Gas- and Particle-Phase Products from the Chlorine-Initiated

Article pubs.acs.org/JPCA

Gas- and Particle-Phase Products from the Chlorine-Initiated Oxidation of Polycyclic Aromatic Hydrocarbons Matthieu Riva,†,‡,# Robert M. Healy,§ Pierre-Marie Flaud,†,‡ Emilie Perraudin,†,‡ John C. Wenger,*,§ and Eric Villenave*,†,‡ †

Univ. Bordeaux, EPOC, UMR 5805, F-33405 Talence cedex, France CNRS, EPOC, UMR 5805, F-33405 Talence cedex, France § Department of Chemistry and Environmental Research Institute, University College Cork, Cork, Ireland ‡

ABSTRACT: The chlorine atom (Cl)-initiated oxidation of three polycyclic aromatic hydrocarbons (PAHs; namely, naphthalene, acenaphthylene, and acenaphthene) was investigated. Experiments were performed in an atmospheric simulation chamber using a proton transfer reaction time-of-flight mass spectrometer (TOF-MS) and an aerosol TOF-MS to characterize the oxidation products in the gas and particle phases, respectively. The major products identified from the reaction of Cl atoms with naphthalene were phthalic anhydride and chloronaphthalene, indicating that H atom abstraction and Cl addition reaction pathways are both important. Acenaphthenone was the principal product arising from reaction of Cl with acenaphthene, while 1,8-naphthalic anhydride, acenaphthenone, acenaphthenequinone, and chloroacenaphthenone were all identified as products of acenaphthylene oxidation, confirming that the cylcopenta-fused ring controls the reactivity of these PAHs toward Cl atoms. Possible reaction mechanisms are proposed for the formation of these products, and favored pathways have been suggested. Large yields of secondary organic aerosol (SOA) were also observed in all experiments, and the major products were found to undergo significant partitioning to the particle-phase. This work suggests that Cl-initiated oxidation could play an important role in SOA formation from PAHs under specific atmospheric conditions where the Cl atom concentration is high, such as the marine boundary layer.

■

produce oxygenated and nitro compounds,8−10 which can partition into the particle phase and participate in SOA formation.11−15 Recently, Riva et al.16 showed the high reactivity of some PAHs with chlorine atoms and the possible importance of this reaction in coastal areas and the marine boundary layer, where elevated concentrations of Cl atoms can exist.17 However, apart from a few studies on biogenic compounds18 and toluene,19,20 SOA formation from Cl atominitiated oxidation processes has largely been unexplored. In this study, we present results from simulation chamber experiments on SOA formation from the gas-phase reaction of Cl atoms with naphthalene, acenaphthene, and acenaphthylene. To the best of our knowledge, this work is the first study to focus on SOA formation from the oxidation of PAHs initiated by Cl atoms. Gaseous and particulate phase products were tentatively identified and used to elucidate reaction mechanisms for the reactions. SOA yields were also determined to evaluate the potential importance of chlorine atom chemistry in SOA formation from PAHs.

INTRODUCTION Atmospheric aerosol plays a major role in many environmental processes, such as the scattering and absorption of solar and terrestrial radiation, which can affect climate on a regional and global scale.1,2 Aerosols also participate in heterogeneous chemical reactions, thereby altering the abundance and distribution of atmospheric trace gases and pollutants.3 Atmospheric particulate matter influences human health and has impacts on the human respiratory and cardiovascular systems.4 Currently, it is recognized that the largest mass fraction of fine atmospheric aerosol is generally organic carbon, which is often dominated by secondary organic aerosol (SOA) formed from the oxidation of volatile organic compounds (VOCs). However, current models generally predict less SOA mass than typically observed.1,3 A large part of this underestimation is due to the incomplete representation of some SOA precursors in models, such as intermediate volatility organic compounds (IVOCs) like long-chain alkanes and polycyclic aromatic hydrocarbons (PAHs).5 Much of the current effort in the research community is therefore focused on trying to identify the missing or misrepresented sources of SOA. PAHs are emitted into the atmosphere from incomplete combustion processes and have been identified as major components in traffic and wood burning emissions.6,7 The PAHs with less than four aromatic rings mainly exist in the gas phase and can undergo gas-phase oxidation processes to © XXXX American Chemical Society

