- Email: [email protected]
Effect of benzene–acetylene compositions on carbon black configurations produced by benzene pyrolysis
Chemical Engineering Journal 215–216 (2013) 128–135
Contents lists available at SciVerse ScienceDirect
Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej
Effect of benzene–acetylene compositions on carbon black configurations produced by benzene pyrolysis Kiminori Ono a,⇑, Miki Yanaka a, Yasuhiro Saito a, Hideyuki Aoki a, Okiteru Fukuda b, Takayuki Aoki b, Togo Yamaguchi b a b
Department of Chemical Engineering, Graduate School of Engineering, Tohoku University, 6-6-07 Aoba, Aramaki, Aoba-ku, Sendai, Miyagi 980-8579, Japan ASAHI CARBON CO., LTD., 2 Kamomejima-cho, Higashi-ku, Niigata 950-0883, Japan
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
" Aggregate shapes become complex
with an increase in acetylene concentration. " The number flow rate increases with an increase in furnace temperature. " The mean primary diameter decreases with an increase in furnace temperature. " The shapes most complicate when benzene–acetylene concentration ratio is about 2:1.
a r t i c l e
i n f o
Article history: Received 28 June 2012 Received in revised form 29 September 2012 Accepted 2 October 2012 Available online 10 November 2012 Keywords: Carbon black Soot Pyrolysis Aggregate
The addition of acetylene causes complication of aggregate shapes
500 nm
500 nm
a b s t r a c t A mixture of benzene and acetylene is pyrolyzed in an inert atmosphere to investigate the influence of the benzene–acetylene composition on the configurations of carbon black. The effects of benzene concentration, acetylene concentration, and furnace temperature on the mean primary particle diameter and the aggregate shape in carbon black are investigated. When the acetylene concentration is varied and the benzene concentration is made constant, aggregate shapes become complex with an increase in acetylene concentration. However, in the case where the acetylene concentration is greater than that of the benzene concentration, the variation of aggregate shapes is small with increasing acetylene concentration. The results of this study suggest that nucleation has progressed and aggregate shapes appear complicated when the ratio of the benzene concentration to the acetylene concentration is appropriate (in this study, the ratio is 2:1) and the furnace temperature is high. However, when the benzene or acetylene concentration is high, and the furnace temperature is low, aggregate shapes are simplified because of the formation of small polycyclic aromatic hydrocarbons (PAHs), which contributes to surface growth. Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction Carbon black is a type of soot that is produced industrially and is used for composite materials such as tires and electrode materials of batteries. Although the smallest individual units of carbon black are aggregates, transmission electron microscopy (TEM) images show that these aggregates appear to be formed by spherical particles that are fused together [1]. The aggregate shape is one ⇑ Corresponding author. Tel.: +81 22 795 7251; fax: +81 22 795 6165. E-mail address: [email protected] (K. Ono). 1385-8947/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.cej.2012.10.085
of the factors that affect the properties of composite materials. In the furnace process, carbon black is produced by the continuous pyrolysis of hydrocarbons, which are sprayed into a high-temperature field (1500–2000 K) inside the furnace. The process is complicated owing to the fact that chemical reactions occur rapidly with heat and mass transfer, and therefore, it is difficult to control the aggregate shape. Hence, the aggregation mechanism of carbon black has not yet been completely understood. At present, because the technique used to control the aggregation of carbon black particles is a trial-and-error process, a theoretical solution is required to precisely control the aggregate shape.
129
K. Ono et al. / Chemical Engineering Journal 215–216 (2013) 128–135
Nomenclature A Ac cp d k L M N P Q
projected area (m2) covered area (m2) mean thermal velocity of particle (m/s) mean primary particle diameter (nm) Boltzmann constant (J/K) maximum length (m) mean aggregate size (m2) number of primary particle in an aggregate () perimeter (m) flux of carbon black particles of carbon black particles (1013 particles/s)
Numerous studies have been conducted on the effects of pyrolysis on the diffusion flame and reaction tube in an attempt to explain the formation mechanism of carbon black and soot [2–7]. The formation mechanism for carbon black is considered to be as follows. Large molecules are considered to be the precursors of carbon black particles. Although various theories have been proposed for the reaction mechanism of pyrolysis, all of the recent studies conclude that in both pyrolytic systems and flame systems, polycyclic aromatic hydrocarbons (PAHs) are the precursors of carbon nuclei [2,3]. The nascent soot particles expand either by the addition of molecules from a gas phase, such as acetylene, or by reaction with smaller PAHs [4]. At the start of the reaction, a large number of particles are produced. These particles collide to produce larger spherical particles, which then aggregate into final carbon black clusters. The particles are converted into amorphous carbon and a progressively more graphitic material in the furnace. This is because graphite is the most thermodynamically stable form of carbon [5]. With a long residence time, it is believed that aggregate growth occurs as a result of fusion among primary particles [6]. Shishido et al. have proposed a sintering model [7], which suggests that the rate of sintering among primary particles depends on the size, temperature, surface energy, and viscosity of the particles. However, these mechanisms are still not entirely understood. Over the last few decades, PAH formation, which is considered to be a precursor of carbon black particles in flames, has been extensively studied by several groups [8–16]. Bonne et al. [8] and Homann and Wagner [9] studied the formation of carbon black in premixed flames using various fuels by mass spectrometric analysis of the flames. This provides the concentration profiles of the intermediate products that play an important role during the formation of carbon particles. Arthur and Napier [10] studied the formation of carbon black in reversed and normal diffusion flames by isolating the various intermediate stages in the flames by properly choosing the combustion and quenching conditions. The formation and growth of PAHs are described by the addition of small building blocks, such as acetylene, to a precursor PAH [11]. Wang and Frenklach [12] performed a computational study for the formation and growth of PAHs in laminar premixed acetylene and ethylene flames. They demonstrated that the reactions of n-C4Hx + C2H2 leading to the formation of one-ring aromatics are as important as the propargyl recombination, and hence, the reactions must be included in the kinetic modeling of PAH formation in hydrocarbon flames. Tesner et al. [13–16] analyzed surface areas and the number density of soot synthesized by a mixture of hydrocarbons using TEM. They suggested that the number density of soot which was produced by methane–acetylene–helium mixture increased with an increase in the acetylene concentration [16]. Watanabe et al. [17] suggested that aggregate shapes became complicated as the particle number density increased. Thus, it is expected that the
Qs
T W X Y Z
a qp
flux of carbon black particles of carbon black particles considered flux of the feedstock’s carbon (1013 particles/s) furnace temperature (K) width (m) anisotropy () complexity () covered ratio () aggregate factor () density of carbon black (kg/m3)
composition of the hydrocarbon, especially the addition of acetylene, affects the aggregate shape of carbon black. Although many investigations have been conducted to measure the concentration of PAHs in the flame, the surface area, and the number density of soot synthesized by a mixture of hydrocarbons, little is known about the effect of changes in the aggregate shape on carbon black produced with a mixture of hydrocarbon. In this study, we form carbon black by pyrolysis of a mixture of benzene and acetylene in an inert atmosphere, and the shape parameters of the carbon black are measured by TEM analysis. The micrographs are analyzed by image analysis to precisely evaluate the aggregate shape. The number of carbon black particles produced per unit time is calculated using the yield of carbon black and the shape parameters. Therefore, these experiments enable us to further understand the influence of the composition of benzene– acetylene on configurations of carbon black, such as the aggregate shape and primary particle diameter. 2. Experimental 2.1. Experimental apparatus Carbon black is produced by the thermal pyrolysis of benzene entrained by a N2 carrier flow. The schematic diagram of the experimental set up used in this study is shown in Fig. 1. Saturated benzene gas is prepared by aerating liquid benzene (>99.5%) with primary N2 (>99.999%) in a bubbling apparatus, and 0–5.0 vol% of the benzene feedstock gas is produced by attenuating the satu-
radial
Heater
axial
Pre-Heater
Primary Secondary N2 N2 Acetylene
Filter
800
Water bath
1000
Vacuum
Benzene
Fig. 1. Schematic diagram showing the experimental apparatus for the production of carbon black.
130
K. Ono et al. / Chemical Engineering Journal 215–216 (2013) 128–135
rated gas with secondary N2. The gas entrained benzene is mixed with 0–5.0 vol% of acetylene. The benzene and acetylene concentration in the feedstock gas is measured by gas chromatography (hydrogen flame ionization detector, GC-17A, SHIMADZU), and is confirmed to be the predefined concentration. The feedstock gas is supplied to an alumina tube heated by a pre-heater (length = 800 mm) at 873 K and an electric furnace (length = 1000 mm). The inner diameter of the tube is 16 mm, and the length is 1970 mm. The glass fiber filter (inner diameter = 68 mm) is produced by ADVANTEC, and the trapping efficiency for particles larger than 0.3 lm is better than 99.9%.
