Physicochemical properties of soot generated from toluene diffusion flames: Effects of fuel flow rate

Combustion and Flame 178 (2017) 286–296

Contents lists available at ScienceDirect

Combustion and Flame journal homepage: www.elsevier.com/locate/combustflame

Physicochemical properties of soot generated from toluene diffusion flames: Effects of fuel flow rate Gerardo D.J. Guerrero Peña a,b, Abhijeet Raj a,∗, Samuel Stephen a, Tharalekshmy Anjana a, Yousef Adnan Said Hammid a, Joaquin L. Brito b, Ahmed Al Shoaibi a a b

Department of Chemical Engineering, The Petroleum Institute, Abu Dhabi, United Arab Emirates Chemistry Center, Venezuelan Institute for Scientific Research, Altos de Pipe, Miranda, Venezuela

a r t i c l e

i n f o

Article history: Received 29 October 2016 Revised 30 November 2016 Accepted 11 January 2017

Keywords: Toluene Soot Nanostructures HRTEM XRD FTIR

a b s t r a c t Aromatic hydrocarbons are commonly found in fossil-derived transportation fuels, and their combustion in engines produce most of the observed soot particles. Toluene is an important component of gasoline (about 6 wt%), diesel (1–2 wt%), and jet fuels (1–2 wt%), and forms a part of their surrogates. This paper reports the nanostructures and chemical constituents of soot, formed in toluene diffusion flames at different fuel flow rates. High resolution transmission electron microscopy (HRTEM) and X-ray diffraction (XRD) are employed to study the physical properties of soot, while Fourier transform infrared spectroscopy (FTIR), electron energy loss spectroscopy (EELS), and elemental analysis are used to investigate its chemical properties. With increasing fuel flow rate, HRTEM and XRD analyses showed that the lateral size of aromatic layers in soot reduced, while the FTIR analysis revealed that the concentration of aliphatic and oxygenated groups decreased, and that of aromatic group increased. The elemental analysis showed that soot from lower fuel flow rates had more hydrogen and oxygen content than those from higher flow rates. The experimental observations indicate that both physical and chemical characteristics of soot derived from toluene flame are dependent on the fuel flow rate used for its production. © 2017 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

1. Introduction Aromatic hydrocarbons are important constituents of fossilderived transportation fuels, representing about 20–50% by weight in gasoline, 10–25% in diesel, and 5–16% in jet fuels [1–6]. They are frequently used as a fuel additive to improve the anti-knock index and oxidative stability of fuels [7]. A major disadvantage of their presence in fossil fuels is that they are responsible for the generation of soot particles when the fuel is burnt, as their aromatic structure provides a base for the formation of polycyclic aromatic hydrocarbons (PAH) and soot [8]. Due to the harmful effects of soot, the environment protection agencies in various countries have strictly implemented emission standards on soot to improve air quality and to avoid possible threats to human health and environment [9,10]. To develop the emission standards and to come up with innovative routes to minimize soot formation, it is imperative to understand the pathways of its formation, and the effects of feed conditions on soot emissions from fuels.

∗

Corresponding author. E-mail address: [email protected] (A. Raj).

Several studies in the literature are focused on understanding the pathways of soot formation during fuel combustion, and their dependence on the type of fuel and the process conditions for combustion [11,12]. Soot formation consists of several steps, including the growth of small aromatics to PAHs, which are known as soot precursors, soot nucleation from PAHs, and its growth through agglomeration and chemical reactions including gas-phase species [13]. Soot can also undergo oxidation and fragmentation in flames [14,15]. Its formation during the combustion of aromatic fuels is important, as the fuel molecules provide some species such as cyclopentadienyl (C5 H5 ) [16–18] and benzyl (C6 H5 CH2 ) [17–19] that promote the formation and growth of PAHs in flames [20,21], and thus, support soot formation. Out of all the aromatic fuels, toluene has gathered significant attention in the literature [18,22–27] due to its presence in an appreciable quantity in transportation fuels (about 6 wt% in gasoline [28], 1–2 wt% in diesel [29] and 1–2 wt% in jet fuels [29]). It also forms an important part of their surrogates [3,30,31], and contributes to soot emissions from these transportation fuels. Some works related to the characterization of soot emitted from toluene flames are highlighted below. In [32], soot particles collected from the flames of n-heptane, toluene, their mixture, and gasoline were analyzed. While soots from these fuels exhibited onion-like arrangements of fringes in

http://dx.doi.org/10.1016/j.combustflame.2017.01.009 0010-2180/© 2017 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

G.D.J.G. Peña et al. / Combustion and Flame 178 (2017) 286–296

their HRTEM images, the degree of order within each soot sample varied. Toluene soot appeared to be the most graphitic with large concentric shell of parallel PAHs and less disordered core. It is shown in earlier studies as well that soot characteristics such as its chemical constituents and the relative composition of crystalline and amorphous-like phases in soot are highly dependent on fuel type and the experimental conditions (e.g., temperature, fuel flow rate, pressure, equivalent ratio, and retention time) [33–37]. In general, the combustion of aliphatic fuels generates soot particles with a lower degree of graphitization than those from the combustion of aromatic fuels [38,39] due to faster carbonization in the post fuel-oxidation zone of aromatic flames [40,41]. In [42], the chemical characterization of soot particles emitted from hexane and diesel surrogate containing toluene was presented. It was found that the surrogate fuel produced much higher quantity of soot as compared to hexane due to its aromatic content, though the particles in hexane flame, on an average, were found to be larger than that from the surrogate fuel, possibly due to high temperatures that allowed soot particles to grow at a fast rate in hexane flame. In [43], the sooting properties of real fuels and their surrogates (such as toluene and n-heptane) were studied, and it was reported that, in order to meet the sooting tendencies of real fuels, surrogate fuels were required to have the appropriate amount of aromatics in them, as they have a good impact on amount of soot formed from real fuels. In [44], the sooting tendency of the surrogates for the aromatic fractions of diesel and gasoline (toluene, tetralin, trimethylbenzene, and butylbenzene) was studied. Though toluene was the smallest aromatic species among the fuels studied, it produced soot particles that were larger than those from butylbenzene and similar to those from trimethylbenzene (but, smaller than the particles from tetralin that is a two-ring species). In [45], the sooting tendencies of heptane and n-heptane/n-propylbenzene premixed flames were studied. It was shown that n-heptane/npropylbenzene flame led to more condensed species on soot at lower flame heights that were responsible for higher soot mass in it as compared to heptane flame. Thus, even among aromatics, fuel structure plays an important role in their sooting tendencies. The physical properties of in-cylinder soot produced from the combustion of n-heptane and toluene/n-heptane mixture in a diesel engine were studied in [46]. Toluene/n-heptane soot exhibited more number of primary particles with larger size as compared to nheptane soot. The differences in soot nanostructures such as the fringe length and interlayer spacing were also observed. In [47], the structures of soot formed from hexane and benzene flames were determined using UV spectroscopy. It was found that the fringes in benzene soot were more organized and curved than those present in hexane soot. Along with the type of fuel, the variation in soot properties with operating conditions such as fuel flow rate has also been discussed in the literature, as highlighted below. In [48], the effect of the variation in fuel flow rate, temperature, and residence time on soot nanostructures was studied. For high fuel flow rate, low temperature, and low residence time, the presence of fullerenic nanostructures or curved PAHs were observed, whereas a low fuel flow rate, high temperature, and high residence time resulted in graphitic soot. In [49], the effect of fuel flow rate and temperature on the evolution of soot particles in diffusion flames was investigated. A higher fuel flow rate resulted in higher temperature in the soot inception region near the burner, but a lower temperature in the sooting zone due to radiative heat loss from soot. With increase in fuel flow rate, the amount of generated soot increased. The changes in temperature, diameter, and the number density of particles with increasing fuel flow rate were studied in [44], where it was observed that increasing flow rate gradually decreased the soot number density and increased the average soot aggregate size. Similar trends in soot parameters were observed in [50,51]. The effect of fuel flow rate on soot growth

0.007 g/min 8±1 mm

287

0.021 g/min 0.039 g/min 0.057 g/min 0.085 g/min 10±1 mm 12±1 mm 14±1 mm 16±1 mm

Fig. 1. Images of toluene flames with different flame vertical extents (mm) and fuel flow rates (g/min) that were used for soot collection.

