Effect Air-fuel Combustion Ratio On Pah Soot

Carbon 42 (2004) 2471–2484 www.elsevier.com/locate/carbon

Effects of air/fuel combustion ratio on the polycyclic aromatic hydrocarbon content of carbonaceous soots from selected fuels C.C. Jones, A.R. Chughtai, B. Murugaverl, D.M. Smith

*

Department of Chemistry and Biochemistry, University of Denver, Denver, CO 80208, USA Received 26 February 2004; accepted 30 April 2004 Available online 15 July 2004

Abstract Patterns of polycyclic aromatic hydrocarbon (PAH) content were observed from GC/MS analysis of the extracts of soots at various air/fuel combustion ratios of three commonly used fuels: n-hexane, JP-8 (Jet fuel), and diesel. With increasing air/fuel ratio, from a simple diffusion flame up to an air/fuel ratio of 3.94, there is a significant loss of high molecular weight PAHs and an increasing abundance of oxidized lower molecular weight aromatics. The formation of high molecular weight PAHs is favored for JP-8 and diesel fuels at higher air/fuel combustion ratios than is the case with n-hexane, probably due to the aromatic content in JP-8 and diesel fuels acting as centers for large aromatic and soot nucleation. The efficiency and reproducibility of two techniques, Soxhlet and supercritical fluid extraction (SFE), used for extraction of PAHs from soot were compared. Electron paramagnetic resonance (EPR) measurements were performed on the soot both before and after supercritical fluid and Soxhlet extraction, and a substantial decrease in the spin density of soot following extraction indicates that extractable molecules are associated with 40–50% of the unpaired electrons in soot. This analysis generally supports trends observed in our earlier work for surface oxidation, surface area, unpaired electron spin density, hydration, and ozone oxidation.
2004 Published by Elsevier Ltd. Keywords: A. Soot; B. Combustion; C. Chromatography; D. Chemical structure

1. Introduction The incomplete combustion of hydrocarbon fuels has been known to produce carcinogenic constituents such as polycyclic aromatic hydrocarbons (PAH). A particularly good overview of nomenclature, roles in tropospheric chemistry, and mutagenic and carcinogenic effects of polycyclic aromatic hydrocarbons may be found in the monograph on atmospheric chemistry by Finlayson-Pitts and Pitts [1]. Many of these compounds are embedded within the pore structure of soot (black carbon) or adsorbed on its surface from gas phase formation [2]. Soot is a randomly formed particulate carbon that, in addition to carbon atoms arranged in an aromatic framework, contains a large variety of organic components to make up an extremely complex system

*

Corresponding author. Fax: +1-303-871-2932. E-mail address: [email protected] (D.M. Smith).

0008-6223/$ - see front matter
2004 Published by Elsevier Ltd. doi:10.1016/j.carbon.2004.04.042

[2]. The annual global emissions of black carbon have been estimated as 12–24 Tg/yr [3,4]. Black carbon, because of its chemical and physical properties, has impacts on the earth’s radiation budget [5–9], human mortality [10–12], and tropospheric chemistry, e.g. [13– 21]. In this analysis we sought to determine the types of PAHs in soot produced by three different fuels at various air/fuel (A/F) ratios. A comparative study on two common extraction methods, traditional Soxhlet and supercritical fluid extraction, is outlined. The Soxhlet technique has been widely used in the extraction of polyaromatic hydrocarbons. We employed Soxhlet extraction in a previous analysis of PAH compounds in n-hexane soot produced by diffusion flame, in which sequential extraction with a series of solvents possessing a range of polarity was carried out [2]. Recovery studies conducted on diesel particulate have shown significant differences between Soxhlet and supercritical fluid extraction [22–26].

2472

C.C. Jones et al. / Carbon 42 (2004) 2471–2484

There is reason to believe that the unpaired electron spin density of black carbon contributes to both its atmospheric reactivity and human health effects. Therefore, it was of interest to determine any association of unpaired electrons with the PAH content. Details of this association are described below. This approach will provide a basis for understanding at which points of oxidation PAHs begin to form and then disappear during combustion processes. With such knowledge, one may begin to estimate the effects of A/F combustion ratios of fuels in engines and in determining combustion parameters to reduce human and environmental exposure to anthropogenic particulate pollution.

prior to burning and the diffusion of oxygen to the flame is over a larger area.

2. Experimental

2.1.2. JP-8 soot Jet fuel (density 0.8086 g/ml) has a composition of about 68% kerosene, which has hydrocarbon chains in the C10 –C11 region. The JP-8 also contains a maximum of 22% aromatics by volume, 3.0% naphthalenes by volume, and a maximum of 0.3% sulfur by weight as well as added approved antioxidants (such as 2,6-ditert-butyl-4-methylphenol and other tert-butylphenols), metal deactivators (N ; N 0 -disalicylidene-1,2-propanediamine), and corrosion inhibitors. JP-8 was supplied by the Continental Oil Company (Denver, Colorado). The same five soots were prepared as for n-hexane with specific A/F ratios, however, of 0.000, 0.184, 0.867, 1.296, and 3.950.

2.1. Preparation of soots Three common liquid hydrocarbon fuels were used to produce the various soot samples in an effort to determine if there was any variation between them in regards to the formation of PAH as a function of A/F ratio. The fuels included reagent n-hexane (Pharmco, ACS), diesel (Total), and JP-8 (Continental Oil Company). Carbonaceous soots of each fuel were collected at air/fuel ratios that we describe in the format (0,X ) 1, (3,X ) 1, (6,X ) 1, (9,X ) 1, and (11,X ) 1 using a premixed flame combustion apparatus [20]. The preparation of the soot samples was carried out in a special fume hood constructed in house. The fuel was introduced by a peristaltic pump delivering the fuel at a specified flow rate into a chamber, which was preheated to 350 C. The gas then was ignited by a small pilot light. Air 2 was introduced directly into the mixing chamber prior to reaching the burner, which allowed thorough mixing before the ignition, as necessary for establishing the specific A/F ratios. Soot produced from the flame was collected in a bell jar measuring 44 cm in length and 15 cm in diameter with a 1.5 cm diameter hole at the top to release combustion gases. A simple diffusion flame burn also was conducted for the n-hexane fuel and the resulting soot denoted as regular. This is prepared by burning 100 ml of the desired fuel in a beaker under the same bell jar and fume hood. This combustion process differs from the (0,X ) combustion in that the fuel is not vaporized 1 The numerical values of (0,X ), (3,X ), (6,X ), (9,X ), and (11,X ) are determined by the volume of air added in ml/min, which are 0 ¼ 0 ml/ min, 3 ¼ 98 ml/min, 6 ¼ 362 ml/min, 9 ¼ 685 ml/min, and 11 ¼ 998 ml/ min, where X refers to the flow of fuel controlled by the peristaltic pump. (0,X ) is essentially a diffusion rather than a premix flame. Corresponding numerical values of A/F are specific for each fuel combustion. 2 Air was supplied by a laboratory compressed air system. The air was run through two silica gel scrubbers to remove moisture and oils prior to the flow meter and air–fuel mixing chamber.