Table 1a Experimental condition for different values of (a) benzene concentration. Benzene concentration (vol%) Acetylene concentration (vol%) Furnace temperature (K)
0, 0.10, 3.0, 5.0 3.0 1573
Table 1b Experimental condition for different values of acetylene concentration. Benzene concentration (vol%) Acetylene concentration (vol%) Furnace temperature (K)
1.0 0, 0.50, 5.0 1473, 1673
2.2. Experimental conditions The axial temperature in the reaction tube was measured using an R-type thermocouple in the preliminary experiment. In this study, the radial temperature distribution is ignored, because the experiment is conducted after attaining a steady state. The axial temperature distribution in the reaction tube is shown in Fig. 2. The maximum temperature occurs near the midpoint of the tube, and temperatures near the inlet and outlet of the tube is low. From this trend, the flow velocity may change depending on the axial location owing to the expansion and contraction that result from temperature changes in the flowing process. In this experiment, we assume that the pressure of the feedstock gas is atmospheric pressure, and we calculate the residence time t [s] using the following equation:
t¼
Z
Ai T 1 ðlÞ dl; V0 T0
ð1Þ
where Ai [m2] is the cross-sectional area of the reaction tube, l [m] is the axial location, T0 [K] is room temperature (293 K), T1 [K] is the measured temperature, and V0 [L/min] is the feedstock flow rate at T0. The average temperature zone (length = 350 mm) is defined as the axial temperature area of the preset temperature ±10 K, and the residence time [s] is defined as the time during which the feedstock gas passes through the zone. The feedstock flow rate is 3.0 L/min. The furnace temperature is defined as the average temperature zone and is set to 1473 K, 1573 K, and 1673 K. The residence time calculated by Eq. (1) is 0.318, 0.308, and 0.296 s when the furnace temperature is 1473 K, 1573 K, and 1673 K, respectively. We vary (a) benzene concentration, (b) acetylene concentration, and (c) furnace temperature, as shown in Tables 1a–c, respectively.
1800
Table 1c Experimental condition for different values of furnace temperature. Benzene concentration (vol%) Acetylene concentration (vol%) Furnace temperature (K)
1.0 0.50, 5.0 1473, 1573, 1673
2.3. Method for image analysis of TEM images 2.3.1. Evaluation of aggregate shape The produced carbon black powders were heated at 110 °C to remove water. The dried powders were dissolved in chloroform, and the solution was dispersed by an ultrasonic bath. The dispersed solution was dropped onto a collodion substrate and examined using a TEM (TecnaiG2 20 ST, FEI) operated at 120 kV. We evaluated the shape characteristics of more than 1000 aggregates in the TEM images using an image analyzer (LUZEX AP, NIRECO) [6,7]. The maximum length (L), width (W), projected area (A), perimeter (P), and covered area (Ac) were obtained from the results of image analysis, as shown in Fig. 3. Three shape parameters— anisotropy (X), complexity (Y), and covered ratio (Z)—are estimated according to Eqs. (2)–(4), respectively. They are defined as follows:
X¼
L ; W
ð2Þ
Y¼
1 P2 ; 4p A
ð3Þ
Z¼
Ac : A
ð4Þ
Using these three shape parameters, the aggregate shapes are classified into four categories—spheroidal, ellipsoidal, linear, and branched—according to the classification category criterion as shown in Figs. 4 and 5 [7,18,19]. Branched and linear shapes are more complicated than those of spheroidal and ellipsoidal.
Temperature [K]
1600
Covered area (Ac)
1400 1200
Perimeter (P)
1673 K 1000
Projected area (A)
1573 K 1473 K
800 600
Width (W)
0
200
400
600
800
1000
Location [mm] Fig. 2. Variations in temperatures of reaction gas in a reactor.
Maximum length (L) Fig. 3. Shape characteristics for a projection image of an aggregate.
131
K. Ono et al. / Chemical Engineering Journal 215–216 (2013) 128–135
2.3.2. Calculation of mean primary particle diameter The mean primary particle diameter is calculated according to ASTM D3849-07 [20],
P dp ¼
dpi ni ; nt
apA P
Spheroidal
ð6Þ
;
P2 a ¼ 13:092 A
!0:92
nt
; c P1000 i¼1 V pi ni
ð7Þ
ð8Þ
where ni [] is the number of particles of an aggregate, nt [] is the total number of particles, q [g/s] is the carbon black yield, which is an increase in the weight of the filter and of the carbon black in the reaction tube divided by total time, Vp [m3/particle] is the mean volume of the aggregate, and c (=1.8 106 g/m3) is the density of carbon black. In this study, because the composition of feedstock varies with constant flow rate, the feedstock amount of carbon per unit time varies for each condition. In order to investigate the effects of the chemical reaction on the feedstock composition, the feedstock amount of carbon per unit time needs to be considered. Thus, we define Q which is considered to be the feedstock amount of carbon per unit time (Qs, particles/s), as shown in Eq. (9).
Qs ¼ Q0
Complicated Fig. 5. Criteria for aggregate shape classification [18,19].
:
2.3.3. Calculation of the number of carbon black particles produced per unit time To evaluate the number of nuclei and coagulation, the number of carbon black particles produced per unit time (Q, particles/s) is calculated based on the formula put forward by Tesner and Shurupov [21], as shown in the following equation:
q
Branched
Linear
where A [m2] is the projected area, P [m] is the perimeter, and a is the aggregate factor, given by
Q¼
Ellipsoidal
ð5Þ
where ni [] is the number of particle of an aggregate, nt [] is the total number of particle, and dpi is the mean particle diameter described as
dpi ¼
Simple
6xbenzene þ 2xacetylene ; 6xbenzene;0 þ 2xacetylene;0
ð9Þ
where Q0 [1013 particles/s] is the number of carbon black particles produced per unit time when the feedstock comprises only benzene or acetylene, x [vol%] is the concentration of hydrocarbons and x0 is the concentration of hydrocarbons when Q0 is calculated.
Fig. 4. Aggregate shape classification categories [7].
In the case where Q > Qs, the formation of large PAHs occurs and nucleation may take place upon the addition of benzene or acetylene. However, this is not the case when the feedstock comprises only benzene or acetylene. On the other hand, when Q < Qs, the number of primary particles is small when compared with the feedstock amount of carbon per unit time for the case where the feedstock is only benzene or acetylene. 3. Results and discussion 3.1. TEM images of carbon black TEM images of carbon black produced for different compositions of feedstock and temperature are shown in Fig. 6a–f. In the case where there is a change in benzene concentration, as shown in Fig. 6a and b, the primary particle diameter increases, and aggregate shapes become complex with an increase in benzene concentration. In the case where there is a change in the acetylene concentration, as shown in Fig. 6c–f, the primary particle diameter increases and aggregate shapes become complex with an increase in the acetylene concentration. When we vary the furnace temperature, as shown in Fig. 6c–f, the primary particle diameter decreases (it increases with an increase in the furnace temperature). In the following section, we quantitatively discuss the effects of benzene and acetylene concentration and the furnace temperature on the configurations of carbon black. 3.2. Effect of benzene concentration Table 2 shows results of Q, Qs and the mean primary particle diameter d of carbon black, which is synthesized for each benzene concentration, where the acetylene concentration is 3.0 vol% and the furnace temperature is 1573 K. Qs is based on the case in which the feedstock is only acetylene. In the case where the benzene concentration is 0.10 vol%, Q is larger than Qs (Q > Qs) and the mean primary particle diameter decreases when compared with the case in which the feedstock comprises only acetylene. In the case of relatively high temperature, because the rate coefficient of the reaction between benzene and acetylene is larger than that of the reaction between benzene and benzene [12], the reaction between benzene and acetylene occurs with the addition of benzene. Thus, in the case where the feedstock is a mixture of benzene and acetylene, an increase in Q implies that the formation of large PAHs [6] has occurred and the number of nuclei increases when compared with the case in which the feedstock comprises only acetylene. There are more large PAHs that contribute to
132
K. Ono et al. / Chemical Engineering Journal 215–216 (2013) 128–135
Fig. 6. TEM images of carbon black, (a) 1573 K, Benzene 0.10 vol%, acetylene 3.0 vol%, 0.158 s, (b) 1573 K, benzene 5.0 vol%, acetylene 3.0 vol%, (c) 1473 K, benzene 1.0 vol%, acetylene 0.50 vol%, (d) 1473 K, benzene 1.0 vol%, acetylene 5.0 vol%, (e) 1673 K, benzene 1.0 vol%, acetylene 0.50 vol% and (f) 1673 K, benzene 1.0 vol%, acetylene 5.0 vol%.
