Comparison of Properties of Carbon Particles Formed by Pyrolysis of C3O2 and C2H2 behind Shock Waves

Article abstract of DOI:10.1023/A:1025177614233

Various carbon particles formed by the pyrolysis of C₃O₂ and C₂H₂ behind shock waves at 1200-3800 were investigated. Particles formed from C₃O₂ at 1500-2000 K looked like usual soot from hydrocarbon, e.g., C₂H₂. At 2100-2600 K, the formation rate and yields of particles decreased in all mixtures. At these temperature range, giant film spherical particles with a size of ≤ 700 nm were observed. The high degree of crystallinity of the resulting particles was a specific feature of the high-temperature formation of the carbon particles by C₃O₂ pyrolysis. The acceleration of crystallization with temperature resulted in a decrease in the agglomeration rate, and the size of the resulting particles again fell within 20-30 nm. The structure of crystals consisted of hexagonal layers with a graphite-like spatial arrangement.

Full text of DOI:10.1023/A:1025177614233

Kinetics and Catalysis, Vol. 44, No. 4, 2003, pp. 463–470. Translated from Kinetika i Kataliz, Vol. 44, No. 4, 2003, pp. 509–517.
Original Russian Text Copyright © 2003 by Wagner, Emelianov, Eremin, Jander.
Comparison of Properties of Carbon Particles Formed
by Pyrolysis of C3O2 and C2H2 behind Shock Waves
H. Gg. Wagner*, A. V. Emelianov**, A. V. Eremin**, and H. Jander*
* Institut für Physikalische Chemie, Georg–August Universität, Göttingen, 37077 Germany
** Institute for High Energy Density Associated Institute for High Temperatures,
Russian Academy of Sciences, Moscow, 127412 Russia
Received March 25, 2002
Abstract—Various carbon particles formed by the pyrolysis of C3O2 and C2H2 behind shock waves in the tem-
perature range 1200–3800 K are studied. The formation of the condensed carbon particles is observed directly
by the multichannel detection of the time profiles of the extinction of the medium in the UV, visible, and near-
IR spectral regions. The samples of carbon material deposited on the walls of a shock tube after an experiment
are analyzed using transmission electron microscopy with different resolutions and electron microdiffraction.
Particles formed from C3O2 and C2H2 at 1500–2000 K are 10–30 nm in size and look like usual soot. The
absence of molecular hydrogen in C3O2 only results in faster formation and graphitization. At 2100–2600 K,
the formation of particles is retarded, and the yield of the carbon particles decreases for both substances. After
experiments on pyrolysis of C3O2 at these temperatures, giant spherical particles up to 700 nm in size are found
on the walls of the shock tube. Carbon particles formed at the highest temperatures (2700–3200 K) in C3O2 pyrol-
ysis have the high degree of crystallinity of particles.
INTRODUCTION
for the particles formed from hydrocarbons. Further-
more, the optical properties of these particles measured
behind the shock wave differ substantially from those
of usual soot [7]. Finally, the samples taken from the
walls of a shock tube after experiment sometimes were
white or light yellow [5, 6] and sometimes looked like
solidified droplets [8].
These results raised the questions, what is the differ-
ence between the mechanism of formation of particles
during the pyrolysis of carbon-containing hydrogen-
free molecules and the mechanism of soot formation
from hydrocarbons? Can this difference result in differ-
ent structures of the carbon particles?
Based on microdiffraction analysis, Frenklach [5]
explained the white color of particles obtained during
CCl4 pyrolysis by the presence of carbyne crystals. On
the other hand, electron microscopic analysis of the
particles obtained by C3O2 pyrolysis [8] showed that
they had a layered crystalline structure with a distance
between layers of ~0.25–0.35 nm, indicating that they
belong to graphite.
The main purpose of this work was to examine the
spectroscopic and structural properties of carbon nano-
particles formed by C3O2 and C2H2 pyrolysis in a wide
temperature range.
