Science of the Total Environment 490 (2014) 639–646

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Mechanistic and kinetic studies on the OH-initiated atmospheric oxidation of fluoranthene Juan Dang, Xiangli Shi, Qingzhu Zhang ⁎, Jingtian Hu, Jianmin Chen, Wenxing Wang Environment Research Institute, Shandong University, Jinan 250100, PR China

H I G H L I G H T S • We studied a comprehensive mechanism of OH-initiated oxidation of fluoranthene. • We reported the formation pathways of fluoranthone, fluoranthenequinone and epoxide. • The rate constants of the crucial elementary steps were evaluated.

a r t i c l e

i n f o

Article history: Received 2 April 2014 Accepted 30 April 2014 Available online xxxx Editor: Pavlos Kassomenos Keywords: Fluoranthene OH radicals Oxidation mechanism Oxidation products Rate constants

a b s t r a c t The atmospheric oxidation of polycyclic aromatic hydrocarbons (PAHs) can generate toxic derivatives which contribute to the carcinogenic potential of particulate organic matter. In this work, the mechanism of the OHinitiated atmospheric oxidation of fluoranthene (Flu) was investigated by using high-accuracy molecular orbital calculations. All of the possible oxidation pathways were discussed, and the theoretical results were compared with the available experimental observation. The rate constants of the crucial elementary reactions were evaluated by the Rice–Ramsperger–Kassel–Marcus (RRKM) theory. The main oxidation products are a range of ring-retaining and ring-opening chemicals containing fluoranthols, fluoranthones, fluoranthenequinones, nitro-fluoranthenes, dialdehydes and epoxides. The overall rate constant of the OH addition reaction is 1.72 × 10−11 cm3 molecule−1 s−1 at 298 K and 1 atm. The atmospheric lifetime of Flu determined by OH radicals is about 0.69 days. This work provides a comprehensive investigation of the OH-initiated oxidation of Flu and should help to clarify its atmospheric conversion. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Polycyclic aromatic hydrocarbons (PAHs), an important fraction of semi-volatile organic compounds (SVOCs), are formed as byproducts of any incomplete combustion from traffic exhausts, industrial activities, domestic heating, forest fires and biomass burnings (Chen et al., 2013; Christie et al., 2012; Seinfeld and Pandis, 2006). The atmospheric emission of 16 priority PAHs in all Asian countries accounted for 53.5% of the total global emissions (504 Gg), with the highest emission from China (106 Gg) and India (67 Gg) during 2007 (Shen et al., 2013). PAHs are hydrophobic, stable, and sparingly soluble in water (Chaudhry, 1994). They can be metabolized to reactive electrophilic intermediates that can form DNA adducts, which may induce mutations and ultimately tumors (Ramírez et al., 2011). Due to the potential mutagenicity and carcinogenicity, their ubiquitous presence in air, water, soil

⁎ Corresponding author. Fax: +86 531 8836 1990. E-mail address: [email protected] (Q. Zhang).

http://dx.doi.org/10.1016/j.scitotenv.2014.04.134 0048-9697/© 2014 Elsevier B.V. All rights reserved.

and vegetation is of major concern (Boström et al., 2002; Xue and Warshawsky, 2004). The European Union has set a target value of 1 ng m−3 of benzo[a]pyrene used as a main indicator of carcinogenic PAHs (EU, 2005). Fluoranthene (Flu), a member of non-alternant PAHs, is one of the most abundant PAHs in the environment (Monte et al., 2012; Wetzel et al., 1994). Because of its high concentration in ambient air and potential carcinogenicity, it is suggested as a complementary indicator to benzo[a]pyrene (Boström et al., 2002). During spring of 1994–2000, the measurement in Stockholm has shown ambient air concentrations of Flu ranging from 8 to 25 ng m−3 (Boström et al., 2002). Air samples collected from eight locations in the Laurentian Great Lakes region revealed that the maximum concentration of Flu is up to 9350 pg m−3 (Galarneau et al., 2006). Atmospheric monitoring in Eordea Basin, west Macedonia, Greece and Kurashiki City has also detected Flu and other PAHs, the gas phase concentration of Flu scattered over the range of 0.623–30.3 ng m−3 (Terzi and Samara, 2004). On account of its prevalent presence in air, it is critical to understand the fate of gaseous Flu. In general, the tropospheric removal of Flu involves wet and dry deposition, and oxidation reactions with OH, NO3 and O3. The