2.1.1. n-Hexane soot n-Hexane (density 0.6548 g/ml), used to prepare a reference standard soot, was introduced into the burner at a flow rate of 0.96 ml/min, which is the optimum flow for maintaining a controlled sooting flame. Five A/F ratios described as (0,X ), (3,X ), (6,X ), (9,X ) and (11,X ) were used having numerical A/F ratios of 0 (diffusion flame), 0.112, 0.527, 1.072, and 2.397, respectively. Regular n-hexane soot was prepared as described above.

2.1.3. Diesel soot Diesel fuel (density 0.8091 g/ml) is composed of hydrocarbon chains larger than C12 –C14 and contains a small percentage of sulfur as well as 5–30% aromatic. The combustion of diesel fuel (supplied by Total Petroleum, Denver) was controlled in the laboratory at a flow of 0.48 ml/min, which is the optimum flow for a reproducible sooting flame. The five specific A/F ratios for the diesel soot were 0.000, 0.184, 0.866, 1.761, and 3.936.

2.2. Soxhlet extraction All Soxhlet extractions were performed in duplicate for 20 h using 200 ml of dichloromethane and a 500 mg soot sample. The cycle time was approximately 15 min. After Soxhlet extraction the samples were first reduced to 5 ml by simple distillation, followed by filtration through glass microfibre filters (Whatman 934-AH) and 2 · 10 ml dichloromethane solvent rinses. The sample then was reduced to 0.1 ml (determined by a set caliper for reproducibility) by a gentle stream of nitrogen in a 2 ml amber auto sample vial. The sample was immediately analyzed by GC/MS using an HP 6890.

C.C. Jones et al. / Carbon 42 (2004) 2471–2484

2.3. Supercritical fluid extractions (SFE) All extractions were performed using a Durability Model MPS/225 supercritical extraction/chromatography system. Extractions were carried out in duplicate on soot samples weighing 300 mg loaded in 4.5-mmID · 250-mm-long stainless steel extraction cells with 0.2 lm pore size end frits. All extractions were at 200 C with supercritical CO2 modified with a 50/50 mixture of dichloromethane and toluene as the extraction fluid. The modifier was added to the high pressure CO2 at 0.1 ml/ min using an HPLC pump (ESA) during the extraction. The CO2 was pressure programmed, starting at 150 atm for 5 min in a static mode followed by ramping the pressure to 400 atm over 15 min in dynamic mode and held at 400 atm for 5 min in static mode. The pressure then was ramped down to 100 atm over 10 min in dynamic mode. The outlet of the extraction cell was connected to a manual high pressure needle valve to control the static and dynamic extraction steps by simply closing the outlet valve during the static mode and opening during the dynamic mode. The outlet valve also was used to regulate the flow rate along with a 0.01-in.ID · 30-cm-long stainless steel tubing that is crimped at the exit end. Extracted analytes were collected by allowing the supercritical fluid with the extracted analytes to decompress into a vial containing 2 ml of dichloromethane. The extract then was reduced to 0.1 ml using a gentle stream of nitrogen gas before analysis.

2.4. GC/MS analysis All soot extracts were performed using a HewlettPackard Model 6890 GC/MSD equipped with a 30 m long Alltech EC-5 fused silica capillary column (0.250 lm ID, 5% phenylmethylpolysiloxane stationary phase, 25 lm film thickness). Samples were injected (0.1 ll) in split mode (1:10), with the injection port at 290 C and an initial oven temperature of 150 C for 2 min. The oven temperature was ramped to 250 C at 50 C/min, followed by a 10 C/min ramp rate to 300 C and held at 300 C for 7 min. Helium was used as the carrier gas at a 1.0 ml/min flow rate. The total ion chromatograms were analyzed qualitatively with the aid of the NIST 2000 mass spectral library after background subtraction. Chromatographic peaks with mass spectral match of 90% or greater were considered as possible components in the soot extracts, while peaks showing significant abundance but less than 90% spectral match were noted by their retention times. These criteria were used in identifying the relationships between various fuels and air/fuel ratios. Twelve compounds were identified by standards which included 9H-fluoren-9-one, anthracene, phenanthrene, pyrene, benz[a]anthracene, triphenylene, chrysene, 7H-benz[de]anthracen-7-one, benzo[k]fluo-

2473

ranthene, benzo[a]pyrene, indeno[1,2,3-cd]pyrene, and benzo[ghi]perylene. 2.5. Electron paramagnetic resonance (EPR) measurements EPR spectra of soot samples were obtained with a Varian E-9 EPR spectrometer with an E-102 microwave bridge. Quartz EPR tubes (Wilmad Glass) were matched (4.0 ± 0.1 mm OD) and soot samples of exact weight (±0.1 mg) added. Each sample was subjected to evacuation to 3 · 102 Pa at 100 C for 20 h, through a specially designed donut-shaped manifold enclosed in a Glas-Col heating mantle, prior to flame cutting and analysis. A sample standard deviation of the EPR signal’s corrected normalized integral (CNI) of within ±2% was achieved in these replicate measurements.

3. Results and discussion 3.1. n-Hexane soots The total ion chromatograms of n-hexane soots (3,X ) and (11,X ) using SFE can be seen in Figs. 1 and 2. (For all sets of experiments, only these two chromatograms will be shown for comparison in order to conserve space. In every case, replicate experiments were run with regular, (0,X ), (3,X ), (6,X ), (9,X ), and (11,X ) soots and the reproducible trends observed as discussed.) As the air/ fuel ratio is increased, there is a reduction in and eventual loss of peaks with retention times >6.35 min. These peaks represent the higher molecular weight (HMW) (and carcinogenic) PAHs such as benzo(ghi)fluoranthene (7.62 min), benz(a)anthracene (7.97 min), chrysene (8.03 min), benzo(k)fluoranthene (9.73 min), benzo(a)pyrene (10.33 min), indeno(1,2,3-cd)perylene (13.02 min), indeno(1,2,3-cd)fluoranthene (13.15 min), and benzo(ghi)perylene (13.91 min). This trend is summarized in Fig. 3; that is, at higher A/F ratios, (6,X )– (11,X ), the percentage area for the HMW PAHs decreases. The soot extracts contain oxidized and unoxidized PAHs formed in the soot. Oxidized PAHs such as 9Hfluorenone, 1,2-acenaphthylenedione, 1H-phenalen-1one, 9,10-anthracenedione, 1,8-naphthalic anhydride, also have been seen in air particulate studies [23–27]. However, the source for the observed oxygenated PAHs could be due to atmospheric reactions [28]. There is an increase in the relative abundance for the lower molecular weight (LMW) oxidized PAHs such as 9H-fluorenone, 1,2-acenaphthylenedione, and 1,8-naphthalic anhydride with increasing A/F ratio, and a decrease in relative abundance of the HMW oxidized PAHs such as 7H-benz[de]anthracen-7-one, and cyclopenta[def]phenanthrenone. This likely is due to the relative availability