The formation of carbon particles in high-tempera-
ture gas reactions is a complicated physicochemical
process. Growing particles have different sizes and
structures, and the kinetics of their formation is rather
complicated. These phenomena have been reported in
many papers, most of which are devoted to the forma-
tion of soot particles in flames and in hydrocarbon
pyrolysis. Various hypotheses, which are under con-
stant discussion and development, were proposed for
the interpretation of the specific features of the pro-
cesses [1, 2].
The possibility of obtaining carbon nanoparticles
with certain properties (fullerenes, nanotubes, car-
bynes, and ultradispersed diamonds) in high-tempera-
ture gas reactions is of special interest. These processes
remain unstudied, and further experimental studies are
needed to obtain a general concept of this diverse phe-
nomenon.
Data of experiments on the pyrolysis of hydrocar-
bons in shock tubes make a substantial contribution to
the understanding of the kinetics of formation of carbon
particles [3, 4]. Pyrolysis of different carbon-contain-
ing molecules has recently been studied in the absence
of hydrogen (CCl4, C3O2), and the condensation of car-
bon vapor to small particles was simulated [5–9]. In
these experiments, the authors observed a substantial
yield of the particles in a wide temperature range from
1200 to 3500 K. The induction period for the formation
EXPERIMENTAL
Experiments were carried out in a standard shock
of carbon particles from CCl4 and C3O2 was shorter and tube with an inner diameter of 70 mm and a low-pres-
the rate constant of their growth was higher than those sure chamber 4.5 m long. The formation of condensed
0023-1584/03/4404-0463$25.00 © 2003 MAIK “Nauka /Interperiodica”

464
WAGNER et al.
carbon particles was monitored by the multichannel
detection of the time profiles of the optical properties of
the medium in the UV, visible, and IR spectral regions.
With this purpose, the attenuation of the transmitted
radiation (extinction) and radiation of the medium
(emission) were measured in each experiment at 400,
473, 516, and 633 nm and at 1.31 µm. The latter was
carried out by the probing radiation modulation method
with a frequency of 300 kHz. The detection of emission
signals made it possible to take into account the contri-
bution of the intrinsic radiation of the medium to the
measured extinction level. A high-pressure discharge
xenon lamp and He–Ne (633 nm) and IR-diode
(1.31 µm) lasers served as radiation sources. Interfer-
ence filters, photoamplifiers, IR detectors, digital oscil-
lographs, and computers were used for the detection of
signals.
RESULTS
1
Time and Spectral Extinction Profiles
The time profiles of the extinction measured in the
pyrolysis of all molecules under study were processed
using the Lambert–Beer law by the equation determin-
ing the optical density (D) at a certain wavelength λ as
a function of time and weakening of the probe signal:
ln(Iλ (t)/Iλ (0))
------------------------------------
D(t) = –
.
(1)
[C]
Here Iλ(t) and Iλ(0) are the intensities of the transmitted
light experimentally measured at the time moments t
and 0, respectively. For the convenience of comparing
data from different experiments, the optical density in
formula (1) was attributed to the maximal achieved
concentration of the carbon atoms [C].
Mixtures consisting of 0.2–2.0% C3O2 or C2H2
diluted with argon were subjected to pyrolysis. Mea-
surements were carried out behind reflected shock
waves in a wide temperature range (1200–3800 K) at
pressures from 10 to 60 atm. The parameters of gases
behind shock waves were calculated from the rate of
the incident shock wave taking into account the real
thermodynamic parameters of the mixtures used. High-
temperature (2500–3800 K) experiments were carried
out at different distances between the measured cross-
section and the edge of the shock tube, which were var-
ied from 4 to 138 mm. The residence time of gas behind
the incident shock wave (first heating) before the arrival
of the reflected shock wave (secondary heating) varied
from 10–15 µs to 1000–1500 µs.