■

EXPERIMENTAL SYSTEM All experiments were performed in the 3910 L atmospheric simulation chamber at University College Cork, previously Received: May 13, 2015 Revised: October 14, 2015

A

DOI: 10.1021/acs.jpca.5b04610 J. Phys. Chem. A XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry A

Table 1. Proposed Structures for Ions Identified by PTR-TOF-MS during the Gas-Phase Oxidation of Naphthalene Initiated by Cl Atoms

described in detail elsewhere.21 Briefly, the chamber is a cylinder consisting of an FEP-Teflon foil tube closed with aluminum end-plates covered with FEP foil. The chamber is surrounded by 16 lamps (Philips TL12, 40 W) with an emission maximum at 310 nm and 12 lamps (Philips TL05, 40 W) with an emission maximum at 360 nm. Before each experiment, the chamber was cleaned by photolysis of oxalyl chloride to produce Cl atoms22 that react with any organic impurities present. The chamber was then flushed with dried purified air until the particle number concentration was less than 200 particles per cubic centimeter. Moreover, flushing reduced the NOx and nonmethane hydrocarbon mixing ratios below 10 ppb. The temperature and water concentration in the chamber were monitored by a dew point meter (Vaisala DM70). The relative humidity in the chamber was typically less than 1%. The experiments were performed at room temperature (293 ± 2 K) and atmospheric pressure. The chamber was equipped with fans to ensure rapid mixing of reactants before turning the lights on. The PAHs used in this study were naphthalene (SigmaAldrich, 99%), acenaphthylene (Sigma-Aldrich, 99%), and acenaphthene (Sigma-Aldrich, 99%). They were introduced into the chamber by flowing dry purified air throughout a heated Pyrex glass bulb containing a known amount of the solid compound. Initial concentrations of reactants during the experiments were (in micrograms per cubic meter): [naphthalene] = 1.76 × 103 and 1.82 × 103 (without and with OH scavenger, respectively); [acenaphthylene] = 2.10 × 103, and [acenaphthene] = 2.00 × 103. One experiment was performed for acenaphthylene and acenaphthene oxidation, while two experiments were conducted using naphthalene. The PAHs and gas-phase oxidation products were monitored during the experiments using a proton transfer reaction time-of-flight mass spectrometer (PTR-TOF-MS, Kore Technology Ltd.).

Details of the instrument and its operating principle are given in Cappelin et al.23 Briefly, H3O+ is produced in a hollow cathode ion source and reacts with organic compounds (M) that have a higher proton affinity than H2O to generate positively charged ions (M + H)+, which are subsequently detected using a time-of-flight mass spectrometer. The PTRTOF-MS was operated over the m/z range of 0−300 using a sampling time of 1 min. The decay of the PAHs was monitored by following the protonated molecular ions: m/z 129 (naphthalene); m/z 153 (acenaphthylene); m/z 155 (acenaphthene). Chlorine atoms were generated by the photolysis of oxalyl chloride (C2O2Cl2, Sigma-Aldrich, 99%), which is recognized as a clean source of Cl atoms.22,24 Oxalyl chloride was injected into the chamber by flowing dry purified air through a heated Pyrex glass bulb containing a known amount (∼30 μL) of the liquid compound, when the PAH concentration had stabilized. The concentration of Cl atoms generated in the experiments was calculated from the measured decay of the PAHs.25 Using the rate coefficients obtained by Riva et al.16 and assuming a steady state of Cl atoms, an average concentration of (1.05 ± 0.48) × 107 molecules per cubic centimeter was determined. Because of the relatively low reactivity of naphthalene with chlorine atoms,16 30 ppm of benzene (Fluka, 99.9%) was added to the chamber to scavenge any hydroxyl (OH) radicals produced during the reaction, as in previous studies.16,26 The aerosol size distribution, number, and mass concentrations were determined using a scanning mobility particle sizer (SMPS, TSI model 3034). The chemical composition of SOA was investigated by online detection using an aerosol time-of-flight mass spectrometer (ATOFMS, TSI model 3800). The instrument is described in detail elsewhere.27 Briefly, single particles are sampled through a critical orifice and focused into a tight beam in the aerodynamic lens before transmission to the B