640

J. Dang et al. / Science of the Total Environment 490 (2014) 639–646

OH E=1.38 H= -19.59

pathway 1

TS1 IM1 OH

E=1.61 H= -17.98

pathway 2

TS2 IM2

+OH

E=0.57 H= -23.39

OH

pathway 3

TS3 fluoranthene

IM3

E=1.47 H= -17.97

pathway 4

TS4

OH IM4

E=1.22 H= -21.74

H= -1.64

TS5

pathway 5

OH

OH

IM5 Fig. 1. The OH addition reaction scheme of Flu embedded with the potential barrier ΔE (in kcal/mol) and reaction heat ΔH (in kcal/mol). ΔH is calculated at 0 K.

wet and dry deposition of gaseous Flu is relatively insignificance as a removal pathway. Among the various oxidants, OH radicals play the most essential role in determining the oxidation power of the atmosphere (Keyte et al., 2013). The reaction with OH radicals is considered to be a dominant removal process for gaseous Flu. However, there exists a notable absence of direct experimental data concerning the reaction mechanism, largely due to the lack of efficient detection schemes for intermediate radical species. In this paper, using quantum mechanics and the RRKM theory, we carried out a theoretical study on the OHinitiated atmospheric oxidation reaction of Flu in the presence of O2/ NO and HO2, which is helpful to clarify the atmospheric fate of Flu.

By means of the MESMER program (Glowacki et al., 2012), the rate constants of the crucial elementary reactions were deduced by using Rice–Ramsperger–Kassel–Marcus (RRKM) theory (Robinson and Holbrook, 1972). The RRKM rate constant is given by:

2. Computational method

kðT Þ ¼

The electronic structure calculations were performed with the Gaussian 03 software package (Frisch et al., 2003). Geometries of the reactants, intermediates, transition states, and products were optimized at the BB1K/6-31 + G(d,p) level, which has yielded satisfying results in the previous research (Qu et al., 2006). The harmonic frequency calculations were also performed at the same level in order to determine the nature of the stationary points, the zero-point energy (ZPE), and the thermal contribution to the free energy of activation. Besides, the intrinsic reaction coordinate (IRC) analysis was carried out to confirm that each transition state connects to the right minima along the reaction path. For a more accurate evaluation of the energetic parameters, a more flexible basis set 6-311 + G(3df,2p), was employed to determine the single point energies of various species.

kðEÞ ¼

W ðEÞ hρðEÞ

ð1Þ

where, W(E) is the rovibrational sum of states at the transition state, ρ(E) is the density of states of reactants, and h is Planck's constant. Then, canonical rate constant k(T) is determined by using the usual equation: 1 ∫kðEÞρðEÞ expð−βEÞdE Q ðTÞ

ð2Þ

where, Q(T) is the reactant partition function. 3. Results and discussion Due to the lack of experimental information on the thermochemical parameters for the present reaction system, it is difficult to compare the calculated results with experimental data directly. To verify the reliability of the computational results, we optimized the geometries and calculated the vibrational frequencies of benzene, phenol and naphthalene. The results agree well with the available experimental values, and the maximum relative errors are less than 3.0% for geometrical parameters and less than 7.2% for vibrational frequencies