2474

C.C. Jones et al. / Carbon 42 (2004) 2471–2484 Abundance 300000

TIC: [BSB1]533CJ.D

6.04 3.10

9.73 10.37

4.34 4.44 4.86

280000 260000

8.17

240000

12.97

9.44

220000 200000 180000

10.20 7.637.95

160000 140000

5.24

120000

4.54

100000

4.51

7.89 7.70 7.41

5.38

80000 60000

6.36 6.11

5.02

3.72

9.58 8.44 8.66

11.15

9.29

13.15 12.84

11.25

7.58 10.11

40000

10.97 11.06

13.93 12.01

20000 0 4.00

Retention

5.00

6.00

7.00

8.00

9.00

10.00

11.00

12.00

13.00

14.00

15.00

Time-->

Fig. 1. Total ion chromatogram of the SFE extract of n-hexane soot (3,X ).

Abundance

TIC: [BSB1]540CJ.D 4.32 4.82

280000

6.08

4.99

260000 6.31

240000 220000 200000 5.46 5.35

180000 160000 140000

5.27

5.69

120000 100000

4.93 5.16 6.48 4.72 5.21 6.50 6.73 4.50 5.39 5.65 6.13 6.53 7.597.89 5.50 5.06 5.56 5.74 5.93 7.54 7.08 8.11 5.826.22 6.63 7.15 7.65

80000 60000 3.11 40000 20000 0

Retention

4.00

5.00

6.00

7.00

8.00

9.00

10.00

11.00

12.00

13.00

14.00

15.00

Time-->

Fig. 2. Total ion chromatogram of the SFE extract of n-hexane soot (11,X ).

of oxygen and thus the flame temperature. At higher A/F ratios, a larger amount of oxygen is available to oxidize the developing aromatic compounds but, since the size of the aromatic is limited by the combustion temperature, this would result in increased oxidized LMW-PAHs and reduced abundance of HMW oxidized species. These are the soot formation conditions under which much of the earlier characterization work was done [29–32]. Recent studies by Fialkov et al. [33], using molecular beam sampling combined with reflectron time-of-flight mass spectrometry of a benzene–oxygen flame, observed that the H-rich hydrocarbons and oxohydrocarbons with masses >200 amu were found in the

early portion of fuel-rich flames, but no systematic description as to why this occurs could be given. Fialkov proposes that the back diffusion of hydrogen atoms into a mixture of benzene and oxygen can initiate hydrogenation of benzene to cyclohexadiene and cyclohexene followed by ring fragmentation to small aliphatic hydrocarbons. With the combination of low-temperature reactions of hydrogenation, partial oxidation, and polymerization of benzene, cyclohexadiene results in the production of a vast number of intermediate products ultimately forming oxidized aromatics. The unoxidized PAHs range from the highest molecular weight as benzo[ghi]perylene down to

C.C. Jones et al. / Carbon 42 (2004) 2471–2484

2475

80.00 70.00 60.00 50.00 % Area 40.00 30.00 20.00 (3,X) (6,X) (9,X) (11,X)

10. 00 0. 00 1

2

3

4

5

6

7

8

9

10

11

12

13

14

Compound #

(11,X) 15

16

17

(9,X) 18

19

20

(6,X) 21

(3,X)

22

23

24

Air/fuel Ratio

25

Fig. 3. PAH content versus air/fuel ratio of n-hexane combustion soots for SFE extractions; compound number begins with 1 as phthalic anhydride moving up to 26 as benzo[ghi]perylene; refer to Table 1 for corresponding compound #s.

It was of interest to compare these results with those from a commonly used low turbulent diffusion flame denoted as regular in which the oxidation is from diffusion of ambient air. This flame is a low oxidizing diffusion flame similar to the (0,X ) flame; however, the fuel is not heated prior to ignition and the combustion surface area is much greater than the fine burner tip for the premixed flames, resulting in diverse oxidation environments of the peripheral and internal regions of the flame. The PAH content shown in Fig. 4 is similar to

anthracene, but the abundance of these compounds shifts from high molecular weight PAHs (>226 MW) for the (0,X ) and (3,X ) soots (lower temperatures, low air/ fuel ratio) to lower molecular weight PAHs (<226 MW) with the (6,X ), (9,X ), and (11,X ) materials. Again, this can be explained by the flame temperature and A/F ratio in which alkyl radical and acetylene abundance begins to drop off as a result of increased combustion to CO2 and H2 O, and it seems this is the definitive factor for the limited size of PAH. Abundance

TIC: [BSB1]543CJ.D

1500000

4.064.38

6.13

9.82

10.50

1400000 1300000 1200000 1100000 1000000 900000

13.28

10.30

8.25

800000 13.41

700000

14.27 600000 6.19

500000

5.08

400000 300000 3.12

5.56

7.718.03 7.97

6.43

200000 4.01 100000

3.32

4.92

7.78 7.66 7.48

5.47 5.31

8.67 8.74

9.57

9.94

8.558.82 9.49 8.93 9.38

11.37 11.61 10.73 11.30 11.15

12.97

10.57

13.56 13.66

12.24

14.72 15.27 15.59

0

Retention

4.00

5.00

6.00

7.00

8.00

9.00

10.00

11.00

12.00

13.00

Time-->

Fig. 4. Total ion chromatogram of the SFE extract of n-hexane soot (regular).

14.00

15.00

2476

C.C. Jones et al. / Carbon 42 (2004) 2471–2484 Abundance 400000

TIC: [BSB1]333CJ.D

5.43 5.41

10.27

13.07

13.9414.38

350000

300000

250000 7.61 200000 7.87

9.63

150000

100000 4.17

5.14 4.98 5.10 4.81 5.48 5.01 4.94

50000

10.11

11.35 12.71

8.11

6.31

7.57 8.05

5.91

9.18 8.96 8.74

9.80

10.39

11.57

13.67 13.46 13.30

11.17

14.12

12.45

3.32 0

Retention

4.00

5.00

6.00

7.00

8.00

9.00

10.00

11.00

12.00

13.00

14.00

15.00

Time-->

Fig. 5. Total ion chromatogram of the SOX extract of n-hexane soot (3,X ).