and oxidation was studied in [52], where, at low flow rates, flames were non-sooting due to rampant soot oxidation. However, at high fuel flow rates, soot oxidation rate was greatly diminished due to the low concentration of O2 molecules, and hence, soot particles were released from the flame tip. While several investigations prove the dependence of experimental condition and fuel properties on soot emissions [48,53], the data related to the physico-chemical properties of soot from smoking aromatic flames remains limited in the literature. Along with fuel sooting tendencies, it is necessary to study the characteristics of soot particles generated from them, since the harmful effects and the reactivity towards oxygen of soot particles depend highly on their nanostructures and the elemental composition [54]. The aim of this study is to evaluate the effect of fuel flow rate on the nanostructures and the chemical constituents of soot formed from toluene in a diffusion flame. Toluene is selected for this study due to its presence in appreciable quantities in real fuels (specially, gasoline), and its high tendency to produce soot. A wick-fed burner will be used to collect soot particles from toluene flames at different fuel flow rates. The collected particles will be characterized using different techniques to understand the reasons behind the observed variation in the chemical composition and the nanostructural parameters of soot with varying fuel flow rate. 2. Experimental details Soot particles were collected from the tip of diffusion flames of toluene at atmospheric pressure, using the smoke point apparatus. A description of this apparatus and soot collection technique are provided in [55]. This apparatus consists of a cylindrical fuel reservoir with a circular nozzle encompassing the wick, which is used to supply the fuel and to generate a diffusion flame. The fuel flow rate to the flame is varied by raising the wick exposure height above the nozzle through a screw mechanism (as a result, the vertical extent or height of the flame also increases). In the rest of the paper, the height of the flame has been referred to as the vertical extent of the flame, since the experimental data at different flame heights in the literature usually refer to the results from different vertical locations in a flame. However, in this paper, the results have been reported for flames of different heights produced with different fuel flow rates. For soot collection from the tip of toluene flames, a filtration system consisting of a particulate filter holder (from Sierra Instruments, USA), a 70-mm borosilicate microfiber filter (TX40H120WW, United Filtration Systems, USA), and a vacuum pump was used to allow the exhaust gas exiting the chimney of the smoke point apparatus to pass through the filter, where soot particles would be entrapped. This apparatus was chosen for this study on soot characteristics due to its capability of generating a stable flame that allows the collection of soot particles at various flame vertical extents above the smoke point of 7.5 mm by varying the fuel flow rate. Figure 1 provides the photographs of the flames with vertical extents of 8, 10, 12, 14 and 16 mm (with an uncertainly of 1 mm). The fuel flow rates are also provided. A higher fuel

288

G.D.J.G. Peña et al. / Combustion and Flame 178 (2017) 286–296

flow rate (leading to a flame vertical extent of more than 16 mm) was avoided since this caused flame instability and fluctuations making it difficult to reliably measure fuel weight loss with time by maintaining a constant flame vertical extent. To eliminate any loosely attached organic fraction and moisture, the soot samples were dried in the N2 environment in a tube furnace prior to conducting any characterization experiments. To identify the functional groups, present on soot surface, infrared spectra were generated using a Fourier transform infrared (FTIR) spectrometer (Bruker Vertex 70) that runs under a spectral range of 40 0–40 0 0 cm−1 and with a resolution of 4 cm−1 . The samples for FTIR analysis were prepared by mixing dried soot with FTIR grade potassium bromide (0.02 mg of soot in 10 mg of potassium bromide), and using a manual press machine to make pellets. The prepared pellets were 1 mm in thickness and 8 mm in diameter. Three spectra were acquired for each sample to make sure that the results are reproducible. To obtain quantitative information about soot elemental constituents, an elemental analyzer (EuroVector EA-30 0 0) was used. Soot samples were weighed (about 0.8–1.1 mg), and then, packed in tin capsules before analysis. These experiments were repeated three times for each sample to ensure reproducibility of the results. The nanostructure of soot particles was visualized using their HRTEM images from FEI Tecnai G20 instrument, operated at a voltage of 200 kV with 0.24 nm point resolution and fitted with an energy-filtered camera. Soot particles were dispersed in ethanol (less than 0.1 mg of soot sample in 4 ml of ethanol), and ultrasonicated for five minutes. Thereafter, a droplet of sonicated suspension was directly placed onto a copper microscope grid covered with perforated carbon that was then dried under an infrared light for 20 min. X-ray diffraction analysis was conducted to observe any change in the crystalline parameters of soot particles with increasing fuel flow rate. XRD patterns were obtained using a Panalytical X’Pert Pro powder X-ray diffractometer by scanning the soot samples in the 2θ range of 10–110° with a step size of 2θ = 0.02° and an an˚ gular speed of 10 s/step. The Cu-Kα X-ray source (λ = 1.5406 A) operated at a voltage of about 45 kV and at a current intensity of 40 mA. To remove the background from XRD patterns, data processing was done using High Score Plus software, where the base points were manually selected to subtract the background from the data. In order to gain information about the relative composition of sigma (σ ) and pi (π ) bonds in soot, electron energy loss spectra (EELS) were recorded using Gatan Image Filter in the TEM mode with energy resolution maintained at 1 eV and with a dispersion rate of 5 eV/channel. 3. Results and discussion 3.1. Fuel flow rate and soot collection The soot yield (mass of collected soot/mass of fuel burnt × 100), the amount of soot collected on the filter per hour, and the fuel flow rate measured at different flame vertical extents, are presented in Fig. 2. For calculating the fuel flow rate at various flame vertical extent, the continuous recording of fuel weight loss in the burner with time was required. Fuel weight loss versus time was then plotted, and an excellent linearity was found. The slope of the plot provided the fuel flow rate. It is an important parameter that mimics the diesel engine experimental conditions, where soot emissions are measured at a constant fuel flow rate [46,56]. From Fig. 2, it can be seen that, with increasing flame vertical extent, while fuel flow rate increased almost linearly, soot production increased rapidly for flame vertical extent above 10 mm. The sharp

increase in soot production in the flames of larger vertical extent is mainly due to higher amount of fuel and lower oxygen concentration in the flame, which leads to incomplete fuel combustion, reduced soot oxidation, and decreased flame speed that supports soot formation [57,58]. Since fuel flow rate increases with increasing flame vertical extent, it would also lead to increased formation of PAHs through fuel pyrolysis or through the combination of unburnt hydrocarbons to enhance the kinetics of soot formation [25]. These results agree with the findings of Kent and Wagner [59], where flames with higher fuel flow rates were found to produce more soot. Since temperature plays an important role in controlling the kinetics of soot formation and oxidation, its role in soot emissions should also be considered. In [44,50], the temperatures at the tip of toluene flames generated using smoke point apparatus (similar to the one used in this work) were found to decrease with increasing flame vertical extent from about 3 to 8.2 mm (with a smoke point of 7.8 mm). While soot production increased with increasing flame vertical extent, flame tip temperature decreased. This is a result of high radiative heat loss from the luminous region of the flame with high amounts of soot. However, a low flame tip temperature may not necessarily mean a low magnitude of maximum temperature in the flame. It has been shown experimentally in [59] that flames with higher fuel flow rate had lower temperature in the post-flame (sooting) region, but had a higher flame temperature near the burner (where soot was incepted). Moreover, it was observed in [59] that high soot emissions at high fuel flow rates are also a result of temperature drop in the luminous region of the flame through radiative heat loss that prohibits soot burnout by oxidants. The flame temperatures could not be measured in this work due to the unavailability of the required setup. Along with temperature, the residence time inside the flames also affects the kinetically controlled chemical processes that lead to soot formation. It has been shown experimentally in [59] that with increasing fuel flow rate, the residence time in the flame increases, thus allowing more time for soot inception and growth. This supports the increasing trend in soot formation rate with rising fuel flow rate. Moreover, along with residence time, the temperature near the fuel nozzle also increases with increasing fuel flow rate that supports the formation and growth of soot precursors (i.e. PAHs) through the reactions involving benzyl radicals and other unburnt hydrocarbons and H atoms. This further aid in the formation of soot particles in high amounts. With increasing fuel flow rate, while the sooting tendency of toluene flame varies significantly, it would be worthwhile to determine the changes in soot characteristics with it. 3.2. High-resolution transmission electron microscopy (HRTEM) Figure 3 presents the TEM images of toluene soots, collected from the flames of different fuel flow rates. Soot aggregates with near-spherical primary particles were observed in the samples. A closer look at the images revealed that soot from higher fuel flow rate contained larger number of primary particles in the clusters, which is in line with the observation of enhanced soot production with increasing fuel flow rate in Fig. 1. The high amounts of soot particles incepted in the flames of higher fuel flow rate along with the low temperatures in the post-flame region and the higher residence time of soot in the flame contribute to the larger size of soot aggregates, as all the above conditions favor particle aggregation after collision [60]. For quantitative analysis, the diameter of the primary particles in soots from the HRTEM images was measured using Gatan software. For each fuel flow rate, several primary particles from different aggregates were selected to determine the average size. For the fuel flow rates of 0.007, 0.021, 0.039, 0.057 and 0.085 g/min, the

a soot collected (mg)