J. Dang et al. / Science of the Total Environment 490 (2014) 639–646

H O +O 2

O

IM15

H HO2

E=10.58 H= -27.25

-HO2

E=9.53 H=-27.18

-HO2 P4

H HO2

+O2

O

E=4.41 H=1.27

E=17.60 H= 9.17

O

+O2

IM17

OH

+HO 2

+O2

O

IM16

E=11.17 +HO 2 H= 1.78 H2O2

O

O

E=31.34 H=-0.41

H=-16.28

OH

H

OONO O -NO2

+NO

E=19.75 H= 9.94

IM14

H

OO

641

HO2

+O2

-HO2

-HO2

IM8

IM9

IM10 -OH

IM6 O

H OH

O

E=8.22 H=-25.58

E=9.30 H=-9.89

O

OOH OH

H shift

IM2

E=25.39 H=-19.38

OO H OH

+O2 E=13.11 H= -3.47

IM7

E=20.72 H=18.03

P1

OH

-HO2 E=12.24 H=-4.40

IM11

P2

OOH O H

H shift

IM12

OOH O H

E=0.75 H= -17.01

IM13

O O H

-OH E=6.82 H=-19.15

P3

Fig. 2. Secondary reaction scheme from the OH-Flu adduct, IM2, embedded with the potential barrier ΔE (in kcal/mol) and reaction heat ΔH (in kcal/mol) in the presence of O2 and NO. ΔH is calculated at 0 K.

(Herzberg, 1966; Shimanouchi, 1972; Hellwege and Hellwege, 1976; Martin et al., 1996).

3.1. OH additions Previous experimental studies suggested that the addition of OH to C_C in PAHs plays a dominant role compared with H abstractions for anthracene, naphthalene and monocyclic aromatic compounds under the general atmospheric conditions (Ananthula et al., 2006; Lorenz and Zellner, 1983; Perry et al., 1977). Thus, only the addition reaction of Flu with OH was discussed in this paper. Analysis of the molecular structure of Flu shows that there are five different kinds of C atoms, leading to five primary processes of the OH addition. The OH addition reaction schemes embedded with the potential barriers (△E) and reaction heats (△H, 0 K) are presented in Fig. 1. The geometrical structures of the transition states are shown in Fig. 5. All of the OH addition pathways are highly exothermic with low energy barriers, which indicate that they can occur readily under the general atmospheric conditions. As shown in Fig. 1, for pathway 5, a van der Waals complex is formed firstly, whereas other pathways proceed via direct OH addition mechanism. At the BB1K/6-311 + G(3df,2p) level, the potential barriers are 0.57 ~ 1.61 kcal/mol, and the reaction heats are from − 23.39 to − 17.97 kcal/mol. The resulting adducts, OH-Flu, from these OH addition reactions will further react with O2/NO or HO2 as their removal.

3.2. Secondary reactions The discussion above shows that the OH addition to Flu is energetically favorable reaction pathways under the general atmospheric conditions. The five OH-Flu adducts (IM1, IM2, IM3, IM4 and IM5) are important intermediates. The conventional view is that these openshell activated radical intermediates could be further oxidized by ubiquitous O2/NO or HO2. Several important secondary pollutants are produced from the secondary reactions with O2/NO or HO2. All of the possible oxidation pathways were discussed in this paper. Besides, the OH-Flu adducts can react with NO2 to form OH–NO2-Flu adducts via barrierless associations. The OH–NO2-Flu adducts may subsequently undergo unimolecular decomposition to yield nitro-fluoranthenes through the direct loss of water. Our research shows that water plays an important role in the formation of nitro-fluoranthenes. This study has been published in Environ. Sci. Technol (Zhang et al., 2014).

3.2.1. O2 abstraction channels of OH-Flu adducts As shown in Figs. 2 and S1 (Supporting Information), the OH-Flu adducts are unstable energy-rich radicals which can readily react with molecular oxygen to form fluoranthols. The potential barriers of H abstraction from the C\H bonds are 12.49 ~ 9.61 kcal/mol at the BB1K/6-311 + G(3df,2p) level. The transition states lie at 5.48 ~ 13.77 kcal/mol below the sum energy of Flu + OH + O2. The processes are strongly exothermic by 21.09 ~ 27.25 kcal/mol. The overall