Article


Figure 1. Temporal profiles of major ions identified by PTR-TOF-MS in the gas-phase oxidation of naphthalene initiated by Cl atoms.

namely, Cl addition to the aromatic ring and H atom abstraction from the ring. The addition pathway produces a Cl-naphthalene adduct, which can subsequently react with molecular oxygen to produce chloronaphthalene, Scheme 1. Addition at the C1 and C2 positions is possible, resulting in the formation of 1-chloronaphthalene and 2-chloronaphthalene, respectively. Although the PTR-TOF-MS technique cannot differentiate between isomers, previous studies suggest that, as for OH and NO3,30,31,35 addition of Cl at the C1 position is strongly favored.33 Moreover, the formation of chloronaphthalene as a major product highlights the importance of the addition pathway. This is in contrast to the (Cl + benzene) reaction where the addition pathway is very minor because the Cl-benzene adduct appears to undergo decomposition to reform Cl and benzene rather than react with O2.34 The results of this work strongly suggest that the Cl-naphthalene adduct is much more stable and does not revert so readily back to reactants. Support for this assertion comes from the measured rate coefficient for reaction of Cl with naphthalene, which is 4 orders of magnitude higher than for benzene.16 Interestingly, this behavior is consistent with that observed for the HO-benzene and HO-naphthalene adducts, where the thermal decomposition of the latter species is ∼1000 times slower.36 The formation of the other major product, phthalic anhydride, can also be explained by the addition pathway. As shown in Scheme 1, addition of O2 to the Cl-naphthalene adduct results in the formation of a peroxy radical that could undergo cyclization and further reaction with O2 followed by ring-opening to yield glyoxal and a chlorinated radical. Oxidation of this radical produces a chlorinated oxy radical, which, in principle, could further react with oxygen to yield 2formylbenzoyl chloride (C8H5ClO2). However, no products with this molecular mass (168 g mol−1) are observed, and inspection of the literature indicates that chlorinated oxy radicals of this type (RCHClO•) tend to undergo HCl elimination.37 As shown in Scheme 1, elimination of HCl would result in an acyl radical, which can react with oxygen and then undergo cyclization to form phthalic anhydride. Although this proposed mechanism is fairly complex, the bicyclic peroxy radical route regularly features in the atmospheric oxidation

sizing region. The velocity of the particles is measured using two continuous wave diode-pumped Nd:YAG lasers operating at 532 nm. The time between these two scattering events is used to obtain the aerodynamic size. The particle is focused in the ionization region of the instrument where desorption/ ionization is performed by a Nd:YAG laser at 266 nm. The resulting ions were analyzed using a time-of-flight mass spectrometer, with the positive mode supplying the best diagnostic information. Usually, the laser is operated at an energy of ∼one millijoule per pulse. However, as described elsewhere,28,29 low pulse energies were also employed in this work to minimize molecular fragmentation.

■

RESULT AND DISCUSSION Naphthalene. PTR-TOF-MS data obtained for the (Cl + naphthalene) oxidation reactions, performed in the presence of excess benzene as OH scavenger, showed six peaks at m/z 147, 149, 163, 165, 183, and 197, and they are reported in Table 1. The temporal profiles of the three main peaks, at m/z 149, 163, and 165, are shown in Figure 1, along with the decay of the naphthalene signal. On the basis of existing knowledge of the gas-phase chemistry of naphthalene, the peak at m/z 149 is assigned to the protonated molecular ion (M + H)+ of phthalic anhydride (C8H4O3, 148 g mol−1). Phthalic anhydride is a wellknown oxidation product of naphthalene, which has also been detected in laboratory studies of the gas-phase OH-initiated oxidation of the PAH.8,9,30,31 The peaks at m/z 163 and 165 exhibit exactly the same temporal profile with a signal intensity ratio of 3:1 and are thus attributed to protonated molecular ions of chloronaphthalene containing the 35Cl and 37Cl isotopes (C10H735Cl and C10H737Cl), respectively.32 Both 1-chloronaphthalene and 2-chloronaphthalene have previously been observed as products resulting from the chlorination of naphthalene in the vapor phase.33 As outlined below, mechanistic considerations also indicate that phthalic anhydride and chloronaphthalene are the only plausible oxidation products with molecular masses of 148 and 162/164 g mol−1. To the best of our knowledge, no mechanism has been proposed in the literature for the reaction of Cl atoms with naphthalene or any other PAHs. On the basis of the (Cl + benzene) reaction,34 two reaction pathways are possible, C