1000

soot yield, % 800 600 400 200

0.12

8

10

12

14

16

y = 0.0096x - 0.0726 R² = 0.984 0.08

0.04

0

0 6

289

b

1200

Fuel flow rate, g/min

20 18 16 14 12 10 8 6 4 2 0

Collected soot mass, mg/hr

Soot yield, %

G.D.J.G. Peña et al. / Combustion and Flame 178 (2017) 286–296

6

18

8

10

12

14

16

18

Flame vertical extent, mm

Flame vertical extent, mm

Fig. 2. (a) Soot yield and collected soot mass in toluene flame with increasing flame vertical extent, (b) fuel flow rate with increasing flame vertical extent, where a linear trend line for fuel flow rate with its equation is shown.

a

c

b

d

e

Fig. 3. TEM images (50 nm resolution) of soot particles at different fuel flow rates: (a) 0.007 g/min, (b) 0.021 g/min, (c) 0.039 g/min, (d) 0.057 g/min, and (e) 0.085 g/min.

average primary particle diameters were 20.11, 21.29, 25.25, 27.95, and 40.95 nm, respectively. Clearly, larger primary particles were formed with increasing fuel flow rate, which points toward an increase in the soot inception and growth rates. Soot fragmentation may also contribute to this trend. At low fuel flow rates (i.e., low sooting flames), with high temperature and oxygen content, soot fragmentation can take place, as observed in [14,15,61] which can lead to soot particles of small size. This phenomenon would be less likely to take place at high fuel flow rates, where temperature and oxygen concentration are low. The residence time of soot particles in flames would also play a role in the size of primary particles. At higher fuel flow rates, the increase in residence time would also support the growth of soot particles to gain larger diameters through PAH condensation and surface growth by chemical reactions. It is worth mentioning that the numbers of the primary particles in the TEM images can be affected by the method of the preparation of the sample on the grid. For example, the amount of ethanol evaporated could affect the experimental observations. This may lead to some uncertainties in the measured particle diameters. Since the standard procedure for sample preparation was followed, where particle dispersion was relatively constant, and the

values reported in this work are consistent with the range of values reported in the literature (an average of 51 nm in [62], 25– 30 nm in [63], and 10–60 nm in [50]), the analysis of the effect of sample preparation on the results is not presented. Figure 4 presents the HRTEM images of toluene soots at a resolution of 5 nm. These images show that all soots have turbostratic structures, which has also been observed in [38,64,65]. Such structures indicate the presence of both graphitic and amorphous states in soot. The near-parallel polycyclic aromatic layers ordered in an onion-like structure constitute the graphitic part of soot, and its amorphous nature arises from the presence of PAH clusters in incoherent arrangement [66]. The core of all the soot samples was composed of PAHs with incoherent orientation, and nebulous boundaries between inner core and outer shell were observed. The particles appeared partially graphitic and partially amorphous, possibly due to relatively short residence times in flames that do not allow complete soot graphitization (graphitization is a process that occurs in carbonaceous materials where they undergo structural modifications under high temperature environment to produce compact structures comprising of large PAHs with parallel orientations and with less defects). The fringes or the fine lines in the images in Fig. 4 represent graphene-like segments or PAHs

290

G.D.J.G. Peña et al. / Combustion and Flame 178 (2017) 286–296

a

c

b

d

e

Fig. 4. HRTEM images (5 nm resolution) of toluene soot samples collected at a fuel flow rate of (a) 0.007 g/min, (b) 0.021 g/min, (c) 0.039 g/min, (d) 0.057 g/min and (e) 0.085 g/min.

3.3. X-ray diffraction (XRD) Figure 5 presents the diffraction patterns of toluene soots collected from the flame tip at different fuel flow rates. Two recognizable peaks in the 2Ѳ range of 0–60° were observed in the XRD patterns for all soots, which goes in line with the results presented in [66,68]. In general, the peaks at 25° and 44° originate from the (002) and (100) Bragg reflection, respectively. The existence of substantial background intensity in the XRD pattern indicates the presence of amorphous carbon in soot, which is difficult to characterize. For all soots collected at different fuel flow rates, the peak near 2θ = 25° was found to be broad, as opposed to a sharp peak for graphite. This shows an inadequate graphitization in soot, which leads to a lower long-range structural order with

0.085 g/min

1 Normalized Intensity

existing in soot [32,67]. The fringes appeared to be more organized at lower fuel flow rate especially near the periphery, where some near-parallel PAHs were found. This is presumably because the higher temperatures at lower fuel flow rate led to restructuring or rearrangement of PAHs in soot. For the quantitative information on the fringes, the HRTEM images were examined using Gatan software to measure the fringe lengths and the distance between the PAH layers. The average fringe lengths were 10.8, 10.5, 10.4, 10.0, and 9.9 A˚ for fuel flow rate of 0.007, 0.021, 0.039, 0.057, and 0.085 g/min, respectively. Evidently, as the fuel flow rate increased, the average fringe lengths slightly reduced. The reduction in average fringe length with increasing fuel flow rate may not be an indication of reduced soot growth rate, since it may also result from the enhanced condensation of small PAHs on soot that are formed in higher amounts at higher fuel flow rates. This trend will be further confirmed and discussed in Section 3.3. The average distance between PAH layers was found to be 4 A˚ for all the soot samples, which is higher than ˚ This further suggests that none of the that for graphite (3.35 A). soot particles generated from toluene flames at different fuel flow rates was completely graphitized.

0.007 g/min 0.021 g/min

0.8

0.039 g/min 0.057 g/min

0.6

0.085 g/min 0.4 0.2

0.007 g/min

0 10

15

20

25

30

35 40 Angle, 2ϴ

45

50

55

60

Fig. 5. X-ray diffraction pattern of toluene soot particles collected at different fuel flow rates.

small crystallites and a high amorphous character [69,70]. The low intensity of the peak near 2θ = 44° suggests a small size of PAHs in soot. It is also an indication that more carbons are embedded in the amorphous matrix and less in the nanocrystallites [66,71]. With increasing fuel flow rate the peak intensities increased, which confirms structural changes in soots. The asymmetric characteristic of the (002) Bragg reflection results from the presence of another peak (γ ) on the left side, which is indicative of the presence of aliphatic chains on PAHs in soot [72]. The lack of any observable peak related to the γ band in the XRD pattern of all the soot samples makes it difficult to quantify the amount of aliphatics in soot, which is also discussed in [55]. Though it is difficult to observe in Fig. 4, but with increasing fuel flow rate, this peak became more symmetric. This suggests that the aliphatic content in soot reduced and the aromatic content in it increased (which is confirmed through the analysis of EELS and FTIR spectra later in this paper).

30.28 26.89 25.05 24.22 23.50

3.38 3.41 3.43 3.47 3.50

700-1000

The analysis of experimental XRD data can reveal quantitative information about the crystalline structure in soot and their changes with increasing flow rate. Four crystallite parameters were calculated from the XRD pattern of each soot sample by following the method given in literature [67,73,74]. The interlayer spacing (d002 ) between PAHs, the stacking height (Lc ) of the PAH stack, and the lateral size (La ) of the aromatic lamellae were determined using the Bragg’s law (Eq. (1)), the Scherrer formula (Eqs. (2) and (3)), respectively. The average number of layers in a stack (N) were calculated using Eq. (4).

d002 =

λ

2sinθ002

1000 - 1300

(1)

Lc =

0.9λ Bc cosθ002

(2)

La =

1.84λ Ba cosθ100

(3)

N=

LC d002

(4)

In these equations, Ba and Bc are the full widths at half maximum (FWHM) for (100) and (002) peaks, respectively, and θ 002 and θ 100 are the angles at which (002) and (100) peaks occur. Matlab software was used to fit Gaussian curves through (100) and (002) peaks to obtain Ba , Bc ,θ 002 , and θ 100 , and to calculate the crystallite parameters. Table 1 summarizes the values of the crystalline parameters for toluene soots collected from different fuel flow rates. The values of d002 were almost identical for all the samples, and exceeded the distinctive value of 3.35 A˚ for graphite [67]. This further confirms the findings from HRTEM that a high level of disorder in PAH arrangement in soot exists. In the literature, the majority of the measurements of interlayer spacing in soot obtained from flames and diesel engines [70,73–76] and in carbon black [72,77,78] have ˚ The d002 values detershowed values in between 3.4 and 3.6 A. ˚ and thus, mined in this work are in the range from 3.58 to 3.60 A, are in good agreement with those from previous investigations. The values of Lc in the soot samples varied from 12.22 to ˚ and La varied from 30.28 to 23.50 A, ˚ as the fuel flow 12.55 A, rate was increased from 0.007 to 0.085 g/min. Some recent studies [76,79,80] on soot nanostructures using XRD have shown that La in ˚ soot is near 30 A˚ which is far from its value for graphite (>100 A) [81,82]. The parameter, Lc varies significantly (usually between 6 and 30 A˚ [67,73–76,78,80,83]) depending upon the source of soot particles. The values of La and Lc are in good agreement with their range of values reported in the literature. The Lc values slightly increased with increasing fuel flow rate, which indicates a small increase in the size of PAH stacks. The number of PAH layers in stacks, N, provided in Table 1, however, showed negligible variation. The parameter, La showed a clear trend, where its value was found to decrease with increasing fuel flow rate. Generally, with decreasing oxygen concentration in the flame (as the fuel flow rate is increased), longer PAHs are expected [73,84], as soot oxidation is less relevant in such a case, which is not in line with the results obtained here. However, it should be noted that, as the fuel flow