642

J. Dang et al. / Science of the Total Environment 490 (2014) 639–646

OH

H OH

+O2 E=24.33 H= 15.95

H OH

+NO

OO

OONO

IM18

OH

E=14.08 H= -35.22

H= -16.18

IM3

H

-NO2

IM20

IM19

O

+O2 E=17.74 -HO2 H=-6.63

+O2 E=35.38 -HO2 H= 16.05

O OH

H O

O

-NO2

+NO

+O2

H=-15.56

E=7.86 H= -10.08

IM5

OONO

OO

OH

P5

OH

OH

IM21

IM22

E=18.08 H=0.66

O OH E=4.94 H=-7.10

IM23

+O2 O O

E=8.65 H=-6.50

-HO2

P6 +O2 OH

E=19.21 H= 8.05

IM4

OO H OH IM25

+O2

OH

IM26

P7 OO H

O

O

+O2

E=11.98 H= 0.39

-HO2

-HO2

OH

+HO 2

E=11.37 H= -24.83

IM24

E=13.94 H= -1.31

OOH

OH

OH H

IM1

E=14.82 H= -32.96

H shift

O OH

E=18.68 H= 9.78

H2O2 IM27

IM28

IM29 +NO H=-17.55

O

O +O2

O H

O

E=8.32 H= -28.97

ONOO H

O

E=15.71 H= 1.29

HO2 P8

- NO2

IM31

IM30

Fig. 3. Secondary reaction scheme from the OH-Flu adducts (IM1, IM3, IM4, IM5), embedded with the potential barrier ΔE (in kcal/mol) and reaction heat ΔH (in kcal/mol) in the presence of O2 and NO.

reactions, Flu + OH + O2 → fluoranthol + HO2, are strongly exothermic by 44.42 ~ 45.63 kcal/mol. This shows that these H abstraction reactions can easily occur. For the intermediate fluoranthen-7-ol generated from the H abstraction reaction of IM2 and O2, it can further react with HO2, followed by twice H abstraction by O2. The calculated profiles of the potential energy surface show that these steps are exoergic and low-barrier reactions. Subsequent reaction of the resulting radical IM10 is the unimolecular decomposition via the cleavage of O\OH bond to form fluoranthenequinone. Calculations indicate that the process has a low potential barrier of 9.30 kcal/mol and is exothermic by

9.89 kcal/mol. Similar products have been detected in the laboratory experiments of naphthalene and OH radical (Kautzman et al., 2009). 3.2.2. O2 addition channels of OH-Flu adducts 3.2.2.1. The formation of dialdehyde and epoxide. The formation schemes of dialdehyde from OH-Flu and O2 are displayed in Figs. 2 and S2 (Supporting Information). For IM2, O2 add to the carbon from the anti-position of hydroxyl, followed by H atom shift via a sixmembered ring transition state. The barrier height of H shift is

J. Dang et al. / Science of the Total Environment 490 (2014) 639–646

643

OH E=33.32 H=19.15

OO

OH

OH +O2

OO

E=13.10 H= -1.80

IM33 OH

IM1

OO

E=21.84 H=-8.41

O

E=22.93 H=-17.83

IM34

OH

OH

O

IM32

O E=11.01 H=0.06

O

IM35

IM36

E=1.63 H=-22.17

E=12.28

OH

+O2 H=-8.12

O

O

OH CHCHOH +

OO

OO

O

O E=54.88 H=47.74

P9

IM37

IM38 +NO

H=-15.51

OH ONOO

OO

OH -NO2 E=31.59 H=-0.68

IM39

O

OO

OH

O E=37.30 H=-59.55

O

O

IM41

IM40

E=4.10 H=-10.50

O O + CHOCHOH P9 Fig. 4. The reaction scheme of the bicyclic peroxy radical embedded with the potential barrier ΔE (in kcal/mol) and reaction heat ΔH (in kcal/mol) in the presence of O2 and NO. ΔH is calculated at 0 K.