After each experiment, the samples of particles col-
lected from the walls of the shock tube were analyzed
using a transmission electron microscope (TEM) with
low and high resolutions and using electron microdif-
fraction measurements using a Philips ELMI 420
instrument. For TEM measurements, the samples were
placed on a copper grid covered by a carbon film 3 mm
in diameter with 400 meshes. This type of grid allows
one to observe particles on the edges of the free space
of the lattice.
A typical example of such experiments is shown in
Fig. 1. It is clearly seen that an increase in the extinction
in the UV spectral region is faster than those in the vis-
ible and, especially, IR regions. That is, the particle
growth is accompanied by the extension of the extinc-
tion spectrum from the short-wave to the long-wave
region. This phenomenon is observed for the particles
formed by C2H2 and C3O2 pyrolysis in the 1500–2100 K
temperature range. Thus, the primarily particles are
“transparent” in the visible and IR spectral regions and
absorb only in the UV region. At the end of the experi-
ment, the particles become completely “black.”
At temperatures higher than 2000–2200 K, the situ-
ation changes. The rate of extinction growth in the UV
region still increases with temperature. However, the
extinction in the visible and IR regions increases more
slowly, and the resulting spectrum is different. The
extinction spectra of the particles obtained from C3O2
measured at different times and temperatures are shown
in Fig. 2. It is seen that the resulting spectrum of the
carbon particles, which are formed very rapidly (within
<15 µs) at T = 2214 K (curve 4) differs from the spec-
trum of particles formed at T = 1646 K for >300 µs
(curve 3). The maxima in the spectra observed at λ =
516 nm in all plots are explainable by the Swan bands
for the C2 radicals.
The particles were analyzed as follows.
1. Measurements by low-resolution TEM (LRTEM)
with an amplification of m < 100000, which allow one
to see the shape of particles and estimate their sizes.
These measurements give the particle size distribution,
the average diameter (d), and the root-mean-square
deviation (σ).
2. Measurements using high-resolution TEM
(HRTEM) with an amplification of 200000 ≤ m ≤
500000 make it possible to observe the fine structure of
a specific particle, the region and state of crystalliza-
tion, and the planes of the layers and distances between
them.
Final Extinction Level
The next step in data processing was the analysis of
the temperature plot of the final extinction level. The
lnImax(t)/I(0)
plots of the final extinction Dmax = ---------------------------------
[C]
vs. í at λ = 633 nm and λ = 1.31 µm for C2H2 and C3O2
pyrolysis are shown in Fig. 3. It is seen that for C3O2 the
plot represents the curve with two maxima whose posi-
tions differ at different wavelengths. For the pyrolysis
3. Electron microdiffraction measurements (EMD)
confirm the crystal structure when the Bragg conditions
for interference are fulfilled.
1
Henceforth, extinction implies the attenuation of the radiation
transmitted through the medium, which is the sum of two pro-
cesses: intrinsic absorption and scattering.
KINETICS AND CATALYSIS Vol. 44 No. 4 2003

COMPARISON OF PROPERTIES OF CARBON PARTICLES
465
D
2
400
473
1
0
516
633
1310 nm
50
100
150
200
250
Time, µs
Fig. 1. Time profiles of the optical densities measured simultaneously at different wavelengths (indicated in the curves) from the
UV to IR spectral regions. Experimental conditions: 1%C O in Ar, T = 1548 K P = 28 atm. Initial peaks: schlieren signals of the
3
2
front of the reflected shock wave.
of a 1%C2H2–Ar mixture (Fig. 3b), the second extinc-
tion maximum at high temperatures was not observed.
However, the second bell-shaped maximum was
observed in preliminary experiments [10] for a
Several different types of particles were observed at
low resolution (LRTEM). Spherical particles agglom-
erated in chains or more compact structures with differ-
ent sizes were the most typical. These particles were
2%C2H2–Ar mixture. It is noteworthy that the extinc- observed at all temperatures, and their average diameter
dg and the root-mean-square deviation σg were deter-
mined. The results are presented in Table 1. The aver-
age diameter of the particles differs at three different
temperatures, and dg at T = 2214 K is four- to fivefold
tion decreases at temperatures higher than 2100 K for
both substances (and in other systems) at all wave-
lengths.