Table 2 Number of carbon black particles produced per unit time for carbon black particles and mean primary particle diameter with benzene concentration at 1573 K. Benzene concentration (vol%)
0
0.10
3.0
5.0
Q (1013 particles/s) Qs (1013 particles/s) d (nm)
0.997 0.997 40
1.82 1.10 35
2.38 3.99 49
2.77 5.98 54
nucleation than small PAHs [6] that contribute to the surface growth. Therefore, the mean primary particle diameter decreases. When the benzene concentration is over 3.0 vol%, Q is smaller than Qs (Q < Qs) and the mean primary particle diameter increases, whereas Q increases when compared with the case in which the benzene concentration is low. Because the reaction between benzene and acetylene occurs with an increase in benzene concentra-
tion, Q increases. However, since the acetylene concentration is constant, the benzene concentration is relatively larger than the acetylene concentration. Hence, the reaction between benzene and benzene occurs in addition to the reaction between benzene and acetylene. The ratio of small PAHs increases relative to the case involving low benzene concentration because the rate coefficient of the reaction between benzene and benzene is low [18]. Thus, the results when Q < Qs and the increase in the mean primary particle diameter indicate that surface growth occurs because of small PAHs. The results of the aggregate shape classification for varying benzene concentration are shown in Fig. 7. When the benzene concentration is 0–3.0 vol%, aggregate shapes become complex as the benzene concentration increases. This result implies that the collision rate has increased. This is because the Q increases significantly with an increase in benzene concentration.
133
K. Ono et al. / Chemical Engineering Journal 215–216 (2013) 128–135
When the benzene concentration is 5.0 vol%, aggregate shapes become simple compared with the case in which the benzene concentration is 3.0 vol%. This leads us to consider the mean thermal velocity of a particle and an aggregate (cp), as shown in Eq. (10) [22]:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 48kT cp ¼ ; 2 p Nqp d3
ð10Þ
where cp [m/s] is the mean thermal velocity of a particle or an aggregate, d [m] is the primary particle diameter, k [J/K] is the Boltzmann constant, N [] is the number of primary particles of an aggregate, T is the temperature [K], and qp [kg/m3] is the particle density. cp is inversely proportional to the two-thirds power of the particle diameter. Thus, a simplification of the aggregate shapes implies that the collision rate decreases, since the effect of a decrease in cp is large because of a significant increase in the mean primary particle. 3.3. Effect of acetylene concentration at low furnace temperature Table 3 shows the results of the Q, Qs and the mean primary particle diameter d of carbon black synthesized for each acetylene concentration, where the benzene concentration is 1.0 vol% and the furnace temperature is 1473 K (which is a relatively low temperature). Qs is based on the case in which the feedstock comprises only benzene. When the acetylene concentration is 0.50 vol% (which is a relatively low acetylene concentration), Q is the same as Qs (Q ’ Qs), and the mean primary particle diameter increases when compared with the case in which the feedstock comprises only acetylene. For the case of relatively low furnace temperatures, the rate coefficient of the reaction between acetylene and acetylene is larger than that of the reaction between benzene and acetylene [18]. Because the reaction between acetylene and acetylene requires time to produce large PAHs, which contribute to nucleation, nucleation does not occur in the case of low acetylene concentration; hence, Q is nearly equal to Qs. Thus, increments in the mean primary particle diameter are due to the surface growth of increments in the feedstock amount of carbon per unit time. When the acetylene concentration is 5.0 vol% (a relatively high acetylene concentration), Q increases. However, Q is smaller than Qs (Q < Qs), and the mean primary particle diameter increases when compared with the case in which the acetylene concentration is low. The reaction rate constant for the reaction between acetylene and acetylene is larger than that of the reaction between benzene and acetylene [11]. Because the reaction between benzene and acetylene occurs with an increase in acetylene concentration, Q
80 51.9
47.8
5.39
35.5
37.3
Spheroidal Ellipsoidal
32.0
25.9
23.2
0.10
5.0
0.916 0.916 42
1.03 1.07 45
1.43 1.07 52
increases compared with the case in which the acetylene concentration is 0.50 vol%. However, since the ratio of acetylene to benzene is large, the reaction between both acetylene molecules occurs significantly in addition to the reaction between benzene and acetylene. The ratio of small PAHs increases because the rate coefficient for reaction between acetylene and acetylene becomes low with an increase in furnace temperature [18]. Thus, the results when Q < Qs and an increase in the mean primary particle diameter indicate that surface growth occurs because of small PAHs. The result of aggregate shape classification in the case of varying acetylene concentration at 1473 K is shown in Fig. 8. In the case where the benzene concentration is 0.50 vol%, the complicated nature of aggregate shapes remains mostly unchanged when compared with the case where the feedstock comprises only benzene. This result implies that the collision rate remains mostly unchanged when compared with the case where the feedstock comprises only benzene. This is because the rate increases with increasing flux of carbon black particles, and the rate decreases with increasing mean primary particle diameter. In the case where the acetylene concentration is 5.0 vol%, the aggregate shapes are complicated when compared with the case for which the acetylene concentration is 0.50 vol%. This result implies that the collision rate increases because the effect of the increase in Q on the increase in the collision rate is larger than that of an increase in the mean primary particle diameter for a decrease in the rate. 3.4. Effect of acetylene concentration at high furnace temperature Table 4 shows the results of Q, Qs and the mean primary particle diameter d of carbon black, which is synthesized for each acetylene concentration, where the benzene concentration is 1.0 vol% and the furnace temperature is relatively high (1673 K). Qs is based on the case in which the feedstock comprises only benzene. When the acetylene concentration is relatively low (0.50 vol%), Q increases and becomes larger than Qs (Q > Qs), and the mean primary particle diameter remains mostly unchanged compared with the case in which the feedstock comprises only benzene. In the case of relatively high furnace temperature, the rate coefficient of the reaction between benzene and acetylene is larger than that
Branched
20
80
25.0
27.2
28.4
0
0.10
3.0
5.0
0
Benzene concentration [vol.%] Fig. 7. Ratio of aggregates with benzene concentration at 1573 K.
38.4
5.71
36.8
3.58
33.5
40
Spheroidal Ellipsoidal
60
20 20.5
1.98
100
Linear 22.1
0
Q (1013 particles/s) Qs (1013 particles/s) d (nm)
8.31
60 40
Acetylene concentration (vol%)
Existence ratio [%]
Existence ratio [%]
3.93
5.49
100
Table 3 Number of carbon black particles produced per unit time for carbon black particles and mean primary particle diameter with acetylene concentration at 1473 K.
28.8
28.8
30.0
Linear Branched
30.9
28.7
32.9
0
0.50
5.0
0
Acetylene concentration [vol.%] Fig. 8. Ratio of aggregates with acetylene concentration at 1473 K.
K. Ono et al. / Chemical Engineering Journal 215–216 (2013) 128–135
Table 4 Number of carbon black particles produced per unit time for carbon black particles and mean primary particle diameter with acetylene concentration at 1673 K. Acetylene concentration (vol%)
0
0.10
5.0
Q (1013 particles/s) Qs (1013 particles/s) d (nm)
2.08 2.08 32
3.32 2.43 30
3.62 5.55 38
6.99
100
2.87
(a) Existence ratio [%]
134
2.58
5.71
100 80
36.8
3.63
2.87
41.3
42.6
Spheroidal Ellipsoidal
60
Linear 40
28.8
25.6
28.4
20 28.7
25.3
30.2
1573
1673
Branched
0
60
41.3
20
Ellipsoidal 25.6
Temperature [K]
Spheroidal
62.3
40
1473
35.8
28.1
Linear
(b)
4.15
3.58
100
2.58
Branched 19.9 30.2
33.5
0.50
5.0
Existence ratio [%]
Existence ratio [%]
80
10.8
0
0
Acetylene concentration [vol.%]
80
37.1
35.8
Spheroidal Ellipsoidal
60 30.0
29.9
40 20
Fig. 9. Ratio of aggregates with acetylene concentration at 1673 K.