calculated to be 21.48 kcal/mol and the process is endothermic by 19.00 kcal/mol. The resulting intermediate is unstable and easy to proceed with a ring-opening reaction. The process has a potential barrier of 0.78 kcal/mol and is strongly exothermic by 17.64 kcal/mol. Through the cleavage of O\OH bond, it ultimately results in OH and dialdehyde. Similar products have been observed in the reaction chamber experiments of gaseous phenanthrene and OH radical (Lee and Lane, 2010). Other OH-Flu adducts (IM1, IM3, IM4 and IM5) can also react with O2 to form dialdehydes via similar mechanisms of IM1. In particular, O2 can add to IM4 from two positions, which are on the left and right side of the carbon linked with the hydroxyl. As shown in Figs. 2 and S2, the process of H shift is the rate-determining step due to its high potential barrier. Based on the study above, the other subsequent reaction of the OHO2-Flu adducts is expected to react with NO, and then the rupture of O\ONO bond occurs. The reaction of NO addition is barrierless and exothermic. For the first three pathways (Route1–3 in Fig. S3), there are two kinds of H abstractions by O2 either from the hydroxyl or from the C\H bond of benzene ring. By comparison, the latter is easily to occur due to the lower potential barrier. It ultimately results in the formation of NO2 and epoxide. Comparison of the reaction scheme presented in Figs. 3 and S3 of Supporting Information shows that the subsequent reactions of IM18 have lower barriers and release more

heat, which indicate that P5 is the dominant product. For the last three reaction routes (Route4–6 in Fig. S3), these processes lead to the formation of NO2 and dialdehydes because of the extra ring-opening reactions. By comparison of the potential barriers and reaction heats of these three reaction routes, the reaction of IM5 is favored over those of IM4. So it is concluded that P6 is the main product. 3.2.2.2. The formation of fluoranthol, fluoranthone and fluoranthenequinone. The formation of fluoranthols from the reaction of OH-Flu adducts and O2 are depicted in Figs. 3 and S2. Similarly, O2 can attack on the C atoms with an unpaired electron in OH-Flu via an addition mechanism, and then the H atom of the benzene ring transfer to the O\O bond via a six-membered ring transition state. Subsequently the C\OOH bond cleavage occurs, leading to the formation of HO2 and fluoranthol. As shown in Fig. 3, the subsequent reaction of IM25 has the lowest barrier and is strongly exothermic, which means this pathway is thermodynamically most favorable. As seen from Figs. S4 and 3, fluoranthols can react with HO2 to produce fluoranthone and H2O2. Subsequently it react with O2/NO and O2, yielding HO2 and fluoranthenequinone. Calculations show that the reaction channel of IM27 can easily proceed due to the low barrier. Similar products have been detected in the experiments of phenanthrene and OH radical (Lee and Lane, 2010).

644

J. Dang et al. / Science of the Total Environment 490 (2014) 639–646

Fig. 5. Transition states for the OH addition reactions of Flu optimized at the BB1K/6-31 + G(d,p) level of theory. Distances are in angstrom.

3.2.2.3. The reaction of bicyclic peroxy radical. Firstly, OH-Flu adducts react with O2 to produce different bicyclic peroxy radicals. As shown in Figs. 4 and S5, two O atoms and the connected C atoms can form a four-membered, a five-membered or a six-membered ring. By comparison of the potential barriers and reaction heats of these reactions, a five-membered ring is easier to form. Due to the structural similarity, just take IM1 as an example. The subsequent reaction of IM34 has two pathways. The first one involves four elementary steps: the O\O bond cleavage, the C\C bond cleavage, the C\C bond rupture and the C\O bond rupture. The calculated profiles of the potential energy surface show that the last step is less likely to occur because of the high potential barrier of 54.88 kcal/mol. The former three steps are energetically favorable and can proceed readily under the general atmospheric conditions. The second pathway includes five elementary reactions: the addition of O2 to form organic peroxy radicals, the reaction of resulting peroxy radicals with NO, the cleavage of O\ONO bond to form alkoxy