Kinetics of the Optical Density Increase
D
Figure 4 represents the Arrhenius plots of the rate
constants of the optical density increase kλ calculated
under the assumption that the first-order kinetic equa-
tion is fulfilled [8]:
2
dD(t)
dt
-------------- = kλ (Dmax – D(t)).
(2)
For both mixtures the plots lnkλ = f(1/T) at λ = 633 nm
and 1.31 µm (curves 2 and 3) have a break in the 1900–
2200 K temperature range. At a higher temperature, the
optical density increase at these wavelengths retards
with temperature. At the same time, the rate constant of
optical density increase measured at λ = 400 nm (curve 1)
continues to rise with temperature.
1
3
4
2
1
0
400
600
800 1000 1200 1400
Analysis of Particles Using TEM and EMD Methods
λ, nm
The solid carbon materials, which formed by C3O2
pyrolysis at 1650–3000 K and deposited on the walls of
the shock tube, were studied with an electron micro-
scope. The materials were predominantly black but in
some high-temperature (T ≥ 3000 K) experiments the
samples were dark brown.
Fig. 2. Extinction spectra of the particles obtained by C O
3
2
pyrolysis at different times and temperatures. Dotted lines
show the current spectra measured at T = 1646 K at the time
moments t = (1) 60 and (2) 150 µs. Solid lines show the
spectra of the resulting particles measured at T = (3) 1646
and (4) 2214 K in the last, stationary step of the process.
KINETICS AND CATALYSIS Vol. 44 No. 4 2003

466
WAGNER et al.
lnkλ [s–1]
Dmax
13
12
11
10
9
6
(‡)
(‡)
5
4
3
2
1
0
1
2
3
1
2
8
11
10
(b)
(b)
2
1
0
9
8
7
3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5
1/T × 104, ä–1
1500
2000
2500
3000
3500
4000
T, ä
Fig. 3. Temperature plots of the maximal optical density
(D ) at (1) 633 nm and (2) 1.31 µm observed for (a)
Fig. 4. Arrhenius plots of the rate constant of the optical
density increase at (1) 400 nm, (2) 633 nm, and (3) 1.31 µm
3 2 2 2
curves were drawn using the least-squares method.
max
1%C O –Ar and (b) 1%C H –Ar mixtures. At T > 2500 K
3
2
2 2
the measurements were carried out at the distance L =
for (a) 1%C O –Ar and (b) 1%C H –Ar mixtures. The
59 mm from the edge of the shock tube.
greater than those at T = 1646 and 3002 K; that is, the ticles. The whole particle is surrounded by this crystal-
temperature plot of dg has a maximum in this tempera- line structure consisting of four to six atomic layers.
ture range. The same behavior of the particles can be When the temperature of particle formation increases
observed by HRTEM analysis of the samples. The (T > 2300 K), these two different types are retained, and
structures of the samples obtained at three different their regions of crystallization increase. The develop-
temperatures are shown in Fig. 5. The particles formed ment of the regions of crystallization is shown in Fig.
at 1646 K (Fig. 5a) have no crystalline structure and 5d. At T = 3002 K they look like developed crystalline
form chains from almost spherical primary particles. structures with different lengths, which sometimes con-
The samples have a developed surface and a multilevel sist of two or three atomic layers.
structure of particular particles. This result reflects the
Particles obtained at T < 2000 K are amorphous and
cannot be studied by EMD. Among the particles
amorphous nature of the particles, which is also con-
firmed by microdiffraction studies.
