008).
75. Ohta T. et al. New J. Phys. 10 023034 (2008).
76. Hass J. et al. Appl. Phys. Lett. 89 143106 (2006).
77. Sutter P.W., Flege J-I., Sutter E.A. Nature Mater. 7 406 (2008).
78. Marchini S., Gunther S., Wintterlin J. Phys. Rev. B 76 075429 (2007).
79. Vazquez de Parga A.L. et al. Phys. Rev. Lett. 100 056807 (2008).
80. Pan Y., Shi D-X., Gao H-J. Chinese Phys. 3151 (2007).
81. N'Diaye A.T. et al. Phys. Rev. Lett. 97 215501 (2006).
82. Coraux J. et al. Nano Lett. 8 565 (2008).
83. Arnoult W.J., McLellan R.B. Scr. Metallurg. 6 1013 (1972).
84. Ferrari A.C., Meyer J.C., Scardaci V., et al Physical Review Letters 97, 187401-4 (2006).
85. Obraztsova E.A., Osadchy A.V., Obraztsova E.D., et al. Physica ststus solidi (b) 245 2055 (2008).
86. Ni Z.H., Wang H.M., Kasim J., et al Nano Lett., Vol. 7, No. 9, 2007 s. 2758-2763.
87. Nair R.R., Blake P., Grigorenko A.N., et al. Science 3201308 (2008).
88. Gao L., Ren W., Li F., Cheng H.M. ACS Nano. 2 (8)1625 (2008).
89. Treossi E., Melucci M., Liscio A., et al. Am. Chem. Soc., 131, 43, 15576-15577 (2009). ES
90. Kelly B.T. Physics of Graphite; Applied Science: London, (1981).
91. Dresselhaus M.S., Dresselhaus G., Eklund P.C. Science of Fullerenes and Carbon Nanotubes;
Academic Press: San Diego, CA, 1996; p 965.
92. Katsnelson M. I. Science, 329, 5996, 1157-1158 (2010).
93. Xu K., Cao P., Heath J.R. Science, 329, 5996, 1188-1191 (2010).
94. Li Z., Zhu H., Xie D., et al. Chem. Commun., 47, 3520-3522 (2011).
95. Memon N.K., Tse S.D., Al-Sharab J.F., et all Carbon 9 5064-5070 (2011).
96. Bhaviripudi S., Jia X., Dresselhaus M.S., Kong J. Nano Lett 10:4128–33 (2010).
97. Prikhodko N.G., Lesbayev B.T., Auyelkhankyzy M., Mansurov Z.A. et al. VII Intern. Symp. "Phys.
Chem. of Carbon Materials / Nanoengineering". Almaty (2012).
98. Bockhorn H. (ed). Soot Formation in Combustion. Springer, Berlin/Heidelberg. 1994:596.
99. Mansurov Z.A.Intern. Symp. "Modern chelenges of higher education and science in the field of
chemistry and chemical engineering". Almaty (2013), P. 14-19.
100. Geim, A.K., Novoselov, K.S., (2007). “The rise of graphene”, Nat. Mater., Vol. 6, p. 183–91.
101. Memon, N.K., Tse, S.D., Al-Sharab, J.Fet. et al. (2011). “Flame synthesis of graphene films in open
environments”, Carbon, Vol. 49, p. 5064-5070.
102. Prikhodko, N.G., Lesbaev, B.T., Auelkhankyzy, M., Mansurov, Z.A. (2014) “Synthesis of Graphene
Films in a Flame” Rus. J. of Phys. Chem. B, Vol. 8 (1), p. 61–64.
103. Z.A.Mansurov. From cool flame to nanomaterials synthesis in flames //Zel’dovich Memorials: Ac-
complishments in the combustion science in the last decade/ [Edited by A.A. Borisov and S.M.
Frolov]. – Moscow: TORUS PRESS, 2014. – Vol. 1. 196 p.
104. Z.A.Mansurov. Mechanism of soot and Nanomaterials formation in flame // 8th International seminar
on Flame Structure. 2014. P. 1-4
105. Denis, P. A.; Iribarne, F. (2013). "Comparative Study of Defect Reactivity in Graphene". Journal of
Physical Chemistry C 117(37): 19048. doi:10.1021/jp4061945.