0.085 g/min

O-H

12.22 12.28 12.34 12.45 12.55

C-H aliphatics C-H aromatic

3.60 3.60 3.59 3.58 3.58

C=C

N, layers

0.007 0.021 0.039 0.057 0.085

780 830 880

La , A˚

0.057 g/min

700

Lc , A˚

0.039 g/min

0.021 g/min

715

d002 −1 , A˚

Absorbance

Fuel flow rate, g min

C=O

Table 1 Structural parameters obtained from XRD pattern analysis.

291

Symmetrical C-H bending Asymmetrical C-H bending

G.D.J.G. Peña et al. / Combustion and Flame 178 (2017) 286–296

0.007 g/min

500

1000

1500

2000

2500

3000

3500

4000

Wavenumber, cm-1 Fig. 6. Fourier transform infrared (FTIR) spectra of soots collected from the tip of toluene flames at different fuel flow rates.

rate was increased, temperature near the flame tip reduced and residence time increased. This would enhance the condensation of small PAHs on soot surface, as their collision efficiency is inversely proportional to temperature [60]. While low temperatures support successful addition of a PAH on soot after their collision, high temperatures can lead to the loss of small PAHs from soot that have smaller binding energy than the larger ones [60]. As mentioned above, the higher residence time for soot particles in flame environments would also enhance the number of collisions of PAHs with soot surface, thus increasing the possibility of their condensation on soot. The significant increase in the primary particle diameter with increasing fuel flow rate, as reported above, also supports the possibility of enhanced soot growth through the condensation of aromatics. While the trend shown by the fringe lengths, acquired from the HRTEM images, is same as that from La values, the latter ones are higher than the former ones at all fuel flow rates. In [32], Botero et al. estimated the fringe lengths from the HRTEM images of soot particles collected from inside and from the tip of gasoline and toluene flames. At the flame tip, average fringe lengths in ˚ respectively. Simigasoline and toluene soots were 9.7 and 9.9 A, lar results on PAH size from HRTEM images in toluene soot were reported in [38], which are comparable to the values obtained in this work. The difference between La from XRD analysis and fringe lengths from HRTEM images has already been discussed in [75,85]. The peak heights in XRD pattern is highly influenced by the amount of large crystallites. Even if such crystallites are small in concentration, they may have large contribution in the peak intensity. Thus, the PAH size may be over-predicted from XRD analysis. Similarly, HRTEM provides a two-dimensional projection of three-dimensional soot particles, and thus, may encompass some inaccuracies, when a quantitative analysis of such images are conducted.

3.4. Fourier transform infrared (FTIR) spectra The infrared analysis of toluene soot was done to assess the variation in the functional groups on soot, as the fuel flow rate increases. Figure 6 provides the baseline-corrected and smoothed FTIR spectra of different soot samples obtained from different fuel flow rates. The absorption bands in the spectrum for all soot samples give an indication that the same functional groups exist in them (aromatic base with aliphatic chains and oxygen-containing groups), but different intensities can be observed for some peaks, which indicate some differences in their chemical composition.

G.D.J.G. Peña et al. / Combustion and Flame 178 (2017) 286–296

Absorbance

292

0.085 g/min 0.057 g/min 0.039 g/min 0.021 g/min 0.007 g/min

1500

1550

1600

1650

1700

1750

1800

Wavenumber, cm -1

Fig. 7. Fourier transform infrared spectra in the range of 150 0–180 0 cm−1 with merging C=O and C=C bands.

The prominent and broad absorption band observed in the spectra between 3650 and 3300 cm−1 corresponds to –OH stretching modes from different carbon environments such as carboxylic acid and phenolic groups on soot surface that are generated during the partial oxidation of soot particles [86–89]. The contribution of –OH stretching modes from carboxylic acids may be dismissed here, because they decompose in flames [89,90]. Though care was taken to ensure that soot and potassium bromide are devoid of moisture, but, while creating their mixture for FTIR experiments, water molecules from ambient air may have got adsorbed on soot surface to contribute to the –OH stretching mode [91,92]. However, the low hydrophilicity of soot suggests that water molecules may be present only in a small quantity [93]. In the spectra, soot from the flame with 0.007 g/min of flow rate had the highest intensity for this band, while the soots from higher fuel flow rates had nearly the same peak intensity. For all the soots, the mode for aromatic C–H stretching (at 3050 cm−1 ) with low absorbance value is attributed to polycyclic aromatic hydrocarbons with low hydrogen content that are present in soot [94]. The modes for aliphatic groups (at 2950, 2925, and 2860 cm−1 ) with similar peak intensities varied with increasing fuel flow rate. These C–H stretching bands stem from methyl symmetric, methyl asymmetric, and methylene asymmetric modes [22,34,86,95,96]. These groups may be present in the form of aliphatic chains or saturated rings linked to the aromatic rings in polycyclic aromatic hydrocarbons, or may be forming methylene bridges to connect different PAHs [22,34,86,95,96]. A high flame temperature in the sooting zone helps in the formation of aromatic-aliphatic linked structures [97,98], which supports the higher intensity of this band at lower fuel flow rate. Moreover, as the higher fuel flow rates provide higher residence time, soot dehydrogenation (the loss of H atoms during the conversion of open structures to rings through bond formation between two nearby C atoms) [99] which is a kinetically controlled process, can also contribute to lower H-content in soots. The absorption band for aromatic C=C stretching mode near 1625 cm−1 was observed to increase with increasing fuel flow rate. This indicates that the aromatic character in soot increases as the fuel flow rate is increased. Figure 7 presents the FTIR spectra in the range of 1500– 1800 cm−1 to highlight that, at all the fuel flow rates, the band for carbonyl group (around 1700 cm−1 ) appears to be merging with the band for C=C bond. This is a result of several factors such as conjugation of C=C and C=O bonds [100], interaction between ke-

tone functional groups and localized C=C bonds [101], and delocalization of the π electrons of both unsaturated groups caused by intermolecular and intramolecular hydrogen bonding with hydroxyl group [101]. The carbonyl band can arise from aldehydes, anhydrides, esters, lactones, ketone, and quinone-type oxygenated group on soot [86]. The contribution of aldehyde group is dismissed because the absorption bands for aldehydic C–H stretch and the first overtone of the aldehydic C–H bending vibration between 2830 and 2700 cm−1 were not observed [86]. The absorption band for carbonyl group appears to decrease, as the fuel flow rate is increased. This is an indication of the reduced availability of oxidizers in the flames of higher fuel flow rates, which leads to a reduction in the oxygenated functional groups on soot A band, observed at 1460 cm−1 , was allotted to methyl asymmetrical C–H bending or methylene scissoring [86]. Another band corresponding to methyl symmetrical C–H bending or methylene wagging at 1419 cm−1 was also observed [86]. The intensity of these bands remains almost constant with increasing fuel flow rate. The bands in the range of 10 0 0 to 130 0 cm−1 near the fingerprinting region is a complex part of the spectra that is difficult to analyze because the signals corresponding to C–H in-plane bending bands of PAHs overlap with the signals corresponding to C–O bond stretching for phenols, anhydrides, esters and ether-like group [86]. For this reason, these absorption bands cannot reliably be allocated to any bond type. The intensity of the absorption bands in this range slightly decreases with increasing fuel flow rate, suggesting that a greater range of oxygenated functionalities are associated with the samples collected at lower fuel flow rates. The fingerprinting region between 10 0 0 and 700 cm−1 is also hard to analyze, but its low intensity bands represent aromatic out-of-plane C–H bending vibrations that are characteristics of diand tri-substituted aromatic compounds or highly substituted aromatic hydrocarbons [102,103]. The infrared absorption bands for peri-condensed aromatic hydrocarbon are located at 880, 830, and 780 cm−1 . The cata-condensed aromatic hydrocarbons show absorption peaks at 880, 830 and 700 cm−1 . The nonplanar PAHs containing five membered rings show absorption bands near 910 cm−1 and between 830 and 715 cm−1 [102,104]. The spectra in this region are of low intensity, which suggests low hydrogen content at the periphery of PAHs. Negligible differences in these bands were observed with increasing fuel flow rate. 3.5. Electron energy loss spectra (EELS) Figure 8 presents the carbon K-edge EELS for toluene soots in the range of 275–315 eV. The EELS peaks can give quantitative data related to the compositions of sp3 and sp2 hybridized carbons in soot, and can aid in interpreting the electronic structure of soot samples. A small and sharp peak near 285 eV is caused by 1s→π ‫٭‬ transitions, and it represents carbon atoms in C=O and C=C bonds with sp2 hybridization in soot. The second peak in the spectra at around 290 eV occurs from 1s→σ ‫ ٭‬transitions, and it represents carbon atoms with sp3 hybridization (i.e. C–C, C–H and C–O bonds) in soot [105–108]. The quantitative information about the electronic structure of soot is readily obtained from the ratio of the two peak intensities, represented by π ‫٭‬/ σ ‫ ٭‬ratio, which is indicative of the relative concentration of the two kind of chemical bonds in soot [109]. A large value of π ‫٭‬/ σ ‫ ٭‬ratio in soot particles gives evidence for a high quantity of aromatic hydrocarbons (π bonding) and a low amount of cyclic or acyclic aliphatics (σ bonding) in them [67,105,106]. For toluene soots, π ‫٭‬/ σ ‫ ٭‬ratios are provided in the figure inset. It can be seen that this ratio increased with increasing fuel flow rate, which clearly indicates an increase in π bonding in soot with increasing fuel flow rate, which is consistent with the results from FTIR spectra. While EELS and FTIR spectra reveal bond-