formed at T > 2000 K, only carbon materials obtained
Particles of two radically different types are formed by C3O2 pyrolysis were subjected to EMD analysis. For
at T = 2214 K (Figs. 5b, 5c). Particles of one type T = 2214 K, the diffraction pattern could only be
(Fig. 5b) have pronounced spherical components obtained for the giant particles, whereas for T = 3002 K
whose size is much greater than the size of particles EMD measurements were possible for both types of
observed at T = 1646 K (see Table 1). The HRTEM particles, which is explained by the extended region of
observations show that these large particles have a crystallization in the latter. The microdiffraction pattern
developed multilayer surface and are similar to those is shown in Fig. 6 and demonstrates the structure of the
formed at 1646 K. However, they contain some small pronounced rings corresponding to the interplanar dis-
regions of crystallization. The particles of the other tances d002, d110, and d100. The results of EMD measure-
type obtained at this temperature (Fig. 5c) are giant ments are presented in Table 2 along with data for
(d ~ 700 nm). They look like thin films where the crys- graphite [11], carbyne [12], and testing measurements
tallization threads pass through the whole region of par- of black carbon for comparison.
KINETICS AND CATALYSIS Vol. 44 No. 4 2003

COMPARISON OF PROPERTIES OF CARBON PARTICLES
467
Table 1. Average diameter of carbon particles (dg) and the
root-mean-square deviation (σg)*
Thus, the properties of particles of two different
types were studied: the optical properties of “young”
particles during their formation and the size and struc-
ture of the samples deposited on the walls of the shock
tube after cooling. The main problem was to combine
the results obtained and to draw some conclusions
about the formation of the particles in different sub-
stances and at different temperatures.
T, K
dg, nm
σg
1646
2214
3002
14.0
65.0
22.4
0.18
0.30
0.21
* A 1%C O –Ar mixture.
3
2
DISCUSSION
for all wavelengths and all substances. Note that the
kλ(T) plot measured for C3O2 at 400 nm in the 1500–
2000 K range is very close to the rate constant of C3O2
decomposition [13]. This indicates the substantial role
of the C3 radicals, which exhibit an intense absorption
band at λ = 400 nm under these conditions.
The specific features of formation and the properties
of the carbon particles make it possible to divide the
temperature range under study from 1200 to 3500 K
into several subranges: 1500–2000, 2100–2600, 2700–
3200, and >3200 K. Our experiments confirmed the
presence of two maxima in the temperature plot of the
extinction in the visible (633 nm) and IR (1310 nm)
spectral regions for the particles obtained from C3O2
[8] (see Fig. 3a). Let us compare these curves with data
from other measurements and attempt to obtain some
idea of the real properties of particles.
The low-temperature range 1500–2000 K is the
most studied. This is the range of the highest yield of
soot in flames and in experiments on hydrocarbon
pyrolysis in shock waves [3]. The results of this work
also showed a substantial yield of particles in this
range. The extinction maximum at different wave-
lengths was observed for the C3O2 mixture at T ~ 1600–
1700 K and for the C2H2 mixture at T ~ 1800–1900 K.
The temperature dependence of the rate constants of
extinction (usually named the “rate constants of particle
formation”) in the range 1500–2000 K are described by
Particles obtained by C3O2 pyrolysis are similar in
size and structure (TEM data) to the soot particles
obtained from C2H2 and by the pyrolysis of other
hydrocarbons (see Fig. 5). Nevertheless, the optical
properties of the particles from C3O2 differ from the
optical properties of soot formed from hydrocarbons.
Dörge et al. [7] pointed to the fact that the molar
absorption coefficients of these particles are two- to
threefold higher than those for soot formed from hydro-
carbons. To describe these particles, they proposed to
use the refractive index m = n – ik with the real part for
soot n = 1.57 and the imaginary part for graphite k =
1.4. Our observations confirm this conclusion and show
a significant deviation of the spectral extinction coeffi-
cient for the particles formed by C3O2 pyrolysis from
the Arrhenius law with the activation energy ~200 kJ/mol that for usual soot particles formed from C2H2.
50 nm
(b)
30 nm
(‡)
30 nm
100 nm
(c)
(d)
Fig. 5. High-resolution electron microscopic patterns of the carbon particles formed by C O pyrolysis at temperatures (a) 1646,
3
2
(b, c) 2214, and (d) 3002 K.