165
106. Yamada, Y.; Murota, K; Fujita, R; Kim, J; et al. (2014). "Subnanometer vacancy defects introduced
on graphene by oxygen gas". Journal of American Chemical Society 136 (6): 2232.
107. Eftekhari, A.; Jafarkhani, P. (2013). "Curly Graphene with Specious Interlayers Displaying Superior
Capacity for Hydrogen Storage". Journal of Physical Chemistry C 117 (48): 25845.
108. Yamada, Y.; Yasuda, H.; Murota, K.; Nakamura, M.; Sodesawa, T.; Sato, S. (2013). "Analysis of
heat-treated graphite oxide by X-ray photoelectron spectroscopy". Journal of Material
Science 48(23): 8171.
109. Yamada, Y.; Kim, J.; Murota, K.; Matsuo, S.; Sato, S. (2014). "Nitrogen-containing graphene
analyzed by X-ray photoelectron spectroscopy". Carbon 70: 59.
110. "Thinnest graphene sheets react strongly with hydrogen atoms; thicker sheets are relatively
unaffected". Phys.org. 1 February 2013.
111. A. Centeno, V.G. Rocha, B. Alonso, A. Fernández, C.F. Gutierrez-Gonzalez, R. Torrecillas, A.
Zurutuza. Graphene for tough and electroconductive alumina ceramics // Journal of the European
Ceramic Society. Vol. 33.15–16, 2013, P. 3201–3210
112. Long-Yue Meng, Soo-Jin Park. Superhydrophobic carbon-based materials: a review of synthesis,
structure, and applications // Carbon Letters Vol. 15, No. 2, 89-104 (2014)
113. Stankovich, S. et al. Graphene-based composite materials. Nature 442, 282-286 (2006).
114. Schedin, F. et al. Detection of individual gas molecules by graphene sensors (cond-mat/0610809).
115. Hill, E.W. Geim, A.K. Novoselov, K. Schedin, F. Blake, P. Graphene spin valve devices. IEEE
Trans. Magn.42, 2694- 2696 (2006).
116. H.B. Heersche, P. Jarillo-Herrero, J.B. Oostinga, L.M.K. Vandersypen, A.F. Morpurgo. Bipolar
supercurrent in graphene Nature (in press, 2007).
117. O. Sofo, A. Chaudhari, G.D. Barber. Graphane: a two-dimensional hydrocarbon (cond-
mat/0606704).

166
 Научное издание

SOOT FORMATION
Mansurov Zulkhair Aimukhametovich

Компьютерная верстка Г.К. Шаккозовой

ИБ № 8051
Подписано в печать 19.02.2015. Формат 70х100/16.
Бумага офсетная. Печать цифровая. Объем 12,7 п.л.
Тираж 150 экз. Заказ №332. Цена договорная.
Издательский дом «Қазақ университеті»
Казахского национального университета имени аль-Фараби.
050040, г. Алматы, пр. аль-Фараби, 71, КазНУ.
Отпечатано в типографии издательского дома «Қазақ университеті».

167
новые книги издательского дома
«ҚАЗАҚ УНИВЕРСИТЕТІ»

Матакова Р. Тестовые задания по аналитической химии: учебное пособие.

– 2015. – 192 с.
ISBN 978-601-04-0992-7
В учебном пособии приведено 600 тестов закрытой формы с выбором од-
ного правильного ответа по трем разделам дисциплины «Аналитиче-
ская химия» – общий теоретический, качественный и количественный
(химические методы) анализ веществ. Представленный комплекс тес-
товых заданий предназначен для текущей и итоговой внешней про-
верки знаний студентов и его самоконтроля при подготовке учебного
материала.