G.D.J.G. Peña et al. / Combustion and Flame 178 (2017) 286–296

a

b

4.0E+5

4.0E+5 EELS Intensity, a.u.

EELS Intensity, a.u.

293

3.0E+5 2.0E+5

0.556 ± 0.003

1.0E+5 0.0E+0

275

285

295

305

3.0E+5 2.0E+5

0.613 ± 0.014

1.0E+5 0.0E+0

315

275

285

Energy loss, eV

c

305

315

d 7.0E+5 EELS Intensity, a.u.

5.6E+5 EELS Intensity, a.u.

295 Energy loss, eV

4.2E+5 2.8E+5

0.643 ± 0.015

1.4E+5 0.0E+0

5.6E+5 4.2E+5

0.668 ± 0.0024

2.8E+5 1.4E+5 0.0E+0

275

285

295

305

315

275

285

Energy loss, eV

295

305

315

Energy loss, eV

e EELS Intensity, a.u.

7.0E+5 5.6E+5 4.2E+5

0.721 ± 0.029

2.8E+5 1.4E+5 0.0E+0 275

285

295

305

315

Energy loss, eV Fig. 8. Electron energy loss spectra of soot samples from toluene flames with fuel flow rates of (a) 0.007 g/min, (b) 0.021 g/min, (c) 0.039 g/min, (d) 0.057 g/min and (e) 0.085 g/min. The values of π ‫٭‬/ σ ‫ ٭‬ratio are provided in the figure inset.

ing information about soot, their quantitative comparison is challenging. The π ‫ ٭‬and σ ‫ ٭‬peaks in EELS present integrated results of all type of pi and sigma bonds, and the contribution of individual bond is unknown, whereas FTIR provides the absorption intensity from different functional groups (with some peaks in the fingerprinting region that are hard to allocate to any functional groups).

3.6. Elemental analysis The elements present in soot particles were identified through elemental analysis by oxidizing the soot samples in the presence of excess oxygen [110]. Table 2 presents the elemental composition (in wt%) of soot particles with increasing fuel flow rate. Soot samples were found to contain C, H, and O atoms only (no N and S were detected). The composition of O atoms in soot was quantified by subtracting the wt% of C and H from 100, as it has been reported in previous studies [55]. With increasing fuel flow rate,

Table 2 Elemental composition (weight %) of toluene soots collected at different fuel flow rates. Fuel flow rate, (g/min)

C (wt%)

H (wt%)

O (wt%)

0.007 0.021 0.039 0.057 0.085

82.12 ± 0.22 84.19 ± 0.65 86.65 ± 0.61 89.38 ± 0.21 92.83 ± 0.35

1.37 ± 0.05 1.30 ± 0.02 1.26 ± 0.03 1.20 ± 0.02 1.17 ± 0.01

16.51 ± 0.45 14.51 ± 0.04 12.09 ± 0.15 9.42 ± 0.22 6.50 ± 0.37

the carbon content in soot increased, and the hydrogen and oxygen content decreased. Figure 9 presents the atomic H/C and O/C ratios in toluene soots. The H/C ratio decreased with increasing fuel flow rate, which suggests an increase in the aromatic character of soot particles. This is because, with increasing fuel flow rate, more PAHs are formed during soot inception to lead to a high amount of soot. At

0.21

0.16

0.2

0.14

0.19

0.12

0.18

0.1

0.17

0.08

0.16

0.06

0.15 0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

O/C ratio

G.D.J.G. Peña et al. / Combustion and Flame 178 (2017) 286–296

H/C ratio

294

0.04 0.09

Fuel flow rate, g/min Fig. 9. The H/C and O/C ratios with increasing fuel flow rate. Solid symbols: H/C ratio. Open symbols: O/C ratio.

low fuel flow rate, higher aliphatic content leads to high hydrogen content in soot. The trend of the absorption bands for aromatic and aliphatic groups, seen in the FTIR spectra, supports the decrease in the H/C ratio of soot with increasing fuel flow rate. The O/C ratio decreased with increasing fuel flow rate, which is in line with the FTIR results, where the intensities related to oxygenated functional groups on soot surface reduced with increasing fuel flow rate. The high oxygen content can be a result of low residence time of soot in the flames of low fuel flow rates, as the oxygenated groups bonded to PAHs comprising soot are unable to desorb by forming CO and CO2 [22,88]. Interestingly, if all the above results are analyzed together, with increasing fuel flow rate, the HRTEM and XRD suggest an increase in the amorphous character of soot, while FTIR, EELS, and elemental analysis indicate a reduction in H and O content in soot along with an increase of aromatic character that generally indicate more ordered (or, graphitic) structure of soot. As the toluene flow rate is increased, the oxygen present in the flame is consumed at an early stage (leading to a high temperature near the nozzle), which makes the sooting zone in the flame oxygen-deficient. While hightemperature, oxygen-deficient zone is ideal for PAH and soot formation, it would lead to a low amount of O atoms in soot. The decrease in H content can be attributed to increased residence time in the flame with increasing fuel flow rate that provides sufficient time for soot dehydrogenation through the conversion of chain-like structures to rings [99]. While these phenomena lead to more aromatic structure of soot, it may be argued that they should also lead to more graphitic structure of soot. However, the role of temperature cannot be ignored here. With increasing fuel flow rate, a higher PAH production due to a higher amount of unburnt fuel lead to increased soot inception and growth through PAH condensation, but the drop in temperature in the sooting zone (due to radiative heat loss from soot) does not support sintering or restructuring of PAHs in soot [111] to lead to their graphitic structure, even with an increase in the residence time. Thus, while the soots produced at higher fuel flow rates have some characteristics of graphitic soot such as low H and O and high aromatic content, the absence of the restructuring of the PAHs and the presence of relatively small size of PAHs in soot (found from XRD analysis) leads to their amorphous character.