KINETICS AND CATALYSIS Vol. 44 No. 4 2003

468
WAGNER et al.
time, the ERmax values for the particles from C3O2
decrease sharply, most likely indicating a decrease in
the average size of the particle. This result agrees with
the decrease in the absolute extinction at all wave-
lengths in this temperature range observed for the par-
ticles for C3O2 pyrolysis (see Fig. 3 and [7, 8]).
The next temperature range from 2100 to 2600 K
seems to be the most special. The optical density of par-
ticles formed behind the reflected shock wave begins to
decrease for C3O2 at T ~ 1800 K and for C2H2 at T ~
2000 K (see Fig. 3).As mentioned above, these data can
indicate a decrease in the size of particles formed at
these temperatures. This logical conclusion agrees with
the results on light scattering [4], the measurements of
the particle size by laser-induced incandescence [15],
and the numerical simulation of the formation of the
particles by C3O2 pyrolysis [16].
(002)
(100)
(110)
(004)
One more specific feature of particle formation
observed at these temperatures is known as a break in
the Arrhenius plot of the rate constant of the extinction
growth. In this work, this break is observed for the
pyrolysis of C2H2 at T ~ 1900 K, and for C3O2 it is
observed in the interval 2000–2100 K in the visible and
IR spectral regions, which is in good accord with previ-
ous data [3, 6, 9].
The particles deposited on the walls of the shock
tube after experiments on C3O2 pyrolysis at these tem-
peratures substantially differ from usual soot particles.
They are mainly black or dark brown. After some
experiments, the spots of a bright black substance,
which look like a solidified liquid, were found on the
walls of the shock tube. Contrary to in situ measure-
ments, TEM analysis of these samples shows unusually
large, giant particles resembling film balls with a diam-
eter of up to 700 nm.
Fig. 6. Microdiffraction pattern of the carbon particles
obtained by C O pyrolysis at T = 2214 K. The lines
3
2
denoted by figures in parentheses characterize the interpla-
nar distances (d ) for the particles under study.
hkl
The temperature plots of the optical density (ERmax
at λ1 = 1310 nm and λ2 = 633 nm are shown in Fig. 7.
)
ln(Imax, 1(t)/I(0)1)
ln(Imax, 2(t)/I(0)2)
--------------------------------------------
≡
ERmax
(3)
σext(λ1)
------------------
σext(λ2)
σscat(λ1) + σabs(λ1)
---------------------------------------------
,
≈
=
σ
scat(λ2) + σabs(λ2)
where σext, σabs, and σscat are the cross-sections of
extinction, absorption, and scattering for the finally
formed particles at both wavelengths. One can see a
gradual increase in ERmax with temperature for the soot
In order to eliminate the contradiction between the
particles from C2H2 and the very specific behavior of small size of the particles observed behind the shock
ERmax for the particles formed by C3O2 pyrolysis. In the wave and the giant solidified droplets on the walls after
1500–2000 K temperature range, the data for soot experiment, it should be taken into account that the rate
obtained from C2H2 agree with λ2/λ1 = 0.48, which fol- of formation of particles sharply decreases in this tem-
lows from the Mie theory describing the light absorp- perature range. Therefore, particles observed behind
tion and scattering by small particles [14]. At the same the shock wave for an accessible time of ~1–1.5 ms are
Table 2. Interplanar distances (dhkl) for different carbon materials
Substance
T, K
d002, Å
d100, Å
d100, Å
d100/d110
Graphite [11]
Carbyne [12]
300
300
3.3756
2.204
3.42
2.1386
4.408
2.09
1.234
2.545
1.21
1.20
1.19
1.17
1.733
1.732
1.73
1.73
1.71
1.70
Black carbon (testing substance)
300
C3O2
2214
3002
3002
3.40
2.07
C3O2* (1)
C3O2* (2)
2.96
2.03
3.27
1.99
* (1) Results for spherical particles and (2) results for giant particles.