Дюсебаева М. Методы анализа контроля качества продукции тонкого и ос-

новного органического синтеза: учебное пособие. – 2015. – 128 с.
ISBN 978-601-04-0991-0
В данном учебном пособии рассматриваются методы анализа органи- М. А. Дюсебаева

ческих соединений – хроматографические и спектральные (УФ- и И МЕТОДЫ АНАЛИЗА

К-спектроскопия, спектроскопия Я МР и масс-спектрометрия). Пред-

КОНТРОЛЯ КАЧЕСТВА
ПРОДУКЦИИ
ТОНКОГО И ОСНОВНОГО
ставлены методики решения спектральных задач и предложены при- ОРГАНИЧЕСКОГО СИНТЕЗА

меры их использования для установления строения органических Учебное пособие

соединений. Также приведен табличный материал для расшифровки

ОН
Тестостерон

спектров указанными методами. Учебное пособие является руковод- О

ством для самостоятельной работы студентов химических специально-

12
11
13 17
1 16
2 14 15

стей высших учебных заведений, а также оно может быть использовано

9
10 8
3 7
5
4 6 Холестен-4

для исследований при определении строения органических соедине-

Алматы 2015

ний спектральными методами.

Жусупова Г. Химия и химическая технология проантоцианидинов: учебное.

– 2014. – 96 с. Г. Е. Жусупова

ISBN 978-601-04-0808-1 ХИМИЯ

Учебное пособие разработано в соответствии с требованиями программы И ХИМИЧЕСКАЯ
ТЕХНОЛОГИЯ
курса «Химия и химическая технология проантоцианидинов» специ- ПРОАНТОЦИАНИДИНОВ

альности 6D072100-ХТОВ. В данном пособии коротко изложены тео- Учебное пособие

ретические основы строения и химические свойства различных форм он

флаванолов, в том числе их димерных, олигомерных и полимерных

он
о
но
n
он он

форм (проантоцианидины), являющихся важнейшими биологическими

он он
но о

активными соединениями растений. Приведены их физико-химические

он
он

характеристики, оигинальные литературные источники и экзамена- Алматы 2014

ционные вопросы для контроля качества полученных знаний.

Каирбеков Ж., Токтамысов М., Жалгасулы Н., Ешова Ж. Комплексная пе-

реработка бурых углей Центрального Казахстана: монография. – 2014.
– 278 с.
Ж. Каирбеков, М.Т. Токтамысов
Н. Жалгасулы, Ж.Т. Ешова

ISBN 978-601-04-0852-4
ЦЕНТРАЛЬНОГО КАЗАХСТАНА
КОМПЛЕКСНАЯ ПЕРЕРАБОТКА БУРЫХ УГЛЕЙ

КОМПЛЕКСНАЯ
ПЕРЕРАБОТКА БУРЫХ УГЛЕЙ

В монографии обобщены результаты комплексной переработки угля Кияк- ЦЕНТРАЛЬНОГО

КАЗАХСТАНА
тинского месторождения. Приведены результаты работ по разработке
технологии брикетирования угля. Разработана технология получения
гуминового препарата из бурого угля. Подробно проанализированы
Ж. Каирбеков, М.Т. Токтамысов
Н. Жалгасулы, Ж.Т. Ешова

теоретические и прикладные аспекты известных способов получения

гуминовых препаратов и приведены наиболее важные области их прак-
тического применения. Представлены результаты исследования по
разработке технологии получения углещелочных реагентов из бурого
угля. Приведе¬ны результаты по опытно-промышленному испытанию и
определен экономический эффект от внедрения разработанной техно-
логии производства углещелочного реагента.

По вопросам приобретения обращаться в отдел продаж и маркетинга издательского дома

«Қазақ университеті». Контактные тел.: 8 (727) 377-34-11. E-mail: [email protected],
cайт: www.read.kz, www.magkaznu.com