4. Conclusion To understand the variations in the morphology and chemical composition of soot particles formed during toluene combustion

with increasing fuel flow rate, soot particles were collected from the flames at different fuel flow rates. The soot samples were characterized using high-resolution transmission electron microscopy (HRTEM), X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), electron energy loss spectra (EELS), and elemental analysis. The HRTEM analysis indicated that, as the fuel flow rate is increased, larger soot aggregates with more number of primary particles are formed, and the randomness in the orientations of their fringes slightly increases, possibly due to decrease in the temperature in the sooting zone (arising from radiative heat loss) and due to increase in the residence time in flames that support particle coagulation and PAH condensation on soots. The XRD analysis showed negligible differences in the interlayer spacing and crystalline thickness, but the average size of PAHs present in soot was reduced with increasing fuel flow rate, which can result from an increased condensation of small PAHs on soot. The FTIR spectra showed that, though soots from different fuel flow rate had similar functional groups, the peak intensities for those functional groups differed, which indicates differences in the chemical composition of soot samples. The presence of aliphatic chains and oxygenated functional groups on soot surface was favored at low fuel flow rates, but with increasing flow rate, the aromatic character of soot particles was found to increase. The EELS analysis further confirmed the findings of high aliphatic content and low aromatic character al low fuel flow rates. The elemental analysis revealed that, while carbon content in soot increased, hydrogen and oxygen content in it decreased with increasing fuel flow rate. The experimental results suggest that the characteristics of soot produced from toluene in a diffusion flame differ, when the fuel flow rate is altered. The differences in those characteristics have been linked to the changes in flame temperature, residence time, and oxygen content. Further studies need to be conducted to examine if the physical and chemical properties of soot from different fuels change with fuel flow rate. References [1] Y. Yano, S.S. Morris, C. Salerno, A.M. Schlapia, M. Stichick, Impact of a new gasoline benzene regulation on ambient air pollutants in Anchorage, Alaska, Atmos. Environ. 132 (2016) 276–282. [2] W. Yinhui, Z. Rong, Q. Yanhong, P. Jianfei, L. Mengren, L. Jianrong, W. Yusheng, H. Min, S. Shijin, The impact of fuel compositions on the particulate emissions of direct injection gasoline engine, Fuel 166 (2016) 543–552. [3] W.J. Pitz, C.J. Mueller, Recent progress in the development of diesel surrogate fuels, Progr. Energy Combust. Sci. 37 (2011) 330–350. [4] A.L. Lown, L. Peereboom, S.A. Mueller, J.E. Anderson, D.J. Miller, C.T. Lira, Cold flow properties for blends of biofuels with diesel and jet fuels, Fuel 117 (Part A) (2014) 544–551.

G.D.J.G. Peña et al. / Combustion and Flame 178 (2017) 286–296 [5] C.T. Chong, S. Hochgreb, Spray flame structure of rapeseed biodiesel and Jet-A1 fuel, Fuel 115 (2014) 551–558. [6] G. Labeckas, S. Slavinskas, Combustion phenomenon, performance and emissions of a diesel engine with aviation turbine JP-8 fuel and rapeseed biodiesel blends, Energy Conversion Manag. 105 (2015) 216–229. [7] J. Yang, L. Zhao, W. Yuan, F. Qi, Y. Li, Experimental and kinetic modeling investigation on laminar premixed benzene flames with various equivalence ratios, Proc. Combust. Inst. 35 (2015) 855–862. [8] H. Anderson, C.S. McEnally, L.D. Pfefferle, Experimental study of naphthalene formation pathways in non-premixed methane flames doped with alkylbenzenes, Proc. Combust. Inst. 28 (20 0 0) 2577–2583. [9] W.L. Chameides, M. Bergin, Soot takes center stage, Science 297 (2002) 2214–2215. [10] V. Ramanathan, G. Carmichael, Global and regional climate changes due to black carbon, Nat. Geosci. 1 (2008) 221–227. [11] M. Keita, A. Nicolle, A.E. Bakali, A wide range kinetic modeling study of PAH formation from liquid transportation fuels combustion, Combust. Flame 174 (2016) 50–67. [12] Z.A. Mansurov, Soot formation in combustion processes (Review), Combust., Explos. Shock Waves 41 (2005) 727–744. [13] H.A. Michelsen, Probing soot formation, chemical and physical evolution, and oxidation: a review of in situ diagnostic techniques and needs, Proc. Combust. Inst. (2016). http://www.sciencedirect.com/science/article/pii/ S1540748916304163. [14] H. Ghiassi, I.C. Jaramillo, P. Toth, J.S. Lighty, Soot oxidation-induced fragmentation: part 2: experimental investigation of the mechanism of fragmentation, Combust. Flame 163 (2016) 170–178. [15] M. Sirignano, J. Kent, A. D’Anna, Further experimental and modelling evidences of soot fragmentation in flames, Proc. Combust. Inst. 35 (2015) 1779–1786. [16] Y. Li, L. Zhang, T. Yuan, K. Zhang, J. Yang, B. Yang, F. Qi, C.K. Law, Investigation on fuel-rich premixed flames of monocyclic aromatic hydrocarbons: part I. Intermediate identification and mass spectrometric analysis, Combust. Flame 157 (2010) 143–154. [17] B. Yang, Y. Li, L. Wei, C. Huang, J. Wang, Z. Tian, R. Yang, L. Sheng, Y. Zhang, F. Qi, An experimental study of the premixed benzene/oxygen/argon flame with tunable synchrotron photoionization, Proc. Combust. Inst. 31 (2007) 555–563. [18] Y. Li, L. Zhang, Z. Tian, T. Yuan, J. Wang, B. Yang, F. Qi, Experimental study of a fuel-rich premixed toluene flame at low pressure, Energy Fuels 23 (2009) 1473–1485. [19] V. Detilleux, J. Vandooren, Experimental and kinetic modeling investigation of toluene combustion in premixed, one-dimensional and laminar toluene-oxygen-argon flames, Proc. Combust. Inst. 33 (2011) 217–224. [20] A. D’Anna, A. Violi, Detailed modeling of the molecular growth process in aromatic and aliphatic premixed flames, Energy Fuels 19 (2005) 79–86. [21] Z. Tian, W.J. Pitz, R. Fournet, P.-A. Glaude, F. Battin-Leclerc, A detailed kinetic modeling study of toluene oxidation in a premixed laminar flame, Proc. Combust. Inst. 33 (2011) 233–241. [22] Y. Liu, C. Song, G. Lv, X. Cao, L. Wang, Y. Qiao, X. Yang, Surface functional groups and sp3/sp2 hybridization ratios of in-cylinder soot from a diesel engine fueled with n-heptane and n-heptane/toluene, Fuel 179 (2016) 108–113. [23] A.E. Daca, Ö.L. Gülder, Soot formation characteristics of diffusion flames of methane doped with toluene and n-heptane at elevated pressures, Proc. Combust. Inst. (2016). http://www.sciencedirect.com/science/article/pii/ S1540748916303042. [24] W. Yuan, Y. Li, P. Dagaut, J. Yang, F. Qi, Investigation on the pyrolysis and oxidation of toluene over a wide range conditions. II. A comprehensive kinetic modeling study, Combust. Flame 162 (2015) 22–40. [25] Y. Li, J. Cai, L. Zhang, T. Yuan, K. Zhang, F. Qi, Investigation on chemical structures of premixed toluene flames at low pressure, Proc. Combust. Inst. 33 (2011) 593–600. [26] D.A. Knyazkov, N.A. Slavinskaya, A.M. Dmitriev, A.G. Shmakov, O.P. Korobeinichev, U. Riedel, Structure of an n-heptane/toluene flame: Molecular beam mass spectrometry and computer simulation investigations, Combust., Explos. Shock Waves 52 (2016) 142–154. [27] W.K. Metcalfe, S. Dooley, F.L. Dryer, Comprehensive detailed chemical kinetic modeling study of toluene oxidation, Energy Fuels 25 (2011) 4915–4936. [28] S. Ito, Analysis of aromatic hydrocarbons in gasoline and naphtha with the agilent 6820 series gas chromatograph and a single polar capillary column, Yokogawa Analytical Systems, Inc., Tokyo, Japan, 2013, pp. 1–16. [29] L.E. Dunlap, Soluble hydrocarbons analysis from kerosene/diesel type hydrocarbons, Proceedings of the Conference on Petroleum Hydrocarbons and Organic Chemicals in Ground Water: Prevention, Detection and Restoration, Houston, TX, USA (1988), pp. 9–11. [30] A. Raj, I.D.C. Prada, A.A. Amer, S.H. Chung, A reaction mechanism for gasoline surrogate fuels for large polycyclic aromatic hydrocarbons, Combust. Flame 159 (2012) 500–515. [31] S. Dooley, S.H. Won, M. Chaos, J. Heyne, Y. Ju, F.L. Dryer, K. Kumar, C.-J. Sung, H. Wang, M.A. Oehlschlaeger, R.J. Santoro, T.A. Litzinger, A jet fuel surrogate formulated by real fuel properties, Combust. Flame 157 (2010) 2333–2339. [32] M.L. Botero, D. Chen, S. González-Calera, D. Jefferson, M. Kraft, HRTEM evaluation of soot particles produced by the non-premixed combustion of liquid fuels, Carbon 96 (2016) 459–473. [33] D. Uy, M.A. Ford, D.T. Jayne, A.E. O‫׳‬, L.P.Haack Neill, J. Hangas, M.J. Jagner, A. Sammut, A.K. Gangopadhyay, Characterization of gasoline soot and comparison to diesel soot: Morphology, chemistry, and wear, Tribol. Int. 80 (2014) 198–209.