that of (0,X ), but more diverse in LMW-PAHs, oxyPAHs and HMW-PAHs due to the varying oxidation regions. Soxhlet extractions (SOX) of n-hexane have proven to be difficult to reproduce, with random abundance of peaks and solvent contamination (because of the extract concentration required) as major problems; for example, the Soxhlet blank contains dibutylphthalate from the ACS grade solvent. Perhaps the molecular size [34] of dichloromethane (DCM) is large enough to inhibit its ability to penetrate deep into the soot pore structure and extract the target analytes. Therefore, analytes exposed on the soot surface that are more readily available for extraction affect overall peak abundance from sample to sample. On the other hand, it may simply be that the extraction conditions make supercritical CO2 a more effective solvent. A total ion chromatogram of the Soxhlet extraction of n-hexane (3,X ) soot can be seen in Fig. 5. The overall trends such as the reduction of HMW-PAHs to LMW-PAHs were observed with increasing air/fuel ratios using Soxhlet extraction and generally are consistent with the results of SFE extracts. However, the dominating peaks vary dramatically in intensity and there are some qualitative differences between the SFE and Soxhlet results as well. This can be attributed to the limited solubility in or availability of the analytes to DCM. Jonker and Koelmans [35], in a Soxhlet extraction study of sediments and environmental carbons (soots, charcoal, coal) found DCM to be the poorest extractant of the 7 studied. With the SFE extractions, 1,10 -methylene-bis-4-methylbenzene was the most abundant, but in the SOX extracts peaks of significant intensity did not appear until beyond 4 min.

retention time. Perhaps this can be attributed to the diverse solubility parameters of the modified CO2 as it is ramped through the various pressures in the supercritical fluid extraction. There is contradictory evidence for some oxidized species such as cyclopenta[def]phenanthrenone and 7H-benz[de]anthracen-7-one, which increased in relative abundance with increase in A/F ratio, but did not disappear as in the SFE extracts of n-hexane soots. Also two new compounds, 1-eicosanol (11.34 min) and 1-hexacosanol (13.49 min), which appeared at higher retention times for n-hexane SOX (9,X ), did not appear in n-hexane SFE (9,X ). There also is discrepancy between the unoxidized PAHs for the Soxhlet extractions and those observed with the SFE extractions. For n-hexane soot, SFE extracts showed unoxidized HMW-PAHs, such as benzo[k]fluoranthene, benzo[a]pyrene, 2,20 -binaphthalene, indeno[1,2,3-cd]pyrene, indeno[1,2,3-cd]fluoranthene, and benzo[ghi]perylene, although these compounds did not appear in extracts of (6,X ) soot. However, for the Soxhlet extractions of nhexane soot, the unoxidized HMW-PAHs remain in large abundance up to (6,X ). Extractions of the soot from the (6,X ) A/F ratio, however, was very difficult to reproduce in Soxhlet extractions. Variability in relative peak heights as well as peak appearances could not be eliminated. Ultimately, replicate Soxhlet extractions of PAHs showed random abundances, regardless of how carefully each sample was handled. 3.2. JP-8 soots Total ion chromatograms of SF extracted JP-8 soot are shown in Figs. 6 and 7 and similar trends to those

C.C. Jones et al. / Carbon 42 (2004) 2471–2484

2477

Abundance 300000

TIC: [BSB1]603CJ.D 7.60 7.88 7.94 8.14

6.00

10.36

280000 260000 13.19

240000 10.20

220000 6.32

200000

9.87

180000 160000 10.46

6.07

140000 120000

7.377.68

100000

7.54

8.63 8.73

5.45

60000 3.994.38

40000 20000

9.45

8.57

80000

3.12

4.99 5.64 5.025.35

13.96

12.82

8.458.83 9.40

6.18

7.97 8.23

6.35

10.63 9.29

9.99

11.23 10.85 11.43

12.23

14.31

12.54 11.9812.38

13.39 13.50

0 4.00

5.00

6.00

7.00

8.00

9.00

10.00

11.00

12.00

13.00

14.00

15.00

14.00

15.00

Retention Time-->

Fig. 6. Total ion chromatogram of the SFE extract of JP-8 soot (3,X ).

Abundance

TIC: [BSB2]606CJ.D 5.48

300000 280000 260000

4.05

240000 220000 200000 180000 160000

4.48

140000 4.88

120000 100000 80000

4.43

60000 40000

3.68 4.78 4.22 3.99 4.36

7.96

6.06 5.19 20000 3.16 5.53 6.21 6.30 4.67 3.62 4.14 4.56 4.71 3.53 6.14 4.98 6.38 3.11 5.05 3.41 3.32 3.94 6.58 5.34 3.82 0 4.00 5.00 6.00 7.00 Retention

7.62 8.11 8.638.98 8.00

9.00

9.67 10.26 10.00

11.00

12.00

13.00

Time-->

Fig. 7. Total ion chromatogram of the SFE extract of JP-8 soot (11,X ).

observed for n-hexane can be seen, such as increasing the air/fuel ratio resulting in a decrease of HMW-PAHs. The dominant compounds for JP-8 also tend to favor higher molecular weight species for the (0,X ), (3,X ), and (6,X ) soots. The dominant oxidized PAH species are apparent in Fig. 8 and consist of 2-naphthalene-carboxaldehyde, 9H-fluoren-9-one, 1,2 acenaphthylenedione, 1H-phenalen-1-one, 1,8-naphthalic anhydride, 7H-benz[de]anthracen-7-one, and cyclopenta[def]phenanthrenone. These are similar to the extracts of n-hexane soot except for 2-naphthalene-carboxaldehyde. Con-

comitant with the formation of the oxidized species, the HMW compounds such as 7H-benz[de]anthracen-7-one and cyclopenta[def]phenanthrenone disappear at (6,X ) which is at a higher A/F than the n-hexane case, due to the 20–30% aromatic content in JP-8 fuel which lowers the barrier of aromatic and soot formation, requiring higher A/F ratios to favor LWM-PAHs and oxidized soot. The oxidized species 1,8-naphthalic anhydride is the most abundant at (0,X ) and (9,X ) but diminishes to about 2% at higher A/F ratios, in contrast to n-hexane. Dominant unoxidized species include fluoranthene, and

2478

C.C. Jones et al. / Carbon 42 (2004) 2471–2484

40.00 35.00 30.00 25.00 % Area 20.00 15.00 10.00 5.00 (3,X)

0.00

(6,X) 1

2 3 4 5

6 7 8 9

(9,X) (11,X) 10 11 12

13 14 15 16

17 18

Compound

19 20

21 22 23

24 25

(9,X) 26

27 28 29

30 31

(3,X)

Air/Fuel Ratio

32 33 34

Fig. 8. PAH content versus air/fuel ratio of JP-8 combustion soots SFE extractions; compound number begins with 1 as phthalic anhydride moving up to 34 as benzo[ghi]perylene; refer to Table 1 for corresponding compound #s.