radicals, the synchronous cleavage of O\O bond and C\C bond and the rupture of C\C bond to form dialdehyde and CHOCHOH. Using gas chromatography/electron ionization-time-of-flight mass spectrometry (GC/EI-TOFMS), the atmospheric oxidation product of naphthalene, phthaldialdehyde was detected, which is structurally similar to the dialdehyde formed by the reactions of OH-Flu adducts and O2/NO (Kautzman et al., 2009). 3.3. Rate constant calculations The prediction of accurate rate constants of the elementary reactions involved in the degradation of PAHs is crucial for the transport and fate of PAHs in the atmosphere (Keyte et al., 2013). On the basis of the BB1K/ 6-311 + G(3df,2p)//BB1K/6-31 + G(d,p) energies, the rate constants of the elementary reactions involved in the OH-initiated oxidation degradation of Flu were evaluated by using Rice–Ramsperger–Kassel–Marcus

J. Dang et al. / Science of the Total Environment 490 (2014) 639–646 Table 1 The rate constants (cm3 molecule−1 s−1) of the crucial elementary reactions involved in the OH-initiated oxidation degradation of Flu at 298 K and 1 atm. Reactions

Rate constants

Flu + OH → OH-Flu Flu + OH → IM1 Flu + OH → IM2 Flu + OH → IM3 Flu + OH → IM4 Flu + OH → IM5 IM2 + O2 → IM7 IM7 → IM11 IM11 → P2 IM7 → IM12 IM12 → IM13 IM13 → P3 IM18 → IM19 IM19 → IM20 IM5 → IM21 IM21 → IM22 IM22 → IM23 IM4 → IM25 IM25 → IM26 IM26 → P7 IM1 → IM32 IM32 → IM33 IM32 → IM34 IM34 → IM35 IM35 → IM36 IM36 → IM37 IM37 → P9 IM34 → IM38 IM38 → IM39 IM39 → IM40 IM40 → IM41 IM41 → P9

(k) 1.72 × 10−11 (k1) 1.74 × 10−13 (k2) 1.05 × 10−13 (k3) 1.73 × 10−12 (k4) 1.26 × 10−13 (k5) 6.46 × 10−12 2.91 × 10−24 1.95 × 10−10 5.31 × 103 3.32 × 10−6 1.10 × 108 1.48 × 107 4.95 × 10−13 2.68 × 103 1.89 × 10−20 5.39 × 10−13 1.77 2.50 × 10−25 1.96 × 10−13 1.75 × 105 2.61 × 10−24 4.56 × 10−13 1.08 × 10−4 1.73 × 10−4 4.08 × 104 1.10 × 108 1.05 × 10−29 4.83 × 10−5 9.05 1.35 × 10−10 6.38 × 10−9 3.24 × 108

(RRKM) theory (Robinson and Holbrook, 1972) at 298 K and 1 atm. The RRKM method has been used to deal with several reactions (Glowacki et al., 2012; Zhou et al., 2011). The calculated rate constants of some important elementary reactions are presented in Table 1. The individual rate constants for the OH addition to Flu are noted as k1, k2, k3, k4, and k5, respectively. The overall rate constant of the OH addition reaction is noted as k, k = (k1 + k2 + k3 + k4 + k5) × 2. By the calculation, the overall rate constant k is 1.72 × 10− 11 cm3 molecule−1 s− 1 at 298 K and 1 atm. which is consistent with the experimental value of 1.1 × 10−11 cm3 molecule−1 s− 1 at 298 K and 1 atm (Brubaker and Hites, 1998). From the good agreement with the experimental value, it can be inferred that the rate constants of other elementary reactions listed in Table 1 are reasonable. We expect that the RRKM values will assist in the construction of detailed kinetic model describing the transport and fate of Flu in the atmosphere. According to the rate constant of the reaction of Flu with OH radicals and a typical OH concentration (COH) of 9.75 × 105 molecule cm− 3 (Prinn et al., 1995), using the expression:

τOH ¼

1 : kðOHþFluÞ  cOH

The atmospheric lifetime of Flu determined by OH radicals is calculated about 0.69 days. Depending on the literature, Brubaker and Hites estimated that the atmospheric lifetime of Flu is 26 h in an experimental study of gas-phase OH reactions (Brubaker and Hites, 1998). The bias of atmospheric lifetime mainly results from the rate constant difference between the calculated value (1.72 × 10− 11 cm3 molecule− 1 s− 1) and the experimental value (1.1 × 10− 11 cm3 molecule− 1 s− 1).