33.5
32.9
28.1
Linear Branched
28.9
33.5
1573
1673
0
of the reaction between acetylene and acetylene [11]. Thus, since there is a significant reaction between benzene and acetylene, nucleation occur The number of small PAHs, which contribute to surface growth, remains mostly unchanged because an increase in the feedstock amount of carbon per unit time with an increase in the acetylene concentration contributes to nucleation, and therefore, the mean primary particle diameter also remains unchanged. In the case where the acetylene concentration is relatively high (5.0 vol%), Q is smaller than Qs (Q < Qs), and the mean primary particle diameter increases compared with the case in which the acetylene concentration is low. Because the ratio of acetylene to benzene is large, as is the case for low furnace temperatures, a significant reaction occurs between acetylene and acetylene in addition to the reaction between benzene and acetylene. For small PAHs, the ratio increases because the rate coefficient for the reaction between acetylene and acetylene is low with an increase in the furnace temperature [11]. The result of aggregate shape classification in the case where the acetylene concentration changes at 1673 K is shown in Fig. 9. The aggregate shapes become complex as the acetylene concentration increases. This result reveals that the collision rate increases because of an increase in Q. However, the variation is small in the case where the acetylene concentration is 5.0 vol%. This result is due to an increase in the mean primary particle diameter. When the acetylene concentration is 5.0 vol%, the aggregate shapes become complex when compared with the case for which the acetylene concentration is 0.50 vol%. This result implies that the collision rate increases because the effect of an increase in Q
1473
Temperature [K] Fig. 10. Ratio of aggregates with temperature, (a) benzene 1.0 vol%, acetylene 0.50 vol% and (b) benzene 1.0 vol%, acetylene 5.0 vol%.
on the increase in the collision rate is larger than that of the increase in the mean primary particle diameter for a decrease in the rate. 3.5. Effect of furnace temperature Table 5 shows the results for Q and the mean primary particle diameter d of carbon black synthesized for each furnace temperature, where the benzene concentration is 1.0 vol% and the acetylene concentration is 0.50 and 5.0 vol%. For each feedstock, Q increases, and the variation is especially large at 1673 K. This result implies that the rate coefficient of the reaction between benzene and acetylene increases with increasing furnace temperature [18], enabling the formation of large PAHs. The mean primary particle diameter decreases as a result of a decrease in small PAHs that contribute to surface growth and an increase in large PAHs. The results of the aggregate shape classification in the case involving a mixture of 1.0 vol% benzene and 0.50 vol% acetylene, and in the case of 1.0 vol% benzene and 5.0 vol% acetylene are shown in Fig. 10a and b. In each case, aggregate shapes are slightly simplified from 1473 K to 1573 K, whereas they are complicated from 1573 K to 1673 K. In the case of relatively long residence
Table 5 Number of carbon black particles produced per unit time for carbon black particles and mean primary particle diameter with furnace temperature. Acetylene concentration (vol%)
Benzene 1.0 vol%–acetylene 0.5 vol%
Temperature (K) Q (1013 particles/s) d (nm)
1473 1.03 45
1573 1.45 42
Benzene 1.0 vol%–acetylene 5.0 vol% 1673 3.32 30
1473 1.43 52
1573 1.86 47
1673 3.62 38
K. Ono et al. / Chemical Engineering Journal 215–216 (2013) 128–135
times (about 300 ms whereas the condition in the industrial furnace allows under 100 ms), the sintering of primary particles occurs in the case of high temperatures [9]. Thus, aggregate shapes are simplified at 1573 K because of sintering. On the other hand, at 1673 K, since Q is approximately twice of 1573 K and the mean primary particle diameter increases significantly, the collision rate also increases significantly. As a result, aggregate shapes are complicated. 3.6. Effect of hydrocarbon addition on aggregation mechanisms of carbon black When additives are added to feedstock, the formation of large PAHs is promoted because the reaction between benzene and acetylene occurs. An increase in large PAHs causes the promotion of nuclei and number of carbon black particles. Thus, aggregate shapes become complex due to an increase in collision rate. When the concentration of additives increases, the mixture ratio that promotes the chemical reaction may be excessed. The excess hydrocarbon inhibits the formation of large PAHs, and small PAHs that contribute to the surface growth increases. An increase in primary particle diameter by surface growth causes a decrease in collision rate. Thus, aggregate shapes tend to simple at this case. 4. Conclusion In this study, we investigated the influence of the composition of benzene–acetylene on carbon black configurations. We formed carbon black by pyrolysis of a mixture of benzene and acetylene in an inert atmosphere. The effects of the benzene concentration, acetylene concentration, and furnace temperature on the mean primary particle diameter and the aggregate shape of carbon black are investigated. The results are summarized as follows. When the acetylene concentration is changed and the benzene concentration is made constant, aggregate shapes become complex as the acetylene concentration increases. However, in the case where the acetylene concentration is larger than the benzene concentration, the variations in the aggregate shapes are small with an increase in acetylene concentration, because surface growth occurs and the collision rate decreases owing to an increase in the mean primary particle diameter. In the case where there is a change in the furnace temperature, the number of carbon black particles produced per unit time increases. Also, the mean primary particle diameter decreases with an increase in furnace temperature because the rate coefficient for the reaction between benzene and acetylene increases. In addition, aggregate shapes become complex with an increase in furnace temperature. Therefore, from the observations we conclude that nucleation occurs and aggregate shapes become complicated when the ratio of the benzene and acetylene concentration is suitable (about 2:1 in this study), and the furnace temperature is high. However, in
135
the case where benzene or acetylene concentration is large and the furnace temperature is low, the aggregate shapes are simplified because the formation of small PAHs, which contributes to surface growth. References [1] T. Ungár, J. Gubicza, G. Tichy, C. Pantea, T.W. Zerda, Size and shape of crystallite and internal stresses in carbon blacks, Compos. A Appl. Sci. Manuf. 36 (2005) 431–436. [2] J. Lahaye, G. Prado, J.B. Donnet, Nucleation and growth of carbon black particles during thermal decomposition of benzene, Carbon 12 (1974) 27–35. [3] J. Lahaye, F. Ehrburger-Dolle, Mechanisms of carbon black formation. Correlation with the morphology of aggregates, Carbon 32 (1994) 1319–1324. [4] 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. 26 (2000) 565–608. [5] J.B. Donnet, R.C. Bansal, M.J. Wang, Carbon Black, Marcel Dekker, New York, 1993. [6] K. Ono, M. Yanaka, S. Tanaka, Y. Saito, H. Aoki, O. Fukuda, T. Aoki, T. Yamaguchi, Influence of furnace temperature and residence time on configurations of carbon black, Chem. Eng. J. 200–202 (2012) 541–548. [7] F. Shishido, H. Hashiguchi, Y. Matsushita, Y. Morozumi, H. Aoki, T. Miura, An investigation of primary particle growth and aggregate formation of soot using a numerical model considering the sintering of primary particles, Kagaku Kogaku Ronubnshu 33 (2007) 306–314. [8] U. Bonne, K.H. Homann, H.Gg. Wagner, Carbon formation in premixed flames. Symposium (International) on Combustion (Pittsburg): The Combustion Institute, 1965; pp. 503–512. [9] K.H. Homann, H.Gg. Wagner, Some New Aspects of the Mechanism of Carbon Formation in Premixed Flames. Symposium (International) on Combustion (Pittsburg): The Combustion Institute, 1967, pp. 371–379. [10] J.R. Arthur, D.H. Napier, Formation of Carbon and Related Materials in Diffusion Flames, Symposium (International) on Combustion (Pittsburg): The Combustion Institute, 1955, pp. 303–316. [11] M. Frenklach, D.W. Clary, W.C. Gardiner, S.E. Stein, Symposium (International) on Combustion (Pittsburg): The Combustion Institute, 1984, pp. 887–901. [12] H. Wang, M. Frenklach, A detailed kinetic modeling study of aromatics formation in laminar premixed acetylene and ethylene flames, Combust. Flame 110 (1997) 173–221. [13] P.A. Tesner, S.V. Shurupov, Soot formation during pyrolysis and combustion of hydrocarbons, Combust. Sci. Technol. 109 (1995) 399–400. [14] P.A. Tesner, S.V. Shurupov, Soot formation during pyrolysis of benzene–xylene mixtures, Combust. Sci. Technol. 109 (1995) 401–404. [15] P.A. Tesner, S.V. Shurupov, Soot formation during pyrolysis of naphthalene, anthracene and pyrene, Combust. Sci. Technol. 126 (1997) 139–151. [16] P.A. Tesner, S.V. Shurupov, Some physico–chemical parameters of soot formation during pyrolysis of methane and methane–acetylene and methane–benzene mixtures. Symposium (International) on Combustion: The Combustion Institute (Pittsburg), 1994; pp. 653–659. [17] R. Watanabe, T. Shindoh, Y. Matsushita, H. Aoki, T. Miura, K. Nishiwaki, H. Yamada, O. Fukuda, T. Yamagychi, A numerical investigation of the factors influencing the aggregate shape of carbon black from the furnace process, J. Chem. Eng. Jpn. 43 (2010) 150–157. [18] C.R. Herd, G.C. Mcdonald, W.M. Hess, Morphology of carbon-black aggregates, Rubber Chem. Tech. 65 (1992) 107–129. [19] C.R. Herd, G.C. Mcdonald, R.E. Smith, W.M. Hess, The use of skeletonization for the shape classification of carbon-black aggregates, Rubber Chem. Tech. 66 (1993) 491–509. [20] ASTM D 3849-07. Standard test method for carbon black – morphological characterization of carbon black using electron microscopy. [21] P.A. Tesner, S.V. Shurupov, Soot formation from acetylene–benzene mixture, Combust. Sci. Technol. 92 (1993) 61–67. [22] A. Einstein, Über die von der molekularkinetischen Theorie der Wärme geforderte Bewegung von in ruhenden Flüssigkeiten suspendierten Teilchen, Annalen der Physik 322 (1905) 549–560.