pyrene at (0,X ), then shift to benzo[k]fluoranthene and 2,20 -binaphthalene for (3,X ) and (6,X ). Then at (9,X ) and (11,X ), 1-methyl-4-phenylmethylbenzene and 1,10 methylene-bis[4-methylbenzene] dominate the chromatograms with an appearance of cyclododecane at (11,X ). The shift is again from high molecular weight PAHs to low molecular weight PAHs as the A/F ratio is increased. There is a large abundance of the HMW-PAHs and commonly known carcinogenic PAHs in the A/F range (0,X )–(6,X ). At (3,X ) and (6,X ) benzo[k]fluoranthene and 2,20 -binaphthalene are the most abundant

peaks, which again can be attributed to the 20–30% aromatic content in the JP-8 fuel. Other large PAHs such as perylene, benzo[a]pyrene, benzo[e]pyrene, indeno[1,2,3-cd]pyrene, indeno[1,2,3-cd]fluoranthene, and benzo[ghi]perylene all show significant abundance in the range of (0,X )–(6,X ). There is an increase in chrysene, benz[a]anthracene, and triphenylene compared to the nhexane soot, but they also disappear at A/F ratios higher than (6,X ). For (9,X ) and (11,X ) the most abundant compound becomes the lower molecular weight aromatic 1-methyl-4-phenylmethylbenzene, also seen in

Abundance TIC: [BSB1]704CJ.D 7.96 8.03

300000 280000

8.24

6.40

260000 240000

7.69

220000 200000

9.98

180000

6.15

10.28

160000 13.33

140000

10.61 10.36

120000 6.10

100000 80000

8.80

5.53

7.77

60000 40000

14.22

7.62

4.04

6.25

7.46

8.68

8.65 8.90 8.17 8.60 8.76 8.07 8.31

9.54 9.63

20000

11.41 10.74 10.95

12.36

12.99 12.95

14.60 13.57

0 Retention Time-->

4.00

5.00

6.00

7.00

8.00

9.00

10.00

11.00

12.00

13.00

Fig. 9. Total ion chromatogram of the SFE extract of diesel soot (3,X ).

14.00

15.00

C.C. Jones et al. / Carbon 42 (2004) 2471–2484

n-hexane soot. Cyclodecane appears in (11,X ) with a percentage area greater than 10 which was not seen in the n-hexane material. It appears that the most drastic change moving from n-hexane to JP-8, is that a larger number of HMWPAHs are formed and the most dominant peaks are HMW-PAHs which prevail at (0,X )–(3,X ) but show similar compositions to the n-hexane soot at (9,X ) and (11,X ). Again, this can be attributed to 20–30% aromatic content and the large and highly branched hydrocarbon chains (C10 –C11 ) of JP-8, allowing a

Abundance

2479

greater opportunity for the larger PAHs to form prior to sooting. 3.3. Diesel soots Diesel soot shows the same trend as JP-8 for supercritical fluid extracts, as shown by Figs. 9–11, in that with increasing A/F ratio the high molecular weight PAHs disappear and the low molecular weight PAHs begin to form. Diesel also contains large and highly branched alkanes (C12 –C14 ) as well as 5–30% aromatic

TIC: [BSB2]702CJ.D 4.04

280000 260000 240000 220000 200000 180000

4.50

160000 140000 120000

4.14

100000

4.84

80000 60000

3.74 3.97 4.32 4.40

40000 20000 0

3.11 3.52 3.67 3.59

Retention

5.18

4.88 4.434.76 5.42 4.27

4.00

5.00

6.00

7.00

8.00

9.00

10.00

11.00

12.00

13.00

14.00

15.00

Time-->

Fig. 10. Total ion chromatogram of the SFE extract of diesel soot (11,X ).

70.00 60.00 50.00 40.00 % Area

30.00

(3,X)

20.00

(6,X) (9,X)

10.00

(11,X)

0.00 1 2

3 4 5

6

7 8 9 10 11 12

13 14 15 16 17 18

Compound

19 20 21 22 23 24

25 26 27 28

(9,X) 29 30 31 32 33

(3,X) Air/Fuel Ratio

Fig. 11. PAH content versus air/fuel ratio of diesel combustion soots SFE extractions; compound number begins with 1 as phthalic anhydride moving up to 33 as benzo[ghi]perylene; refer to Table 1 for corresponding compound #s.

2480

C.C. Jones et al. / Carbon 42 (2004) 2471–2484

content which is considered to be a source of nucleation to larger aromatic compounds and soot. For example benzo(k)fluoranthene and benzo(j)fluoranthene dominate the chromatogram and continue up to (6,X ). The same oxidized PAHs are formed except cyclopenta[def]phenanthrenone, which appears in n-hexane and JP-8 soots, but did not appear in diesel soot. Also 1,8naphthalic anhydride was not seen in the extract of the diesel (11,X ). The extract of diesel (11,X ) shows no components at retention times higher than 5.42 min. Large alkyl chains, such as 6-methyltridecane, 3,6-methyl-

undecane, 3-methyl-undecane, and hexadecane were extracted from the diesel soots but did not appear in the n-hexane or JP-8 extracts. 3.4. EPR studies Unpaired electron spin density has been measured by EPR in n-hexane, JP-8, and diesel soots at 1018 –1020 spins per gram [36] and it was of interest to determine if any unpaired electrons associated with the PAH constituents persisted in the solvent extracts of soots.

4000

(B) 3000

2000

1000

0 3240

3250

3260

3270

3280

3290

3300

3310

3300

3310

3320

3330

3340

-1000

-2000

-3000

-4000 Gauss 6000

(A) 4000

2000

0 3240

3250

3260

3270

3280

3290

3320

3330

3340

-2000

-4000

-6000 Gauss

Fig. 12. A. EPR signal of n-hexane regular soot before SFE extraction, 9.41 · 1018 spins g1 and B. EPR signal of the n-hexane regular soot after SFE extraction, 5.66 · 1018 spins g1 .