645

4. Conclusions A theoretical study is presented on the reaction mechanism of OH radical-initiated atmospheric oxidation of Flu. The rate constants were calculated by using the RRKM method. Two specific conclusions can be drawn from this study: (1) The OH-initiated atmospheric oxidation of Flu generates a range of ring-retaining and ring-opening products containing fluoranthols, fluoranthones, fluoranthenequinones, nitro-fluoranthenes, dialdehydes and epoxides. (2) The calculated overall rate constant matches well with the available experimental value. The atmospheric lifetime of Flu by OH radicals is about 0.69 days.

Acknowledgment The work was financially supported by NSFC (National Natural Science Foundation of China, project Nos. 21337001, 21377073 and 21177076), Taishan Grand (No. ts20120522) and Independent Innovation Foundation of Shandong University (IIFSDU, project No. 2012JC030). Appendix A. Supplementary data H abstraction reaction scheme of OH-Flu adducts by O2. O2 addition reaction scheme of OH-Flu adducts. NO addition reaction scheme of OH-Flu-O2 adducts. The formation reaction scheme of fluoranthenequinones. The reaction scheme of the bicyclic peroxy radical. The material is available free of charge via the Internet at http://www.elsevier.com/. Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10. 1016/j.scitotenv.2014.04.134. References Ananthula R, Yamada T, Taylor PH. Kinetics of OH radical reaction with anthracene and anthracene-d10. J Phys Chem A 2006;110:3559–66. Boström CE, Gerde P, Hanberg A, Jernström B, Johansson C, Kyrklund T, et al. Cancer risk assessment, indicators, and guidelines for polycyclic aromatic hydrocarbons in the ambient air. Environ Health Perspect 2002;110:451–88. Brubaker WW, Hites RA. OH reaction kinetics of polycyclic aromatic hydrocarbons and polychlorinated dibenzo-p-dioxins and dibenzofurans. J Phys Chem A 1998;102: 915–21. Chaudhry GR. Biological degradation and bioremediation of toxic chemicals. Portland, OR, USA: Dioscorides Press; 1994. Chen F, Hu W, Zhong Q. Emissions of particle-phase polycyclic aromatic hydrocarbons (PAHs) in the Fu Gui-shan Tunnel of Nanjing, China. Atmos Res 2013;124: 53–60. Christie S, Raper D, Lee DS, Williams PI, Rye L, Blakey S, et al. Polycyclic aromatic hydrocarbon emissions from the combustion of alternative fuels in a gas turbine engine. Environ Sci Technol 2012;46(11):6393–400. EU. Directive 2004/107/EC of the European Parliament and of the Council of 15 December 2004 relating to arsenic, cadmium, mercury, nickel and polycyclic aromatic hydrocarbons in ambient air. Off J L 2005;23. [of 26.01.2005]. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, Cheeseman JR, et al. Gaussian 03, Revision A.1. Pittsburgh, PA: Gaussian Inc.; 2003. Galarneau E, Bidleman TF, Blanchard P. Seasonality and interspecies differences in particle/gas partitioning of PAHs observed by the Integrated Atmospheric Deposition Network (IADN). Atmos Environ 2006;40:182–97. Glowacki DR, Liang CH, Morley C, Pilling MJ, Robertson SH. MESMER: an open-source master equation solver for multi-energy well reactions. J Phys Chem A 2012;116: 9545–60. Hellwege KH, Hellwege AM. Landolt–Bornstein: group II: atomic and molecular physics. Structure Data of Free Polyatomic Molecules, vol. 7. Berlin: SpringerVerlag; 1976. Herzberg G. Electronic spectra and electronic structure of polyatomic molecules. New York: Van Nostrand; 1966. Kautzman KE, Surratt JD, Chan MN, Chan AWH, Hersey SP, Chhabra PS, et al. Chemical composition of gas- and aerosol-phase products from the photooxidation of naphthalene. J Phys Chem A 2009;114:913–34. Keyte IJ, Harrison RM, Lammel G. Chemical reactivity and long-range transport potential of polycyclic aromatic hydrocarbons — a review. Chem Soc Rev 2013; 42:9333–91.