Contents lists available at SciVerse ScienceDirect
Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej
Effect of benzene–acetylene compositions on carbon black configurations produced by benzene pyrolysis Kiminori Ono a,⇑, Miki Yanaka a, Yasuhiro Saito a, Hideyuki Aoki a, Okiteru Fukuda b, Takayuki Aoki b, Togo Yamaguchi b a b
Department of Chemical Engineering, Graduate School of Engineering, Tohoku University, 6-6-07 Aoba, Aramaki, Aoba-ku, Sendai, Miyagi 980-8579, Japan ASAHI CARBON CO., LTD., 2 Kamomejima-cho, Higashi-ku, Niigata 950-0883, Japan
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
" Aggregate shapes become complex
with an increase in acetylene concentration. " The number flow rate increases with an increase in furnace temperature. " The mean primary diameter decreases with an increase in furnace temperature. " The shapes most complicate when benzene–acetylene concentration ratio is about 2:1.
a r t i c l e
i n f o
Article history: Received 28 June 2012 Received in revised form 29 September 2012 Accepted 2 October 2012 Available online 10 November 2012 Keywords: Carbon black Soot Pyrolysis Aggregate
The addition of acetylene causes complication of aggregate shapes
500 nm
500 nm
a b s t r a c t A mixture of benzene and acetylene is pyrolyzed in an inert atmosphere to investigate the influence of the benzene–acetylene composition on the configurations of carbon black. The effects of benzene concentration, acetylene concentration, and furnace temperature on the mean primary particle diameter and the aggregate shape in carbon black are investigated. When the acetylene concentration is varied and the benzene concentration is made constant, aggregate shapes become complex with an increase in acetylene concentration. However, in the case where the acetylene concentration is greater than that of the benzene concentration, the variation of aggregate shapes is small with increasing acetylene concentration. The results of this study suggest that nucleation has progressed and aggregate shapes appear complicated when the ratio of the benzene concentration to the acetylene concentration is appropriate (in this study, the ratio is 2:1) and the furnace temperature is high. However, when the benzene or acetylene concentration is high, and the furnace temperature is low, aggregate shapes are simplified because of the formation of small polycyclic aromatic hydrocarbons (PAHs), which contributes to surface growth. Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction Carbon black is a type of soot that is produced industrially and is used for composite materials such as tires and electrode materials of batteries. Although the smallest individual units of carbon black are aggregates, transmission electron microscopy (TEM) images show that these aggregates appear to be formed by spherical particles that are fused together [1]. The aggregate shape is one ⇑ Corresponding author. Tel.: +81 22 795 7251; fax: +81 22 795 6165. E-mail address: [email protected] (K. Ono). 1385-8947/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.cej.2012.10.085
of the factors that affect the properties of composite materials. In the furnace process, carbon black is produced by the continuous pyrolysis of hydrocarbons, which are sprayed into a high-temperature field (1500–2000 K) inside the furnace. The process is complicated owing to the fact that chemical reactions occur rapidly with heat and mass transfer, and therefore, it is difficult to control the aggregate shape. Hence, the aggregation mechanism of carbon black has not yet been completely understood. At present, because the technique used to control the aggregation of carbon black particles is a trial-and-error process, a theoretical solution is required to precisely control the aggregate shape.
129
K. Ono et al. / Chemical Engineering Journal 215–216 (2013) 128–135
Nomenclature A Ac cp d k L M N P Q
projected area (m2) covered area (m2) mean thermal velocity of particle (m/s) mean primary particle diameter (nm) Boltzmann constant (J/K) maximum length (m) mean aggregate size (m2) number of primary particle in an aggregate () perimeter (m) flux of carbon black particles of carbon black particles (1013 particles/s)
Numerous studies have been conducted on the effects of pyrolysis on the diffusion flame and reaction tube in an attempt to explain the formation mechanism of carbon black and soot [2–7]. The formation mechanism for carbon black is considered to be as follows. Large molecules are considered to be the precursors of carbon black particles. Although various theories have been proposed for the reaction mechanism of pyrolysis, all of the recent studies conclude that in both pyrolytic systems and flame systems, polycyclic aromatic hydrocarbons (PAHs) are the precursors of carbon nuclei [2,3]. The nascent soot particles expand either by the addition of molecules from a gas phase, such as acetylene, or by reaction with smaller PAHs [4]. At the start of the reaction, a large number of particles are produced. These particles collide to produce larger spherical particles, which then aggregate into final carbon black clusters. The particles are converted into amorphous carbon and a progressively more graphitic material in the furnace. This is because graphite is the most thermodynamically stable form of carbon [5]. With a long residence time, it is believed that aggregate growth occurs as a result of fusion among primary particles [6]. Shishido et al. have proposed a sintering model [7], which suggests that the rate of sintering among primary particles depends on the size, temperature, surface energy, and viscosity of the particles. However, these mechanisms are still not entirely understood. Over the last few decades, PAH formation, which is considered to be a precursor of carbon black particles in flames, has been extensively studied by several groups [8–16]. Bonne et al. [8] and Homann and Wagner [9] studied the formation of carbon black in premixed flames using various fuels by mass spectrometric analysis of the flames. This provides the concentration profiles of the intermediate products that play an important role during the formation of carbon particles. Arthur and Napier [10] studied the formation of carbon black in reversed and normal diffusion flames by isolating the various intermediate stages in the flames by properly choosing the combustion and quenching conditions. The formation and growth of PAHs are described by the addition of small building blocks, such as acetylene, to a precursor PAH [11]. Wang and Frenklach [12] performed a computational study for the formation and growth of PAHs in laminar premixed acetylene and ethylene flames. They demonstrated that the reactions of n-C4Hx + C2H2 leading to the formation of one-ring aromatics are as important as the propargyl recombination, and hence, the reactions must be included in the kinetic modeling of PAH formation in hydrocarbon flames. Tesner et al. [13–16] analyzed surface areas and the number density of soot synthesized by a mixture of hydrocarbons using TEM. They suggested that the number density of soot which was produced by methane–acetylene–helium mixture increased with an increase in the acetylene concentration [16]. Watanabe et al. [17] suggested that aggregate shapes became complicated as the particle number density increased. Thus, it is expected that the
Qs
T W X Y Z
a qp
flux of carbon black particles of carbon black particles considered flux of the feedstock’s carbon (1013 particles/s) furnace temperature (K) width (m) anisotropy () complexity () covered ratio () aggregate factor () density of carbon black (kg/m3)
composition of the hydrocarbon, especially the addition of acetylene, affects the aggregate shape of carbon black. Although many investigations have been conducted to measure the concentration of PAHs in the flame, the surface area, and the number density of soot synthesized by a mixture of hydrocarbons, little is known about the effect of changes in the aggregate shape on carbon black produced with a mixture of hydrocarbon. In this study, we form carbon black by pyrolysis of a mixture of benzene and acetylene in an inert atmosphere, and the shape parameters of the carbon black are measured by TEM analysis. The micrographs are analyzed by image analysis to precisely evaluate the aggregate shape. The number of carbon black particles produced per unit time is calculated using the yield of carbon black and the shape parameters. Therefore, these experiments enable us to further understand the influence of the composition of benzene– acetylene on configurations of carbon black, such as the aggregate shape and primary particle diameter. 2. Experimental 2.1. Experimental apparatus Carbon black is produced by the thermal pyrolysis of benzene entrained by a N2 carrier flow. The schematic diagram of the experimental set up used in this study is shown in Fig. 1. Saturated benzene gas is prepared by aerating liquid benzene (>99.5%) with primary N2 (>99.999%) in a bubbling apparatus, and 0–5.0 vol% of the benzene feedstock gas is produced by attenuating the satu-
radial
Heater
axial
Pre-Heater
Primary Secondary N2 N2 Acetylene
Filter
800
Water bath
1000
Vacuum
Benzene
Fig. 1. Schematic diagram showing the experimental apparatus for the production of carbon black.
130
K. Ono et al. / Chemical Engineering Journal 215–216 (2013) 128–135
rated gas with secondary N2. The gas entrained benzene is mixed with 0–5.0 vol% of acetylene. The benzene and acetylene concentration in the feedstock gas is measured by gas chromatography (hydrogen flame ionization detector, GC-17A, SHIMADZU), and is confirmed to be the predefined concentration. The feedstock gas is supplied to an alumina tube heated by a pre-heater (length = 800 mm) at 873 K and an electric furnace (length = 1000 mm). The inner diameter of the tube is 16 mm, and the length is 1970 mm. The glass fiber filter (inner diameter = 68 mm) is produced by ADVANTEC, and the trapping efficiency for particles larger than 0.3 lm is better than 99.9%.