Article

The Journal of Physical Chemistry A Scheme 1. Proposed Mechanism for the Cl Atom-Initiated Oxidation of Naphthalenea

a

Observed products are boxed and reported in Table 1. The phase of the observed products (gas = g; particle = p) is mentioned in the left lower corner.

mechanisms of aromatic compounds36 and has also been proposed by Kautzman et al.9 as one possible pathway for the formation of analogous C8 ring-opening products in the OH + naphthalene oxidation reaction. There are two further aspects relating to the formation of phthalic anhydride that should be discussed. First, in the OHinitiated oxidation of naphthalene, phthalic anhydride is believed to be a second-generation product resulting from further oxidation of phthaldialdehyde.9,31 However, this is clearly not the case here. As shown in Figure 1, the temporal profile of phthalic anhydride resembles that of a primary product, and there is also no evidence for phthaldialdehyde among the gas-phase products identified by PTR-TOF-MS. Second, the possibility that phthalic anhydride is formed from the H atom abstraction pathway must be considered. H atom abstraction would result in the formation of HCl and a

naphthyl radical, which would undergo addition of molecular oxygen. The chemistry of the resulting peroxy radical could proceed through the bicyclic peroxy radical route in a fashion analogous to that shown in Scheme 1. However, with a H atom replacing Cl in the Scheme 1, the molecular product would be phthaldialdehyde, which was not observed. Alternatively the peroxy radical could react with other peroxy radicals to produce the naphthoxy radical. The chemistry of this radical has been studied before,38 and at the low levels of NOx present in these experiments, the main reaction pathways for naphthoxy are dimerization and reaction with oxygen and HO2 radicals. None of these reactions result in the generation of ring-opening products in appreciable yields.38 The signal intensities of the three other product peaks at m/z 147, 183, and 197 are at least an order of magnitude smaller than those discussed above. The peak at m/z 183 exhibits a D


Article


Figure 2. Average mass spectra (in the positive ion mode) for SOA formed from the reaction of Cl atoms with naphthalene at different laser pulse energies (red at 1 mJ; green at 0.2 mJ).

Figure 3. Temporal profiles of major ions identified by PTR-TOF-MS in the gas-phase oxidation of acenaphthene initiated by Cl atoms (red scale corresponds to the acenaphthenone curve).

temporal profile that is typical of a secondary product and is assigned to chlorophthalic anhydride (C8H5O235Cl, 182 g mol−1), which can be formed from further reaction of phthalic anhydride with Cl atoms, Scheme 1. Similarly, the peak at m/z 197 can be attributed to dichloronaphthalene (C10H635Cl2, 196 g mol−1), formed from further reaction of chloronaphthalene with Cl atoms. Note that the corresponding peaks for these products containing the 37Cl isotope were not observed due to signal limitations. The final peak, at m/z 147, corresponds to a product with molecular mass of 146 g mol−1. Two products with this molecular mass have been previously identified in the photooxidation of naphthalene,8 namely, 1,2-benzopyrone (coumarin) and 1,3-indene-dione. Although it is not possible to distinguish between these two compounds using PTR-TOFMS, we prefer to assign this peak to 1,2-benzopyrone because it

has a more plausible formation mechanism. A possible formation pathway is presented in Scheme 1, which proceeds via ring-opening of a chlorinated naphthoxy radical followed by HCl elimination and cyclization. Similar cyclization processes have been proposed to explain the formation of products from the OH-oxidation of phthaldialdehyde.9 The composition of SOA particles produced from the (Cl + naphthalene) reaction was characterized using ATOFMS. The average mass spectra (positive mode only) obtained using laser pulse energies of 1 and 0.2 mJ are shown in Figure 2. At 1 mJ laser pulse energy the mass spectrum is dominated by the following ions: 37 (C3H)+, 39 (C3H3)+, 51 (C4H3)+, 63 (C5H3)+, 74 (C6H2)+, 89 (C7H5)+, 115 (C9H7)+, and 128 (C10H8)+.38 The smaller ions are typical of aromatic species but not particularly helpful for product identification, while the fragments at m/z 115 and 128 can be considered as diagnostic E