C.C. Jones et al. / Carbon 42 (2004) 2471–2484

Electron paramagnetic resonance (EPR) was performed on the n-hexane regular soot before and after SFE. No signal from the filtered extract attributable to extracted free radical species could be observed, undoubtedly due to their rapid combination reactions in the liquid phase. However, the decrease in spin density of the soot following extraction is evidence that a significant fraction, 40%, of the unpaired electrons, is associated with extracted molecules. Fig. 12 shows the EPR spectra of nhexane (regular) soot before and after SFE extraction in illustration of these results. In separate experiments, the EPR spectrum of the Soxhlet extraction of diesel (0,X ) soot yielded the same result, with somewhat more of the spin density associated with the extracted molecules (55%). The high density of unpaired electrons in combustion soots appears to be divided between associated PAH molecules and either the aromatic soot structure and/or unextractable moieties. 3.5. Relationship to other A/F-dependent soot properties Extensive studies have been performed on the properties and reactivity of combustion soots that corre-

2481

spond well with the polycyclic aromatic hydrocarbon content as determined by this analysis. Particularly, the effects of A/F ratio on properties and reactivity of combustion soots now can be compared to the specific polyaromatic content of the n-hexane, JP-8 and diesel combustion soots. These properties include surface oxidation, surface area, unpaired electron spin density, hydration, and ozone oxidation [36]. It was observed by FTIR spectroscopy that for increasing A/F ratio, an index of surface oxidation (A1725 =A1590 ) also increased. This observed effect in conjunction with that of the PAH content observed in this analysis, demonstrates that increasing A/F ratio results in larger amounts of oxidized PAHs seen in the soots formed from all three fuels. A significant property influencing soot reactivity is that of unpaired electron spin density which has been measured as 1018 –1020 spins per gram and decreases linearly with increasing A/F combustion ratio used in forming the soot [36]. At high air/fuel ratios the larger aromatic structures are replaced by LMW-PAHs and there is a greater abundance of oxidized functionalities, as seen in this analysis, thus fewer unpaired electrons.

Table 1 Compounds identified in the extractions of each prepared soot at air/fuel combustion ratios of (0,X ), (3,X ), (6,X ), (9,X ), and (11,X ) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

n-Hexane SFE

JP-8 SFE

Diesel SFE

Phthalic anhydride 1-Methyl-3-phenylmethyl benzene 1-Methyl-4-phenylmethyl benzene 1,10 -Methylene-bis-4-methyl benzene 2-Methylphenyl-phenyl methanone 9H-Fluoren-9-one Anthracene Phenanthrene 1,2-Acenaphthylenedione 1H-Phenalen-1-one 2-Phenylnaphthalene 9,10-Anthracenedione 1,8-Napthalic anhydride Fluoranthene Pyrene Benzo[ghi]fluoranthene Benz[a]anthracene Chrysene 7H-Benz[de]anthracen-7-one Cyclopenta[def]phenanthrenone Benzo(k)fluoranthene Benzo[a]pyrene 2,20 -Binaphthalene Indeno[1,2,3-cd]pyrene Indeno[1,2,3-cd]fluoranthene Benzo[ghi]perylene

Phthalic anhydride 4-Methylphthalic anhydride 2-Naphthalene-carboxaldehyde 1-Methyl-4-phenylmethylbenzene 1,10 -Methylene-bis-4-methylbenzene Cyclododecane 9H-Fluoren-9-one Anthracene Phenanthrene 1,2-Acenaphthylenedione 1H-Phenalen-1-one 2-Phenylnaphthalene 9,10-Anthracenedione 1,8-Napthalic anhydride Fluoranthene Pyrene ISOMER ISOMER Cyclopenta[cd]pyrene ISOMER Benz[a]anthracene Triphenylene Chrysene 7H-Benz[de]anthracen-7-one Phenanthrene, 2-phenyl Cyclopenta[def]phenanthrenone Benzo(k)fluoranthene Perylene Benzo[a]pyrene 2,20 -Binaphthalene Benzo(e)pyrene Indeno[1,2,3-cd]pyrene Indeno[1,2,3-cd]fluoranthene Benzo[ghi]perylene

Phthalic anhydride 2-Naphthalene carboxaldehyde Benzene, 1-methyl-4-[phenylmethyl Octane, 2,4,6-trimethylBenzene, 1,10 -methylene-bis[4-methyl Tridecane, 6-methyl Undecane, 3,6-dimethylUndecane, 3-methylAnthracene Phenanthrene Hexadecane 1,2-Acenaphthylenedione 1H-Phenalen-1-one 9,10-Anthracenedione 1,8-Napthalic anhydride Fluoranthene Pyrene Benzo[ghi]fluoranthene Benz[a]anthracene Triphenylene Chrysene 7H-Benz[de]anthracen-7-one Phenanthrene, 2-phenyl Benzo(k)fluoranthene Perylene Benzo[a]pyrene Benzo[e]pyrene Benzo(j)fluoranthene Benz[e]acephenanthrylene Indeno[1,2,3-cd]pyrene Indeno[1,2,3-cd]fluoranthene Dibenz[a,h]anthracene Benzo[ghi]perylene

2482

C.C. Jones et al. / Carbon 42 (2004) 2471–2484

Soot hydration is another important property that affects the kinetics and mechanisms of atmospheric reactions and ties nicely with observations in this study. It was observed that an increase of soot particle hydration occurs with increased air/fuel ratio [36]. This is due to the larger amounts of C–O functionalities serving as hydration sites, which is consistent with the observations of this analysis that an abundance of oxidized PAHs are observed at higher air/fuel ratios. Ozone oxidation is an important reaction due to unresolved questions as to the potential role of soot in the upper troposphere and lower stratosphere. As seen by Chughtai et al. [36], as the A/F ratio and the particle oxidation is increased, which diminishes the more reactive unsaturated linkages, the soot reactivity with ozone decreases. The principal reaction pathway of ozone with soot is the formation of surface carboxylic functionalities, which are seen to be abundant in this analysis at higher air/fuel ratios. Such oxidized PAH compounds such as 2-naphthalene-carboxaldehyde, 4-methylphthalic anhydride, 1,2-acenaphthylenedione, 1H-phenalen1-one, 9,10-anthracenedione, 1,8-napthalic anhydride, reflect compounds the formation of which results in the diminution of the reaction centers for ozone. A summary of the dominant species of PAHs observed in these studies is provided in Table 1.

4. Conclusions The analysis of carbonaceous soot for polycyclic aromatic hydrocarbons (PAH) at selected A/F ratios was conducted on those produced by combustion of nhexane, JP-8, and diesel, using SFE followed by GC/ MS. While this analysis undoubtedly reveals less than the complete pattern of combustion soot PAH content, because of the molecular weight limitations of GC/MS, the effects of varying air/fuel combustion ratio on the nature of the extractable constituents is clear. Total ion chromatograms of the n-hexane, JP-8 and diesel soot extracts show a consistent progression from a large array of high and low molecular weight PAHs at lower A/ F ratios of (0,X ), (3,X ), and (6,X ) to a significant reduction of these compounds, leaving only lower molecular weight aromatics at higher A/F ratios such as (9,X ) and (11,X ). A reduction in total extractable compounds with increasing A/F ratio was observed in this analysis, and large (>202 MW) polycyclic aromatic hydrocarbons diminished rapidly as the air/fuel combustion ratio increased. Although no single existing theory of PAH and soot formation explains all experimental observations, there is extensive evidence that PAH formation and soot particle growth is substantially due to the sequential combination of aromatic and/or polyyne radicals with acetylene or polyacetylene species [37–48]. It is not