Table 1a Experimental condition for different values of (a) benzene concentration. Benzene concentration (vol%) Acetylene concentration (vol%) Furnace temperature (K)
0, 0.10, 3.0, 5.0 3.0 1573
Table 1b Experimental condition for different values of acetylene concentration. Benzene concentration (vol%) Acetylene concentration (vol%) Furnace temperature (K)
1.0 0, 0.50, 5.0 1473, 1673
2.2. Experimental conditions The axial temperature in the reaction tube was measured using an R-type thermocouple in the preliminary experiment. In this study, the radial temperature distribution is ignored, because the experiment is conducted after attaining a steady state. The axial temperature distribution in the reaction tube is shown in Fig. 2. The maximum temperature occurs near the midpoint of the tube, and temperatures near the inlet and outlet of the tube is low. From this trend, the flow velocity may change depending on the axial location owing to the expansion and contraction that result from temperature changes in the flowing process. In this experiment, we assume that the pressure of the feedstock gas is atmospheric pressure, and we calculate the residence time t [s] using the following equation:
t¼
Z
Ai T 1 ðlÞ dl; V0 T0
ð1Þ
where Ai [m2] is the cross-sectional area of the reaction tube, l [m] is the axial location, T0 [K] is room temperature (293 K), T1 [K] is the measured temperature, and V0 [L/min] is the feedstock flow rate at T0. The average temperature zone (length = 350 mm) is defined as the axial temperature area of the preset temperature ±10 K, and the residence time [s] is defined as the time during which the feedstock gas passes through the zone. The feedstock flow rate is 3.0 L/min. The furnace temperature is defined as the average temperature zone and is set to 1473 K, 1573 K, and 1673 K. The residence time calculated by Eq. (1) is 0.318, 0.308, and 0.296 s when the furnace temperature is 1473 K, 1573 K, and 1673 K, respectively. We vary (a) benzene concentration, (b) acetylene concentration, and (c) furnace temperature, as shown in Tables 1a–c, respectively.
1800
Table 1c Experimental condition for different values of furnace temperature. Benzene concentration (vol%) Acetylene concentration (vol%) Furnace temperature (K)
1.0 0.50, 5.0 1473, 1573, 1673
2.3. Method for image analysis of TEM images 2.3.1. Evaluation of aggregate shape The produced carbon black powders were heated at 110 °C to remove water. The dried powders were dissolved in chloroform, and the solution was dispersed by an ultrasonic bath. The dispersed solution was dropped onto a collodion substrate and examined using a TEM (TecnaiG2 20 ST, FEI) operated at 120 kV. We evaluated the shape characteristics of more than 1000 aggregates in the TEM images using an image analyzer (LUZEX AP, NIRECO) [6,7]. The maximum length (L), width (W), projected area (A), perimeter (P), and covered area (Ac) were obtained from the results of image analysis, as shown in Fig. 3. Three shape parameters— anisotropy (X), complexity (Y), and covered ratio (Z)—are estimated according to Eqs. (2)–(4), respectively. They are defined as follows:
X¼
L ; W
ð2Þ
Y¼
1 P2 ; 4p A
ð3Þ
Z¼
Ac : A
ð4Þ
Using these three shape parameters, the aggregate shapes are classified into four categories—spheroidal, ellipsoidal, linear, and branched—according to the classification category criterion as shown in Figs. 4 and 5 [7,18,19]. Branched and linear shapes are more complicated than those of spheroidal and ellipsoidal.
Temperature [K]
1600
Covered area (Ac)
1400 1200
Perimeter (P)
1673 K 1000
Projected area (A)
1573 K 1473 K
800 600
Width (W)
0
200
400
600
800
1000
Location [mm] Fig. 2. Variations in temperatures of reaction gas in a reactor.
Maximum length (L) Fig. 3. Shape characteristics for a projection image of an aggregate.
131
K. Ono et al. / Chemical Engineering Journal 215–216 (2013) 128–135
2.3.2. Calculation of mean primary particle diameter The mean primary particle diameter is calculated according to ASTM D3849-07 [20],
P dp ¼
dpi ni ; nt
apA P
Spheroidal
ð6Þ
;
P2 a ¼ 13:092 A
!0:92
nt
; c P1000 i¼1 V pi ni
ð7Þ
ð8Þ
where ni [] is the number of particles of an aggregate, nt [] is the total number of particles, q [g/s] is the carbon black yield, which is an increase in the weight of the filter and of the carbon black in the reaction tube divided by total time, Vp [m3/particle] is the mean volume of the aggregate, and c (=1.8 106 g/m3) is the density of carbon black. In this study, because the composition of feedstock varies with constant flow rate, the feedstock amount of carbon per unit time varies for each condition. In order to investigate the effects of the chemical reaction on the feedstock composition, the feedstock amount of carbon per unit time needs to be considered. Thus, we define Q which is considered to be the feedstock amount of carbon per unit time (Qs, particles/s), as shown in Eq. (9).
Qs ¼ Q0
Complicated Fig. 5. Criteria for aggregate shape classification [18,19].
:
2.3.3. Calculation of the number of carbon black particles produced per unit time To evaluate the number of nuclei and coagulation, the number of carbon black particles produced per unit time (Q, particles/s) is calculated based on the formula put forward by Tesner and Shurupov [21], as shown in the following equation:
q
Branched
Linear
where A [m2] is the projected area, P [m] is the perimeter, and a is the aggregate factor, given by
Q¼
Ellipsoidal
ð5Þ
where ni [] is the number of particle of an aggregate, nt [] is the total number of particle, and dpi is the mean particle diameter described as
dpi ¼
Simple
6xbenzene þ 2xacetylene ; 6xbenzene;0 þ 2xacetylene;0
ð9Þ
where Q0 [1013 particles/s] is the number of carbon black particles produced per unit time when the feedstock comprises only benzene or acetylene, x [vol%] is the concentration of hydrocarbons and x0 is the concentration of hydrocarbons when Q0 is calculated.
Fig. 4. Aggregate shape classification categories [7].
In the case where Q > Qs, the formation of large PAHs occurs and nucleation may take place upon the addition of benzene or acetylene. However, this is not the case when the feedstock comprises only benzene or acetylene. On the other hand, when Q < Qs, the number of primary particles is small when compared with the feedstock amount of carbon per unit time for the case where the feedstock is only benzene or acetylene. 3. Results and discussion 3.1. TEM images of carbon black TEM images of carbon black produced for different compositions of feedstock and temperature are shown in Fig. 6a–f. In the case where there is a change in benzene concentration, as shown in Fig. 6a and b, the primary particle diameter increases, and aggregate shapes become complex with an increase in benzene concentration. In the case where there is a change in the acetylene concentration, as shown in Fig. 6c–f, the primary particle diameter increases and aggregate shapes become complex with an increase in the acetylene concentration. When we vary the furnace temperature, as shown in Fig. 6c–f, the primary particle diameter decreases (it increases with an increase in the furnace temperature). In the following section, we quantitatively discuss the effects of benzene and acetylene concentration and the furnace temperature on the configurations of carbon black. 3.2. Effect of benzene concentration Table 2 shows results of Q, Qs and the mean primary particle diameter d of carbon black, which is synthesized for each benzene concentration, where the acetylene concentration is 3.0 vol% and the furnace temperature is 1573 K. Qs is based on the case in which the feedstock is only acetylene. In the case where the benzene concentration is 0.10 vol%, Q is larger than Qs (Q > Qs) and the mean primary particle diameter decreases when compared with the case in which the feedstock comprises only acetylene. In the case of relatively high temperature, because the rate coefficient of the reaction between benzene and acetylene is larger than that of the reaction between benzene and benzene [12], the reaction between benzene and acetylene occurs with the addition of benzene. Thus, in the case where the feedstock is a mixture of benzene and acetylene, an increase in Q implies that the formation of large PAHs [6] has occurred and the number of nuclei increases when compared with the case in which the feedstock comprises only acetylene. There are more large PAHs that contribute to
132
K. Ono et al. / Chemical Engineering Journal 215–216 (2013) 128–135
Fig. 6. TEM images of carbon black, (a) 1573 K, Benzene 0.10 vol%, acetylene 3.0 vol%, 0.158 s, (b) 1573 K, benzene 5.0 vol%, acetylene 3.0 vol%, (c) 1473 K, benzene 1.0 vol%, acetylene 0.50 vol%, (d) 1473 K, benzene 1.0 vol%, acetylene 5.0 vol%, (e) 1673 K, benzene 1.0 vol%, acetylene 0.50 vol% and (f) 1673 K, benzene 1.0 vol%, acetylene 5.0 vol%.