KINETICS AND CATALYSIS Vol. 44 No. 4 2003

COMPARISON OF PROPERTIES OF CARBON PARTICLES
469
probably far from being in the thermodynamic equilib-
ERmax
rium. These young particles most likely represent linear
carbon polymers (carbynes), which could later form
“liquid clusters and droplets” [17] and are agglomer-
ated during cooling to form giant “consolidated drop-
lets” or “spherules” [18]. It should be kept in mind that
the microdiffraction analysis of giant droplets revealed
no carbyne crystals but indicated their graphite struc-
ture. This fact can be explained by the strong depen-
dence of the stability of carbyne on the rate of its cool-
ing [17]. Evidently, in experiments in the shock tube the
cooling rate is not too high and cooling occurs only in
rarefied waves and during diffusion in the boundary
layer.
0.8
C3O2
2
0.7
0.6
0.5
0.4
0.3
0.2
C2H2
1
The next interesting temperature range is 2700–
3200 K. In this range, the yield of particles was
observed for C3O2 pyrolysis only behind the reflected
shock wave (“two-step heating”). In fact, direct heating
(in the incident wave) did not result in a noticeable for-
mation of particles. This fact was attributed in [8, 19] to
the strong overexcitation of primary carbon clusters
(C2 radicals) at these temperatures, which could retard
the further growth of the clusters. Two-step heating
helps to surmount this phenomenon due to the efficient
formation of small clusters behind the incident wave at
temperatures of 1400–1600 K without overexcitation
of the C2 radicals. Thus, the further growth of the parti-
cles behind the reflected shock wave uses these nuclei.
0.1
1500
2000
2500
3000
3500
T, K
Fig. 7. Temperature plots of the ratios of extinctions mea-
sured (ER ) at λ = 1310 nm and λ = 633 nm for the
pyrolysis of (1) C O and (2) C H at maximum times of 1–
max
1
2
3
2
2 2
1.5 ms. Solid lines show the experimental data processed by
the least-squares method. The dotted line shows the theoret-
ical value ER
≈ λ /λ = 0.48, which follows from the Mie
theor
2 1
theory for the absorption of small spherical particles (d Ӷ λ).
walls reveals the specific features of the formation of
the carbon particles under different conditions.
The optical density at different wavelengths and the
average particle size measured by TEM in this temper-
ature range are again similar to those observed at 1500–
2000 K; that is, in the region of the first maximum in the
temperature plot. A significant difference and strong
crystallization of the samples is shown by microdiffrac-
tion analysis only. The interplanar distances of these
crystals also lie in the interval inherent in graphite.
Nevertheless, d002 and d100 decrease insignificantly (see
Table 2). Due to the strong (and probably, fast) crystal-
lization at high temperatures (2700–3200 K), the
observed extinctions and the resulting size of the parti-
cles coincide with the corresponding values for the par-
ticles formed at low temperatures of 1500–2000 K.
1. Particles formed from C3O2 at standard tempera-
tures of flames (1500–2000 K) look like usual soot from
hydrocarbons (for example, C2H2): spheres 10–30 nm in
size, and the absence of hydrogen results in the faster
formation and graphitization of particles.
2. At 2100–2600 K, the rate of formation and the
yields of particles decrease in all mixtures. After exper-
iments on C3O2 pyrolysis, analysis of the EMD patterns
of the samples of the particles taken from the walls of
the shock tube showed that at these temperatures the
giant film spherical particles with a size of up to 700 nm
were observed. They could be formed due to the fast
agglomeration of small “liquid clusters” during cooling
of the particles on passing through rarefied waves.
The temperature range T > 3200 K is characterized
by a decrease in the yield of particles during C3O2
pyrolysis. The temperature increase most likely results
in an increase in the rate of cluster decomposition. At
T ~ 3500–3800 K, the decomposition rates exceed the
rates of cluster growth [8], and no particles are formed.