surprising that the presence of higher concentrations of free radical-reactive O2 (which yields the strong oxidant HO ), and decreased residence time of soot precursors in a fuel rich region of the flame, result in lower concentrations and a larger number of smaller oxidized PAH molecules. That the growth rate of soot particles in the flame diminishes with A/F ratio is consistent with this interpretation, as is the greater surface oxidation and lower spin density of the particles [36]. The A/F ratios used in these experiments are well below stoichiometry and the O2 concentration, while decreasing the soot precursor concentrations and thus the soot formation rate, also leaves the extractable HMW-PAH concentrations too low for detection at the higher A/F ratios. While this interpretation is consistent with a large body of research on soot formation in flame, alternative mechanisms cannot be ruled out on the basis of the data. For example, some larger PAH molecules formed as a result of the reaction with oxygen, more tightly bound to the soot matrix, may not be extractable. Sooting continues to occur even for the higher air/fuel ratios used in this work, (9,X ) and (11,X ), still below stoichiometry, and the presence of the smaller aromatics in the (9,X ) and (11,X ) materials cannot be disregarded as possible reactants and/or health hazards. The presence of both oxidized and unoxidized PAH species was observed for n-hexane, JP-8 and diesel. However, the relative abundance of the oxidized and unoxidized species differed for each fuel and air/fuel ratio. It was seen that JP-8 and diesel yield a larger abundance of higher molecular weight PAHs, such as benzo(a)pyrene and benzo(k)fluoranthene at the low A/ F ratios, which is attributed to their 20–30% aromatic contents. It also was seen that an increase in A/F ratio resulted in a larger abundance of low molecular weight oxidized and unoxidized PAHs such as 9H-fluorenone, 1,2-acenaphthylenedione, 1,8-naphthalic anhydride, 1methyl-(4-phenylmethyl)-benzene and 1,10 -methylene(bis-4-methyl)-benzene. This shows that both the molecular composition of the fuel and level of oxidation during combustion determine the relative abundance of oxidized and unoxidized species. The supercritical fluid extractions were compared as to completeness and reproducibility with Soxhlet extractions of the same soots. Supercritical fluid extraction is superior to Soxhlet extraction for the PAH analysis of soots in every respect. This undoubtedly is due to both the molecular size of CO2 , which allows greater molecular penetration into the pore structure of the soot than that of dichloromethane, and the nature of the fluid phase at the pressure and temperature of the extraction itself. Unpaired electron spin studies on the soot extracts of both supercritical and Soxhlet extractions displayed only very small amounts of unpaired electrons in the soot extracts. This could be due to a small amount of

C.C. Jones et al. / Carbon 42 (2004) 2471–2484

residual fine (non-filterable) soot particulate as attempts to completely remove such particles was of uncertain success. PAH free radicals likely combine rapidly in the liquid phase. The decrease in spin density of the soot following extraction, however, is evidence that a significant fraction of the unpaired electrons are associated with extracted molecules. The combustion trends of the three selected fuels correspond well with measurements performed in these laboratories on the reactivity and properties of combustion soots. For the most part, the polycyclic aromatic hydrocarbon contents determined in these analyses support the trends with A/F observed for surface oxidation, surface area, unpaired electron spin density, hydration, and ozone oxidation of the combustion soots. FT-IR measurements of soots have shown oxidized functional groups whose identities are confirmed by the identification of extracted oxidized PAHs. We now are developing instrumental techniques to determine if, and the extent to which, higher molecular weight extractables (i.e., MW > 300) are present in these soots.

Acknowledgements The authors gratefully acknowledge support from the National Science Foundation through Grants ATM9618495 and ATM-0000911. Gratitude also is expressed to G.R. Eaton and S.S. Eaton of the Department of Chemistry and Biochemistry for assistance with the EPR measurements. References [1] Finlayson-Pitts BJ, Pitts JN. Chemistry of the upper and lower atmosphere. New York: Academic Press; 2000. p. 436–546. [2] Akhter MS, Chughtai AR, Smith DM. The structure of hexane soot II: extraction studies. Appl Spectrosc 1985;39:154–67. [3] Penner JE, Eddleman H, Novakov T. Towards the development of a global inventory for black carbon emissions. Atmos Environ 1993;27:1277–95. [4] Cooke WF, Wilson JJN. A global black carbon aerosol model. J Geophys Res 1996;101(D-14):19395–409. [5] Charlock TP, Sellers WD. Aerosol effects on climate: calculations with time-dependent and steady-state radiative–convective models. J Atmos Sci 1980;37:1327–41. [6] Toon OW, Pollack JB. Atmospheric aerosol and climate. Am Sci 1980;68:268–77. [7] Chylek PV, Ramaswamy V, Cheng RJ. Effect of graphitic carbon on the albedo of clouds. J Atmos Sci 1984;41:3076–84. [8] Charlson RJ, Schwartz SE, Hales JM, Cess RO, Coakley Jr JA, Hansen JE, et al. Climate forcing by anthropogenic aerosols. Science 1992;255:423–30. [9] Haywood JM, Ramaswamy V. Global sensitivity studies of the direct radiative forcing due to anthropogenic sulfate and black carbon aerosols. J Geophys Res 1998;103:6043–58.