Table 2 Number of carbon black particles produced per unit time for carbon black particles and mean primary particle diameter with benzene concentration at 1573 K. Benzene concentration (vol%)
0
0.10
3.0
5.0
Q (1013 particles/s) Qs (1013 particles/s) d (nm)
0.997 0.997 40
1.82 1.10 35
2.38 3.99 49
2.77 5.98 54
nucleation than small PAHs [6] that contribute to the surface growth. Therefore, the mean primary particle diameter decreases. When the benzene concentration is over 3.0 vol%, Q is smaller than Qs (Q < Qs) and the mean primary particle diameter increases, whereas Q increases when compared with the case in which the benzene concentration is low. Because the reaction between benzene and acetylene occurs with an increase in benzene concentra-
tion, Q increases. However, since the acetylene concentration is constant, the benzene concentration is relatively larger than the acetylene concentration. Hence, the reaction between benzene and benzene occurs in addition to the reaction between benzene and acetylene. The ratio of small PAHs increases relative to the case involving low benzene concentration because the rate coefficient of the reaction between benzene and benzene is low [18]. Thus, the results when Q < Qs and the increase in the mean primary particle diameter indicate that surface growth occurs because of small PAHs. The results of the aggregate shape classification for varying benzene concentration are shown in Fig. 7. When the benzene concentration is 0–3.0 vol%, aggregate shapes become complex as the benzene concentration increases. This result implies that the collision rate has increased. This is because the Q increases significantly with an increase in benzene concentration.
133
K. Ono et al. / Chemical Engineering Journal 215–216 (2013) 128–135
When the benzene concentration is 5.0 vol%, aggregate shapes become simple compared with the case in which the benzene concentration is 3.0 vol%. This leads us to consider the mean thermal velocity of a particle and an aggregate (cp), as shown in Eq. (10) [22]:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 48kT cp ¼ ; 2 p Nqp d3
ð10Þ
where cp [m/s] is the mean thermal velocity of a particle or an aggregate, d [m] is the primary particle diameter, k [J/K] is the Boltzmann constant, N [] is the number of primary particles of an aggregate, T is the temperature [K], and qp [kg/m3] is the particle density. cp is inversely proportional to the two-thirds power of the particle diameter. Thus, a simplification of the aggregate shapes implies that the collision rate decreases, since the effect of a decrease in cp is large because of a significant increase in the mean primary particle. 3.3. Effect of acetylene concentration at low furnace temperature Table 3 shows the results of the Q, Qs and the mean primary particle diameter d of carbon black synthesized for each acetylene concentration, where the benzene concentration is 1.0 vol% and the furnace temperature is 1473 K (which is a relatively low temperature). Qs is based on the case in which the feedstock comprises only benzene. When the acetylene concentration is 0.50 vol% (which is a relatively low acetylene concentration), Q is the same as Qs (Q ’ Qs), and the mean primary particle diameter increases when compared with the case in which the feedstock comprises only acetylene. For the case of relatively low furnace temperatures, the rate coefficient of the reaction between acetylene and acetylene is larger than that of the reaction between benzene and acetylene [18]. Because the reaction between acetylene and acetylene requires time to produce large PAHs, which contribute to nucleation, nucleation does not occur in the case of low acetylene concentration; hence, Q is nearly equal to Qs. Thus, increments in the mean primary particle diameter are due to the surface growth of increments in the feedstock amount of carbon per unit time. When the acetylene concentration is 5.0 vol% (a relatively high acetylene concentration), Q increases. However, Q is smaller than Qs (Q < Qs), and the mean primary particle diameter increases when compared with the case in which the acetylene concentration is low. The reaction rate constant for the reaction between acetylene and acetylene is larger than that of the reaction between benzene and acetylene [11]. Because the reaction between benzene and acetylene occurs with an increase in acetylene concentration, Q
80 51.9
47.8
5.39
35.5
37.3
Spheroidal Ellipsoidal
32.0
25.9
23.2
0.10
5.0
0.916 0.916 42
1.03 1.07 45
1.43 1.07 52
increases compared with the case in which the acetylene concentration is 0.50 vol%. However, since the ratio of acetylene to benzene is large, the reaction between both acetylene molecules occurs significantly in addition to the reaction between benzene and acetylene. The ratio of small PAHs increases because the rate coefficient for reaction between acetylene and acetylene becomes low with an increase in furnace temperature [18]. Thus, the results when Q < Qs and an increase in the mean primary particle diameter indicate that surface growth occurs because of small PAHs. The result of aggregate shape classification in the case of varying acetylene concentration at 1473 K is shown in Fig. 8. In the case where the benzene concentration is 0.50 vol%, the complicated nature of aggregate shapes remains mostly unchanged when compared with the case where the feedstock comprises only benzene. This result implies that the collision rate remains mostly unchanged when compared with the case where the feedstock comprises only benzene. This is because the rate increases with increasing flux of carbon black particles, and the rate decreases with increasing mean primary particle diameter. In the case where the acetylene concentration is 5.0 vol%, the aggregate shapes are complicated when compared with the case for which the acetylene concentration is 0.50 vol%. This result implies that the collision rate increases because the effect of the increase in Q on the increase in the collision rate is larger than that of an increase in the mean primary particle diameter for a decrease in the rate. 3.4. Effect of acetylene concentration at high furnace temperature Table 4 shows the results of Q, Qs and the mean primary particle diameter d of carbon black, which is synthesized for each acetylene concentration, where the benzene concentration is 1.0 vol% and the furnace temperature is relatively high (1673 K). Qs is based on the case in which the feedstock comprises only benzene. When the acetylene concentration is relatively low (0.50 vol%), Q increases and becomes larger than Qs (Q > Qs), and the mean primary particle diameter remains mostly unchanged compared with the case in which the feedstock comprises only benzene. In the case of relatively high furnace temperature, the rate coefficient of the reaction between benzene and acetylene is larger than that
Branched
20
80
25.0
27.2
28.4
0
0.10
3.0
5.0
0
Benzene concentration [vol.%] Fig. 7. Ratio of aggregates with benzene concentration at 1573 K.
38.4
5.71
36.8
3.58
33.5
40
Spheroidal Ellipsoidal
60
20 20.5
1.98
100
Linear 22.1
0
Q (1013 particles/s) Qs (1013 particles/s) d (nm)
8.31
60 40
Acetylene concentration (vol%)
Existence ratio [%]
Existence ratio [%]
3.93
5.49
100
Table 3 Number of carbon black particles produced per unit time for carbon black particles and mean primary particle diameter with acetylene concentration at 1473 K.
28.8
28.8
30.0
Linear Branched
30.9
28.7
32.9
0
0.50
5.0
0
Acetylene concentration [vol.%] Fig. 8. Ratio of aggregates with acetylene concentration at 1473 K.
K. Ono et al. / Chemical Engineering Journal 215–216 (2013) 128–135
Table 4 Number of carbon black particles produced per unit time for carbon black particles and mean primary particle diameter with acetylene concentration at 1673 K. Acetylene concentration (vol%)
0
0.10
5.0
Q (1013 particles/s) Qs (1013 particles/s) d (nm)
2.08 2.08 32
3.32 2.43 30
3.62 5.55 38
6.99
100
2.87
(a) Existence ratio [%]
134
2.58
5.71
100 80
36.8
3.63
2.87
41.3
42.6
Spheroidal Ellipsoidal
60
Linear 40
28.8
25.6
28.4
20 28.7
25.3
30.2
1573
1673
Branched
0
60
41.3
20
Ellipsoidal 25.6
Temperature [K]
Spheroidal
62.3
40
1473
35.8
28.1
Linear
(b)
4.15
3.58
100
2.58
Branched 19.9 30.2
33.5
0.50
5.0
Existence ratio [%]
Existence ratio [%]
80
10.8
0
0
Acetylene concentration [vol.%]
80
37.1
35.8
Spheroidal Ellipsoidal
60 30.0
29.9
40 20
Fig. 9. Ratio of aggregates with acetylene concentration at 1673 K.