295

[34] A. Santamaria, N. Yang, E. Eddings, F. Mondragon, Chemical and morphological characterization of soot and soot precursors generated in an inverse diffusion flame with aromatic and aliphatic fuels, Combust. Flame 157 (2010) 33–42. [35] Y. Jung, C. Bae, Immaturity of soot particles in exhaust gas for low temperature diesel combustion in a direct injection compression ignition engine, Fuel 161 (2015) 312–322. [36] A. La Rocca, F. Bonatesta, M.W. Fay, F. Campanella, Characterisation of soot in oil from a gasoline direct injection engine using transmission electron microscopy, Tribol. Int. 86 (2015) 77–84. [37] S. di Stasio, A.G. Konstandopoulos, M. Kostoglou, Cluster–Cluster aggregation kinetics and primary particle growth of soot nanoparticles in flame by light scattering and numerical simulations, J. Colloid Interface Sci. 247 (2002) 33–46. [38] M.L. Botero, E.M. Adkins, S. González-Calera, H. Miller, M. Kraft, PAH structure analysis of soot in a non-premixed flame using high-resolution transmission electron microscopy and optical band gap analysis, Combust. Flame 164 (2016) 250–258. [39] C. Russo, M. Alfè, J.-N. Rouzaud, F. Stanzione, A. Tregrossi, A. Ciajolo, Probing structures of soot formed in premixed flames of methane, ethylene and benzene, Proc. Combust. Inst. 34 (2013) 1885–1892. [40] Y. Kobayashi, T. Furuhata, K. Amagai, M. Arai, Soot precursor measurements in benzene and hexane diffusion flames, Combust. Flame 154 (2008) 346–355. [41] P. Minutolo, M. Commodo, M. Sirignano, A. D’Anna, Experimental and modeling study on soot inception in rich premixed aliphatic and aromatic flames, Proceedings of 4th European Combustion Meeting ECM, Vienna, Austria (2009), pp. 14–17. [42] M. Velásquez, F. Mondragón, A. Santamaría, Chemical characterization of soot precursors and soot particles produced in hexane and diesel surrogates using an inverse diffusion flame burner, Fuel 104 (2013) 681–690. [43] D. Witkowski, K. Kondo, G. Vishwanathan, D. Rothamer, Evaluation of the sooting properties of real fuels and their commonly used surrogates in a laminar co-flow diffusion flame, Combust. Flame 160 (2013) 1129–1141. [44] M.L. Botero, S. Mosbach, J. Akroyd, M. Kraft, Sooting tendency of surrogates for the aromatic fractions of diesel and gasoline in a wick-fed diffusion flame, Fuel 153 (2015) 31–39. [45] A. D’Anna, M. Alfè, B. Apicella, A. Tregrossi, A. Ciajolo, Effect of fuel/air ratio and aromaticity on sooting behavior of premixed heptane flames, Energy Fuels 21 (2007) 2655–2662. [46] J. Wei, C. Song, G. Lv, J. Song, L. Wang, H. Pang, A comparative study of the physical properties of in-cylinder soot generated from the combustion of n-heptane and toluene/n-heptane in a diesel engine, Proc. Combust. Inst. 35 (2015) 1939–1946. [47] M. Alfè, B. Apicella, R. Barbella, A. Tregrossi, A. Ciajolo, Similarities and dissimilarities in n-hexane and benzene sooting premixed flames, Proc. Combust. Inst. 31 (2007) 585–591. [48] R.L. Vander Wal, A.J. Tomasek, Soot nanostructure: dependence upon synthesis conditions, Combust. Flame 136 (2004) 129–140. [49] R.J. Santoro, H.G. Semerjian, Soot formation in diffusion flames: flow rate, fuel species and temperature effects, Symp. (Int.) Combust. 20 (1985) 997–1006. [50] M.L. Botero, S. Mosbach, M. Kraft, Sooting tendency and particle size distributions of n-heptane/toluene mixtures burned in a wick-fed diffusion flame, Fuel 169 (2016) 111–119. [51] M.L. Botero, S. Mosbach, M. Kraft, Sooting tendency of paraffin components of diesel and gasoline in diffusion flames, Fuel 126 (2014) 8–15. [52] M.M. Constantine, A.D. Richard, Comparison of soot growth and oxidation in smoking and non-smoking ethylene diffusion flames, Combust. Sci. Technol. 66 (1989) 1–16. [53] M. Alfè, B. Apicella, R. Barbella, J.N. Rouzaud, A. Tregrossi, A. Ciajolo, Structure–property relationship in nanostructures of young and mature soot in premixed flames, Proc. Combust. Inst. 32 (2009) 697–704. [54] Y. Zhang, R. Zhang, S. Kook, Nanostructure analysis of in-flame soot particles under the influence of jet-jet interactions in a light-duty diesel engine, SAE Int. J. Engines 8 (2015) 2213–2226. [55] G.D.J. Guerrero Peña, M.M. Alrefaai, S.Y. Yang, A. Raj, J.L. Brito, S. Stephen, T. Anjana, V. Pillai, A. Al Shoaibi, S.H. Chung, Effects of methyl group on aromatic hydrocarbons on the nanostructures and oxidative reactivity of combustion-generated soot, Combust. Flame 172 (2016) 1–12. [56] X. Wang, C. Song, G. Lv, J. Song, H. Li, B. Li, Evolution of in-cylinder polycyclic aromatic hydrocarbons in a diesel engine fueled with n-heptane and n-heptane/toluene, Fuel 158 (2015) 322–329. [57] L. Sileghem, V.A. Alekseev, J. Vancoillie, K.M. Van Geem, E.J.K. Nilsson, S. Verhelst, A.A. Konnov, Laminar burning velocity of gasoline and the gasoline surrogate components iso-octane, n-heptane and toluene, Fuel 112 (2013) 355–365. [58] D. Han, S. Deng, W. Liang, P. Zhao, F. Wu, Z. Huang, C.K. Law, Laminar flame propagation and nonpremixed stagnation ignition of toluene and xylenes, Proc. Combust. Inst. (2016). http://www.sciencedirect.com/science/article/pii/ S154074891630013X. [59] J.H. kent, H.G. Wagner, Who do diffusion flames emit smoke, Combust. Sci. Technol. 41 (1984) 245–269. [60] A. Raj, M. Sander, V. Janardhanan, M. Kraft, A study on the coagulation of polycyclic aromatic hydrocarbon clusters to determine their collision efficiency, Combust. Flame 157 (2010) 523–534. [61] M. Sirignano, H. Ghiassi, A. D’Anna, J.S. Lighty, Temperature and oxygen effects on oxidation-induced fragmentation of soot particles, Combust. Flame 171 (2016) 15–26.