3. The high degree of crystallinity of the resulting
particles is a specific feature of the high-temperature
(2700–3200 K) formation of the carbon particles by
C3O2 pyrolysis. The acceleration of crystallization with
temperature leads to a decrease in the agglomeration
rate, and the size of the resulting particles again falls
within the range 20–30 nm. The structure of crystals
observed consists of hexagonal layers with a graphite-
like spatial arrangement.
CONCLUSIONS
A variety of the carbon particles formed by the pyrol-
ysis of C3O2 and C2H2 behind the shock waves were
studied in a wide temperature range (1200–3800 K). A
comparison of the properties of the extinction of the
growing particles with the electron microscopic analy-
Special experiments with monitoring of the cooling
rate of the particles are needed to obtain further infor-
mation on the influence of the agglomeration process
sis data for the samples of particles deposited on the on the resulting structure of the particles.
KINETICS AND CATALYSIS Vol. 44 No. 4 2003

470
WAGNER et al.
9. Wagner, H.Gg., Vlasov, P.A., Dörge, K.Yu., et al., Kinet.
ACKNOWLEDGMENTS
Katal., 2001, vol. 42, no. 5, p. 645.
The authors thank Prof. E.I. Asinovskii for helpful
discussions of the results. This work was supported by
the Russian Foundation for Basic Research and DFG.
10. Deppe, J., Emelianov, A.V., Eremin, A.V., et al., Z. Phys.
Chem., 2000, vol. 214, p. 129.
11. JCPDS, International Center of Diffraction Data, 1997.
12. Borodina, T., Fortov, V., et al., J. Appl. Phys., 1996,
vol. 80, p. 3757.
REFERENCES
13. Friedrichs, G. and Wagner, H.Gg., Z. Phys. Chem., 1998,
1. Frenklach, M. and Wang, H., Proc. Combust. Inst., 1991,
vol. 203, p. 1.
vol. 23, p. 1559.
14. Bohren, C. and Huffman, D., Absorption and Scattering
of Light by Small Particles, New York: Wiley, 1983.
15. Eremin, A., Ziborov, V., Shumova, V., Starke, R., and
Roth, P., WIP Abstr. of XXIX Int. Symp. on Combustion,
Sapporo, 2002, p. 129.
16. Soika, J., Warnatz, J., Vlasov, P., and Zaslonko, I., Com-
bust. Sci. Technol., 2000, vol. 158, p. 439.
17. Asinovskii, E.I., Asinovskii, S.E., Borodina, T.I., et al.,
Karbin na fazovoi diagramme ugleroda (Carbyne on the
Phase Diagram of Carbon). Moscow: OIVT RAN, 2000,
preprint no. 1-449, p. 13.
2. Krestinin, A.V., Proc. Combust. Inst., 1998, vol. 27,
p. 1557.
3. Bauerle, St., Karasevich, Yu., Slavov, St., et al., Proc.
Combust. Inst., 1994, vol. 25, p. 627.
4. Kellerer, H., Koch, R., and Wittig, S., Combust. Flame,
2000, vol. 120, p. 188.
5. Frenklach, M., Combust. Sci. Technol., 1990, vol. 74,
p. 283.
6. Starikovsky, A., Thienel, Th., Wagner, H.Gg., and
Zaslonko, I., Ber. Bunsen-Ges. Phys. Chem., 1998,
vol. 102, p. 1815.
7. Dörge, K.J., Tanke, D., and Wagner, H.Gg., Z. Phys.
Chem., 1999, vol. 212, p. 219.
18. Wittaker, A. and Kintner, P., Carbon, 1985, vol. 23,
p. 255.
19. Wagner, H.Gg., Deppe, I., Emelianov, A.V., Ere-
min, A.V., et al., Dokl. Ross. Akad. Nauk, 2001, vol. 379,
no. 1, p. 63.
8. Deppe, J., Emelianov, A., Eremin, A., et al., Proc. Com-
bust. Inst., 2000, vol. 28, p. 2515.
KINETICS AND CATALYSIS Vol. 44 No. 4 2003