646

J. Dang et al. / Science of the Total Environment 490 (2014) 639–646

Lee JY, Lane DA. Formation of oxidized products from the reaction of gaseous phenanthrene with the OH radical in a reaction chamber. Atmos Environ 2010;44: 2469–77. Lorenz K, Zellner R. Kinetics of the reactions of hydroxyl radicals with benzene, benzened6 and naphthalene. Ber Bunsen Ges 1983;87:629–36. Martin JML, El-Yazal J, Francois J-P. Structure and vibrational spectrum of some polycyclic aromatic compounds studied by density functional theory. 1. naphthalene, azulene phenanthrene, and anthracene. J Phys Chem 1996;100:15358–67. Monte MJS, Notario R, Pinto Sónia P, Ferreira Lobo, Ana IMC, da Silva Ribeiro, et al. Thermodynamic properties of fluoranthene: an experimental and computational study. J Chem Thermodyn 2012;49:159–64. Perry RA, Atkinson R, Pitts JN. Kinetics and mechanism of the gas phase reaction of hydroxyl radicals with aromatic hydrocarbons over the temperature range 296–473 K. J Phys Chem 1977;81:296–304. Prinn RG, Weiss RF, Miller BR, Huang J, Alyea FN, Cunnold DM, et al. Atmospheric trends and lifetime of CH 3 CCI 3 and global OH concentrations. Science 1995; 269:187–92. Qu XH, Zhang QZ, Wang WX. Mechanism for OH-initiated photooxidation of naphthalene in the presence of O2 and NOx: a DFT study. Chem Phys Lett 2006;429: 77–85. Ramírez N, Cuadras A, Rovira E, Marcé RM, Borrull F. Risk assessment related to atmospheric polycyclic aromatic hydrocarbons in gas and particle phases near industrial sites. Environ Health Perspect 2011;119:1110–6. Robinson PJ, Holbrook KA. Unimolecular reactions. John Wiley & Sons; 1972.

Seinfeld JH, Pandis SN. Atmospheric chemistry and physics; Wiley-Interscience: New York; 2006 [ISBN 9471720186]. Shen HZ, Huang Y, Wang R, Zhu D, Li W, Shen GF, et al. Global atmospheric emissions of polycyclic aromatic hydrocarbons from 1960 to 2008 and future predictions. Environ Sci Technol 2013;47(12):6415–24. Shimanouchi, T. Tables of Molecular Vibrational Frequencies, Consolidated Volume 1, 1972, NSRDS NBS-39. Terzi E, Samara C. Gas-particle partitioning of polycyclic aromatic hydrocarbons in urban, adjacent coastal, and continental background sites of western Greece. Environ Sci Technol 2004;38:4973–8. Wetzel A, Alexander T, Brandt S, Haas R, Werner D. Reduction by fluoranthene of copper and lead accumulation in Triticum aestivum L. Environ Contam Toxicol 1994;53: 856–62. Xue WL, Warshawsky D. Metabolic activation of polycyclic and heterocyclic aromatic hydrocarbons and DNA damage: a review. Toxicol Appl Pharmacol 2004;206(1): 73–93. Zhang QZ, Gao R, Xu F, Zhou Q, Jiang GB, Wang T, et al. Role of water molecule in the gasphase formation process of nitrated polycyclic aromatic hydrocarbons in the atmosphere. 2014;48(9):5051–7. Zhou J, Chen JW, Liang C-H, Xie Q, Wang Y-N, Zhang SY, et al. Quantum chemical investigation on the mechanism and kinetics of PBDE photooxidation by OH: a case study for BDE-15. Environ Sci Technol 2011;45:4839–45.