2483

[10] Sagai M, Saito H, Ichinosa T, Kodama M, Mori Y. Biological effects of diesel exhaust particles. In vitro production of superoxide and in vivo toxicity in mouse. Free Rad Biol Med 1993;14:37– 47. [11] Pope III CA, Thun MJ, Namboodiri MM, Dockery DW, Evans JS, Speizer FE, et al. Particulate air pollution as a prediction of mortality in a prospective study of U.S. adults. Am J Respir Crit Care Med 1995;151:669–74. [12] Pope III CA, Venier RL, Lovett EG, Larson AC, Raizenne ME, Kanner RE, et al. Heart rate variability associated with particulate air pollution. Am Heart J 1999;138(5):890–9. [13] Chang SG, Brodzinsky R, Gundell LA, Novakov T. Chemical and catalytic properties of elemental carbon. In: Wolff GT, Klimisch RL, editors. Particulate carbon: atmospheric life cycle. New York: Plenum Press; 1982. p. 159–79. [14] Goldberg ED. Black carbon in the environment. New York: Wiley; 1985. [15] Chughtai AR, Gordon SA, Smith DM. Kinetics of the hexane soot reaction with NO2 /N2 O4 at low concentration. Carbon 1994;32(3):405–16. [16] Smith DM, Chughtai AR. Reaction kinetics of ozone at low concentration with n-hexane soot. J Geophys Res 1996;101(D14): 19607–20. [17] Ammann M, Kalberer M, Jost DT, Tobler L, Rossler E, Piguet D, et al. Heterogeneous production of nitrous acid on soot in polluted air masses. Nature 1998;395:157–60. [18] Lary DJ, Schallcross DE, Toumi R. Carbonaceous aerosols and their potential role in atmospheric chemistry. J Geophys Res 1999;104(D13):15929–40. [19] Kamm S, Mohler O, Naumann KH, Saathoff H, Schurath U. The heterogeneous reaction of ozone with soot aerosol. Atmos Environ 1999;33:4651–61. [20] Chughtai AR, Miller NJ, Pitts JR, Smith DM. Carbonaceous particle hydration III. J Atmos Chem 1999;34:259–79. [21] Kirchner V, Scheer V, Vogt R. FTIR spectroscopic investigation of the mechanism and kinetics of the heterogeneous reactions of NO2 and HNO3 with soot. J Phys Chem A 2000;104:8908–15. [22] Fernandes MB, Brooks P. Characterization of carbonaceous combustion residues: II. Nonpolar organic compounds. Chemosphere 2003;53:447–58. [23] Hawthorne SB. Analytical-scale supercritical fluid extraction. Anal Chem 1990;62:633A–42A. [24] Benner Jr BA. Summarizing the effectiveness of supercritical fluid extraction of polycyclic aromatic hydrocarbons from natural matrix environmental samples. Anal Chem 1998;70:4594– 601. [25] Langenfeld JJ, Hawthorne SB, Miller DJ, Pawliszyn J. Effects of temperature and pressure on supercritical fluid extraction efficiencies of polycyclic aromatic hydrocarbons and polychlorinated biphenyls. Anal Chem 1993;65:338–44. [26] Langenfeld JJ, Hawthorne SB, Miller DJ, Pawliszyn J. Role of modifiers for analytical-scale supercritical fluid extraction of environmental samples. Anal Chem 1994;66:909–16. [27] Jensen TE, Hites RA. Aromatic diesel emissions as a function of engine conditions. Anal Chem 1983;55:594–9. [28] Decesari S, Facchini MC, Matta E, Mircea M, Fuzzi S, Chughtai AR, et al. Water soluble organic compounds formed by oxidation of soot. Atmos Environ 2002;36:1827–32. [29] K€ onig J, Balfanz E, Funcke W, Romanowski T. Determination of oxygenated polyaromatic hydrocarbons in airborne particulate matter by capillary gas chromatography and gas chromatography/ mass spectrometry. Anal Chem 1983;55:599–603. [30] Allen JO, Dookeran NM, Taghizadeh K, Lafleur AL, Smith KA, Sarofim AF. Measurement of polycyclic aromatic hydrocarbons associated with a size-segregated urban aerosol. Environ Sci Technol 1997;31:2064–70.

2484

C.C. Jones et al. / Carbon 42 (2004) 2471–2484

[31] Hannigan MP, Cass GR, Penman BW, Crespi CL, Lafleur AL, Busby Jr WF, et al. Bioassay-directed chemical analysis of Los Angeles airborne particulate matter using a human cell mutagenicity assay. Environ Sci Technol 1998;32:3502–14. [32] Fraser MP, Cass GR, Simoneit BRT, Rasmussen RA. Air quality model evaluation data for organics. 5. C6 –C22 nonpolar and semipolar aromatic compounds. Environ Sci Technol 1998;32: 1760–70. [33] Fialkov AB, Dennebaum J, Homann KH. Large molecules, ions, radicals, and small soot particles in fuel-rich hydrocarbon flames. Part V: positive ions of polycyclic aromatic hydrocarbons (PAH) in low-pressure premixed flames of benzene and oxygen. Combust Flame 2001;125:763–77. [34] McClellan AL, Harnsberger HF. Cross-sectional areas of molecules adsorbed on solid surfaces. J Colloid Interface Sci 1967;23:577–99. [35] Jonker MTO, Koelmans AA. Extraction of polycyclic aromatic hydrocarbons from soot and sediment: solvent evaluation and implications for sorption mechanism. Environ Sci Technol 2002;36:4107–13. [36] Chughtai AR, Kim JM, Smith DM. The effect of air/fuel ratio on properties and reactivity of combustion soots. J Atmos Chem 2002;43:21–43. [37] Bonne U, Homann K-H, Wagner HGg. Carbon formation in premixed flames. In: Proceedings, Symposium (International) on Combustion. Germany: University Goettingen; 1965. p. 503–10. [38] Wagner HGg. Soot formation––an overview. In: Siegla DC, Smith GW, editors. Particulate carbon formation during combustion. New York: Plenum Press; 1981. p. 1–29. [39] Bittner JD, Howard JB. Pre-particle chemistry in soot formation. In: Siegla DC, Smith GW, editors. Particulate carbon formation during combustion. New York: Plenum Press; 1981. p. 109–42. [40] Bauerle St, Karasevich Y, Slavov St, Tanke D, Tappe M, Thienel Th, et al. Soot formation at elevated pressures and carbon concentrations in hydrocarbon pyrolysis. In: Proceedings, Sym-

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

posium (International) on Combustion. Germany: University Goettingen; 1994. p. 627–34. Knorre VG, Tanke D, Thienel Th, Wagner HGg. Soot formation in the pyrolysis of benzene/acetylene and acetylene/hydrogen mixtures at high carbon concentrations. In: Proceedings, Symposium (International) on Combustion. Russia: Moscow State Technical University; 1996. p. 2303–10. Homann K-H, Wagner HGg. Some aspects of soot formation. In: Combustion science and technology book series 2 Dynamics of exothermicity. Darmstadt, Germany: Institut fur Physikalische Chemie; 1996. p. 151–84. McEnally CS, Schaffer AM, Long MB, Pfefferle LD, Smooke MD, Colket MB, et al. Computational and experimental study of soot formation in a coflow, laminar ethylene diffusion flame. In: Proceedings, Symposium (International) on Combustion. New Haven, CT: Yale University; 1998. p. 1497–505. McEnally CS, Pfefferle LD, Robinson AG, Zwier TS. Aromatic hydrocarbon formation in nonpremixed flames doped with diacetylene, vinylacetylene, and other hydrocarbons: evidence for pathways involving C4 species. Combust Flame 2000;123(3): 344–57. Roesler JF, Martinot S, McEnally CS, Pfefferle LD, Delfau J-L, Vovelle C. Investigating the role of methane on the growth of aromatic hydrocarbons and soot in fundamental combustion processes. Combust Flame 2003;134(3):249–60. Zelepouga SA, Saveliev AV, Kennedy LA, Fridman AA. Relative effect of acetylene and PAHs addition on soot formation in laminar diffusion flames of methane with oxygen and oxygenenriched air. Combust Flame 2000;122(1/2):76–89. Appel J, Bockhorn H, Frenklach M. Kinetic modeling of soot formation with detailed chemistry and physics: laminar premixed flames of C2 hydrocarbons. Combust Flame 2000;121(1/2):122– 36. D’Anna A, Violi A, D’Alessio A. Modeling the rich combustion of aliphatic hydrocarbons. Combust Flame 2000;121(1/2):418–29.