296

G.D.J.G. Peña et al. / Combustion and Flame 178 (2017) 286–296

[62] G.M. Faeth, Ü.Ö. KÖYlÜ, Soot morphology and optical properties in nonpremixed turbulent flame environments, Combust. Sci. Technol. 108 (1995) 207–229. [63] C. Han, Y. Liu, C. Liu, J. Ma, H. He, Influence of combustion conditions on hydrophilic properties and microstructure of flame soot, J. Phys. Chem. A 116 (2012) 4129–4136. [64] B. Apicella, M. Alfè, A. Ciajolo, V. Gargiulo, P. Pré, J.N. Rouzaud, Soot structure by high resolution electron transmission microscopy (HR-TEM) image analysis: current potentialities and perspectives, XXXVI Meeting of the Italian Section of the Combustion Institute, Italy Combustion Institute, Procida (NA) (2013), pp. 13–15. [65] B. Apicella, P. Pré, M. Alfè, A. Ciajolo, V. Gargiulo, C. Russo, A. Tregrossi, D. Deldique, J.N. Rouzaud, Soot nanostructure evolution in premixed flames by high resolution electron transmission microscopy (HRTEM), Proc. Combust. Inst. 35 (2015) 1895–1902. [66] D. Pantea, S. Brochu, S. Thiboutot, G. Ampleman, G. Scholz, A morphological investigation of soot produced by the detonation of munitions, Chemosphere 65 (2006) 821–831. [67] A. Raj, R. Tayouo, D. Cha, L. Li, M.A. Ismail, S.H. Chung, Thermal fragmentation and deactivation of combustion-generated soot particles, Combust. Flame 161 (2014) 2446–2457. [68] L. Frusteri, C. Cannilla, K. Barbera, S. Perathoner, G. Centi, F. Frusteri, Carbon growth evidences as a result of benzene pyrolysis, Carbon 59 (2013) 296–307. [69] W.J. Grieco, J.B. Howard, L.C. Rainey, J.B. Vander Sande, Fullerenic carbon in combustion-generated soot, Carbon 38 (20 0 0) 597–614. [70] C.R. Shaddix, Á.B. Palotás, C.M. Megaridis, M.Y. Choi, N.Y.C. Yang, Soot graphitic order in laminar diffusion flames and a large-scale JP-8 pool fire, Int. J. Heat Mass Transf. 48 (2005) 3604–3614. [71] A. Oberlin, Application of dark-field electron microscopy to carbon study, Carbon 17 (1979) 7–20. [72] B. Manoj, A.G. Kunjomana, Study of stacking structure of amorphous carbon by x-ray diffraction technique, Int. J. Electrochem. Sci. 7 (2012) 3127–3134. [73] B. Gogoi, A. Raj, M.M. Alrefaai, S. Stephen, T. Anjana, V. Pillai, S. Bojanampati, Effects of 2,5-dimethylfuran addition to diesel on soot nanostructures and reactivity, Fuel 159 (2015) 766–775. [74] H.X. Chen, R.A. Dobbins, Crystallogenesis of particles formed in hydrocarbon combustion, Combust. Sci. Technol. 159 (20 0 0) 109–128. [75] K. Yehliu, R.L. Vander Wal, O. Armas, A.L. Boehman, Impact of fuel formulation on the nanostructure and reactivity of diesel soot, Combust. Flame 159 (2012) 3597–3606. [76] K. Al-Qurashi, A.L. Boehman, Impact of exhaust gas recirculation (EGR) on the oxidative reactivity of diesel engine soot, Combust. Flame 155 (2008) 675–695. [77] M.A. Short, P.L. Walker, Measurement of interlayer spacings and crystal sizes in turbostratic carbons, Carbon 1 (1963) 3–9. [78] C. Shaddix, T. Williams, Soot structure and dimensionless extinction coefficient in diffusion flames: implications for index of refraction, combustion generated fine carbonaceous particles, KIT Scientific Publishing, Karlsruhe, Germany, 2009, pp. 17–33. [79] H.J. Seong, A.L. Boehman, Impact of intake oxygen enrichment on oxidative reactivity and properties of diesel soot, Energy Fuels 25 (2011) 602–616. [80] Anu.N. Mohan, B. Manoj, Synthesis and characterization of carbon nanospheres from hydrocarbon soot, Int. J. Electrochem. Sci. 7 (2012) 9537–9549. [81] C.R. Houska, B.E. Warren, X-ray study of the graphitization of carbon black, J. Appl. Phys. 25 (1954) 1503–1509. [82] E.A. Belenkov, Formation of graphite structure in carbon crystallites, Inorgan. Mater. 37 (2001) 928–934. [83] K. Alexandrino, J. Salinas, Á. Millera, R. Bilbao, M.U. Alzueta, Sooting propensity of dimethyl carbonate, soot reactivity and characterization, Fuel 183 (2016) 64–72. [84] C.-H. Huang, R.L. Vander Wal, Partial premixing effects upon soot nanostructure, Combust. Flame 168 (2016) 403–408. [85] A. Sharma, T. Kyotani, A. Tomita, Comparison of structural parameters of PF carbon from XRD and HRTEM techniques, Carbon 38 (20 0 0) 1977–1984. [86] B.H. Stuart, Organic molecules, infrared spectroscopy: fundamentals and applications, John Wiley & Sons, Ltd, 2005, pp. 71–93. [87] J.M. O’Reilly, R.A. Mosher, Functional groups in carbon black by FTIR spectroscopy, Carbon 21 (1983) 47–51.

[88] L. Wang, C. Song, J. Song, G. Lv, H. Pang, W. Zhang, Aliphatic C–H and oxygenated surface functional groups of diesel in-cylinder soot: characterizations and impact on soot oxidation behavior, Proc. Combust. Inst. 34 (2013) 3099–3106. [89] E. Zervas, Formation of organic acids from propane, isooctane and toluene/isooctane flames, Fuel 84 (2005) 691–700. [90] H. Muckenhuber, H. Grothe, The heterogeneous reaction between soot and NO2 at elevated temperature, Carbon 44 (2006) 546–559. [91] O.B. Popovicheva, N.M. Persiantseva, V. Tishkova, N.K. Shonija, N.A. Zubareva, Quantification of water uptake by soot particles, Environ. Res. Lett. 3 (2008) 025009. [92] V.M. Malhotra, S. Jasty, R. Mu, FT-IR spectra of water in microporous KBr pellets and water’s desorption kinetics, Appl. Spectrosc. 43 (1989) 638–645. [93] C. Alcala-Jornod, M.J. Rossi, Chemical kinetics of the interaction of H2 O vapor with soot in the range 190 K ≤ T ≤ 300 K: a diffusion tube study, J. Phys. Chem. A 108 (2004) 10667–10680. [94] A.S. Scott, P.B. Max, K.M. Christopher, The infrared spectra of polycyclic aromatic hydrocarbons with excess peripheral H atoms (H n -PAHs) and their relation to the 3.4 and 6.9 μm PAH emission features, Astrophys. J. Suppl. Ser. 205 (2013) 8. [95] H.M. Daly, A.B. Horn, Heterogeneous chemistry of toluene, kerosene and diesel soots, Phys. Chem. Chem. Phys. 11 (2009) 1069–1076. [96] J.T. McKinnon, E. Meyer, J.B. Howard, Infrared analysis of flame-generated PAH samples, Combust. Flame 105 (1996) 161–166. [97] JeremyP. Cain, PaulL. Gassman, Hai Wang, A. Laskin, Micro-FTIR study of soot chemical composition—evidence of aliphatic hydrocarbons on nascent soot surfaces, Phys. Chem. Chem. Phys. 2012 (2010) 5206–5218. [98] A. D’Anna, Combustion-formed nanoparticles, Proc. Combust. Inst. 32 (2009) 593–613. [99] A. Raj, P.L.W. Man, T.S. Totton, M. Sander, R.A. Shirley, M. Kraft, New polycyclic aromatic hydrocarbon (PAH) surface processes to improve the model prediction of the composition of combustion-generated PAHs and soot, Carbon 48 (2010) 319–332. [100] M.S. Akhter, J.R. Keifer, A.R. Chughtai, D.M. Smith, The absorption band at 1590 cm−1 in the infrared spectrum of carbons, Carbon 23 (1985) 589–591. [101] U.J. Kim, C.A. Furtado, X. Liu, G. Chen, P.C. Eklund, Raman and IR spectroscopy of chemically processed single-walled carbon nanotubes, J. Am. Chem. Soc. 127 (2005) 15437–15445. [102] J.A. Jassim, H.P. Lu, A.R. Chughtai, D.M. Smith, The origin of infrared absorption bands of soot in the region 750 to 900 cm−1 , Appl. Spectrosc. 40 (1986) 113–116. [103] T.F. Yen, W.H. Wu, G.V. Chilingar, A study of the structure of petroleum asphaltenes and related substances by infrared spectroscopy, Energy Sources 7 (1984) 203–235. [104] C.W. Bauschlicher Jr., The infrared spectra of nonplanar polycyclic aromatic hydrocarbons with five- or seven-membered rings, Chem. Phys. 448 (2015) 43–52. [105] K.A. Katrinak, P. Rez, P.R. Buseck, Structural variations in individual carbonaceous particles from an urban aerosol, Environ. Sci. Technol. 26 (1992) 1967–1976. [106] C. Jäger, T. Henning, R. Schlögl, O. Spillecke, Spectral properties of carbon black, J. Non-Cryst. Solids 258 (1999) 161–179. [107] C. Jager, H. Mutschke, T. Henning, Optical properties of carbonaceous dust analogues, Astron. Astrophys. 332 (1998) 291–299. [108] A. Braun, F.E. Huggins, N. Shah, Y. Chen, S. Wirick, S.B. Mun, C. Jacobsen, G.P. Huffman, Advantages of soft X-ray absorption over TEM-EELS for solid carbon studies—-a comparative study on diesel soot with EELS and NEXAFS, Carbon 43 (2005) 117–124. [109] S. di Stasio, A. Braun, Comparative NEXAFS study on soot obtained from an ethylene/air flame, a diesel engine, and graphite, Energy Fuels 20 (2006) 187–194. [110] L.P. Eksperiandova, O.I. Fedorov, N.A. Stepanenko, Estimation of metrological characteristics of the element analyzer EuroVector EA-30 0 0 and its potential in the single-reactor CHNS mode, Microchem. J. 99 (2011) 235–238. [111] K. Ono, K. Dewa, Y. Matsukawa, Y. Saito, Y. Matsushita, H. Aoki, K. Era, T. Aoki, T. Yamaguchi, Experimental evidence for the sintering of primary soot particles, J. Aerosol Sci. 105 (2017) 1–9.