33.5
32.9
28.1
Linear Branched
28.9
33.5
1573
1673
0
of the reaction between acetylene and acetylene [11]. Thus, since there is a significant reaction between benzene and acetylene, nucleation occur The number of small PAHs, which contribute to surface growth, remains mostly unchanged because an increase in the feedstock amount of carbon per unit time with an increase in the acetylene concentration contributes to nucleation, and therefore, the mean primary particle diameter also remains unchanged. In the case where the acetylene concentration is relatively high (5.0 vol%), Q is smaller than Qs (Q < Qs), and the mean primary particle diameter increases compared with the case in which the acetylene concentration is low. Because the ratio of acetylene to benzene is large, as is the case for low furnace temperatures, a significant reaction occurs between acetylene and acetylene in addition to the reaction between benzene and acetylene. For small PAHs, the ratio increases because the rate coefficient for the reaction between acetylene and acetylene is low with an increase in the furnace temperature [11]. The result of aggregate shape classification in the case where the acetylene concentration changes at 1673 K is shown in Fig. 9. The aggregate shapes become complex as the acetylene concentration increases. This result reveals that the collision rate increases because of an increase in Q. However, the variation is small in the case where the acetylene concentration is 5.0 vol%. This result is due to an increase in the mean primary particle diameter. When the acetylene concentration is 5.0 vol%, the aggregate shapes become complex when compared with the case for which the acetylene concentration is 0.50 vol%. This result implies that the collision rate increases because the effect of an increase in Q
1473
Temperature [K] Fig. 10. Ratio of aggregates with temperature, (a) benzene 1.0 vol%, acetylene 0.50 vol% and (b) benzene 1.0 vol%, acetylene 5.0 vol%.
on the increase in the collision rate is larger than that of the increase in the mean primary particle diameter for a decrease in the rate. 3.5. Effect of furnace temperature Table 5 shows the results for Q and the mean primary particle diameter d of carbon black synthesized for each furnace temperature, where the benzene concentration is 1.0 vol% and the acetylene concentration is 0.50 and 5.0 vol%. For each feedstock, Q increases, and the variation is especially large at 1673 K. This result implies that the rate coefficient of the reaction between benzene and acetylene increases with increasing furnace temperature [18], enabling the formation of large PAHs. The mean primary particle diameter decreases as a result of a decrease in small PAHs that contribute to surface growth and an increase in large PAHs. The results of the aggregate shape classification in the case involving a mixture of 1.0 vol% benzene and 0.50 vol% acetylene, and in the case of 1.0 vol% benzene and 5.0 vol% acetylene are shown in Fig. 10a and b. In each case, aggregate shapes are slightly simplified from 1473 K to 1573 K, whereas they are complicated from 1573 K to 1673 K. In the case of relatively long residence
Table 5 Number of carbon black particles produced per unit time for carbon black particles and mean primary particle diameter with furnace temperature. Acetylene concentration (vol%)
Benzene 1.0 vol%–acetylene 0.5 vol%
Temperature (K) Q (1013 particles/s) d (nm)
1473 1.03 45
1573 1.45 42
Benzene 1.0 vol%–acetylene 5.0 vol% 1673 3.32 30
1473 1.43 52
1573 1.86 47
1673 3.62 38
K. Ono et al. / Chemical Engineering Journal 215–216 (2013) 128–135
times (about 300 ms whereas the condition in the industrial furnace allows under 100 ms), the sintering of primary particles occurs in the case of high temperatures [9]. Thus, aggregate shapes are simplified at 1573 K because of sintering. On the other hand, at 1673 K, since Q is approximately twice of 1573 K and the mean primary particle diameter increases significantly, the collision rate also increases significantly. As a result, aggregate shapes are complicated. 3.6. Effect of hydrocarbon addition on aggregation mechanisms of carbon black When additives are added to feedstock, the formation of large PAHs is promoted because the reaction between benzene and acetylene occurs. An increase in large PAHs causes the promotion of nuclei and number of carbon black particles. Thus, aggregate shapes become complex due to an increase in collision rate. When the concentration of additives increases, the mixture ratio that promotes the chemical reaction may be excessed. The excess hydrocarbon inhibits the formation of large PAHs, and small PAHs that contribute to the surface growth increases. An increase in primary particle diameter by surface growth causes a decrease in collision rate. Thus, aggregate shapes tend to simple at this case. 4. Conclusion In this study, we investigated the influence of the composition of benzene–acetylene on carbon black configurations. We formed carbon black by pyrolysis of a mixture of benzene and acetylene in an inert atmosphere. The effects of the benzene concentration, acetylene concentration, and furnace temperature on the mean primary particle diameter and the aggregate shape of carbon black are investigated. The results are summarized as follows. When the acetylene concentration is changed and the benzene concentration is made constant, aggregate shapes become complex as the acetylene concentration increases. However, in the case where the acetylene concentration is larger than the benzene concentration, the variations in the aggregate shapes are small with an increase in acetylene concentration, because surface growth occurs and the collision rate decreases owing to an increase in the mean primary particle diameter. In the case where there is a change in the furnace temperature, the number of carbon black particles produced per unit time increases. Also, the mean primary particle diameter decreases with an increase in furnace temperature because the rate coefficient for the reaction between benzene and acetylene increases. In addition, aggregate shapes become complex with an increase in furnace temperature. Therefore, from the observations we conclude that nucleation occurs and aggregate shapes become complicated when the ratio of the benzene and acetylene concentration is suitable (about 2:1 in this study), and the furnace temperature is high. However, in
135
the case where benzene or acetylene concentration is large and the furnace temperature is low, the aggregate shapes are simplified because the formation of small PAHs, which contributes to surface growth. References [1] T. Ungár, J. Gubicza, G. Tichy, C. Pantea, T.W. Zerda, Size and shape of crystallite and internal stresses in carbon blacks, Compos. A Appl. Sci. Manuf. 36 (2005) 431–436. [2] J. Lahaye, G. Prado, J.B. Donnet, Nucleation and growth of carbon black particles during thermal decomposition of benzene, Carbon 12 (1974) 27–35. [3] J. Lahaye, F. Ehrburger-Dolle, Mechanisms of carbon black formation. Correlation with the morphology of aggregates, Carbon 32 (1994) 1319–1324. [4] 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. 26 (2000) 565–608. [5] J.B. Donnet, R.C. Bansal, M.J. Wang, Carbon Black, Marcel Dekker, New York, 1993. [6] K. Ono, M. Yanaka, S. Tanaka, Y. Saito, H. Aoki, O. Fukuda, T. Aoki, T. Yamaguchi, Influence of furnace temperature and residence time on configurations of carbon black, Chem. Eng. J. 200–202 (2012) 541–548. [7] F. Shishido, H. Hashiguchi, Y. Matsushita, Y. Morozumi, H. Aoki, T. Miura, An investigation of primary particle growth and aggregate formation of soot using a numerical model considering the sintering of primary particles, Kagaku Kogaku Ronubnshu 33 (2007) 306–314. [8] U. Bonne, K.H. Homann, H.Gg. Wagner, Carbon formation in premixed flames. Symposium (International) on Combustion (Pittsburg): The Combustion Institute, 1965; pp. 503–512. [9] K.H. Homann, H.Gg. Wagner, Some New Aspects of the Mechanism of Carbon Formation in Premixed Flames. Symposium (International) on Combustion (Pittsburg): The Combustion Institute, 1967, pp. 371–379. [10] J.R. Arthur, D.H. Napier, Formation of Carbon and Related Materials in Diffusion Flames, Symposium (International) on Combustion (Pittsburg): The Combustion Institute, 1955, pp. 303–316. [11] M. Frenklach, D.W. Clary, W.C. Gardiner, S.E. Stein, Symposium (International) on Combustion (Pittsburg): The Combustion Institute, 1984, pp. 887–901. [12] H. Wang, M. Frenklach, A detailed kinetic modeling study of aromatics formation in laminar premixed acetylene and ethylene flames, Combust. Flame 110 (1997) 173–221. [13] P.A. Tesner, S.V. Shurupov, Soot formation during pyrolysis and combustion of hydrocarbons, Combust. Sci. Technol. 109 (1995) 399–400. [14] P.A. Tesner, S.V. Shurupov, Soot formation during pyrolysis of benzene–xylene mixtures, Combust. Sci. Technol. 109 (1995) 401–404. [15] P.A. Tesner, S.V. Shurupov, Soot formation during pyrolysis of naphthalene, anthracene and pyrene, Combust. Sci. Technol. 126 (1997) 139–151. [16] P.A. Tesner, S.V. Shurupov, Some physico–chemical parameters of soot formation during pyrolysis of methane and methane–acetylene and methane–benzene mixtures. Symposium (International) on Combustion: The Combustion Institute (Pittsburg), 1994; pp. 653–659. [17] R. Watanabe, T. Shindoh, Y. Matsushita, H. Aoki, T. Miura, K. Nishiwaki, H. Yamada, O. Fukuda, T. Yamagychi, A numerical investigation of the factors influencing the aggregate shape of carbon black from the furnace process, J. Chem. Eng. Jpn. 43 (2010) 150–157. [18] C.R. Herd, G.C. Mcdonald, W.M. Hess, Morphology of carbon-black aggregates, Rubber Chem. Tech. 65 (1992) 107–129. [19] C.R. Herd, G.C. Mcdonald, R.E. Smith, W.M. Hess, The use of skeletonization for the shape classification of carbon-black aggregates, Rubber Chem. Tech. 66 (1993) 491–509. [20] ASTM D 3849-07. Standard test method for carbon black – morphological characterization of carbon black using electron microscopy. [21] P.A. Tesner, S.V. Shurupov, Soot formation from acetylene–benzene mixture, Combust. Sci. Technol. 92 (1993) 61–67. [22] A. Einstein, Über die von der molekularkinetischen Theorie der Wärme geforderte Bewegung von in ruhenden Flüssigkeiten suspendierten Teilchen, Annalen der Physik 322 (1905) 549–560.