Article


Table 2. Proposed Structures for Ions Identified by PTR-TOF-MS during the Gas-Phase Oxidation of Acenaphthene Initiated by Cl Atoms

thene (C12H935Cl, 188 g mol−1), which exhibits a weak signal. A summary of the proposed products and their identifying ions is listed in Table 2. The reaction of Cl atoms with acenaphthene can, in principle, proceed via three possible pathways; Cl addition to the aromatic ring, H atom abstraction from the aromatic ring, H atom abstraction from the saturated cylcopenta-fused ring. However, the rate coefficient for reaction of Cl atoms with acenaphthene is ∼100 times faster than that for naphthalene,16 indicating that H atom abstraction from the cyclopenta-fused ring is by far the dominant reaction pathway for Cl + acenaphthene. As shown in Scheme 2, four of the products observed by PTR-TOF-MS can be explained from this reaction pathway. The initial attack by Cl atoms is followed by addition of oxygen to produce a peroxy radical. Subsequent reaction with other peroxy radicals proceeds via two pathways: (i) the molecular channel, which results in equimolar amounts of acenaphthenone and acenaphthenol, and (ii) the radical channel, which produces an oxy radical, which can react further with O2 to form acenaphthenone as the sole product. Given that acenaphthenone is produced in much higher yield than acenaphthenol, the radical channel appears clearly dominant. An alternative pathway, that is, C−C bond cleavage of the oxy radical followed by further oxidation, could lead to the formation of naphthalene-1,8-dicarbaldehyde and 1,8-naphthalic anhydride, which are observed as minor products. As shown in Scheme 2, 1,8-naphthalic anhydride can also be formed as a secondary product from the Cl-initiated oxidation of naphthalene-1,8-dicarbaldehyde. Finally, chloroacenaphthenone is produced via Cl addition to the aromatic ring of acenaphthene, which seems to be a very minor pathway. The reaction mechanism shown in Scheme 2 is based on that previously proposed for OH and NO3 attack at the cyclopentafused ring.14 H atom abstraction at this site in acenaphthene

fragmentation ions for substituted naphthalene products. At 0.2 mJ laser energy the intensity of the smaller fragment ions is reduced, and the peaks for all larger ions increase substantially. This effect has been observed in previous single-particle desorption/ionization mass spectrometry studies and is attributed to less fragmentation of the aromatic compounds.28,38 On the basis of the results presented above, the peaks clustered around m/z 148 and 162 in the lower-energy mass spectrum are assigned to the main oxidation products phthalic anhydride and chloronaphthalene, respectively, which have undergone partitioning to the particle phase. However, the large peak at m/z 134 is not so easily assigned. It is probably due to (C8H6O2)+, and although some oxygenated aromatics (e.g., phthalide, phthaldiadehyde) possess this molecular formula, they were not observed as gas-phase oxidation products. Conventional fragmentation pathways for phthalic anhydride and chloronaphthalene are also very unlikely to produce the large peak at m/z 134. Thus, it is possible that this peak is linked to a species that is either produced during the laser desorption/ionization process or from chemical reactions occurring in or on the surface of the SOA. Finally, there is little evidence for the presence of molecular ions for the three other oxidation products identified in the gas phase, although these compounds were of course only produced in small amounts. Acenaphthene. PTR-TOF-MS data obtained for the (Cl + acenaphthene) oxidation reactions shows one very dominant peak at m/z 169, along with much smaller peaks at m/z 171, 185, 189, and 199. The temporal profiles for selected peaks are presented in Figure 3, along with the decay of the acenaphthene signal. Four of the peaks correspond to molecular masses of products previously observed in smog chamber experiments of the OH- and NO3-initiated oxidation of acenaphthene.14,39,40 The fifth product, at m/z 189, is assigned to the protonated molecular ion of chloroacenaphF


Article

The Journal of Physical Chemistry A Scheme 2. Proposed Mechanism for the Cl Atom-Initiated Oxidation of Acenaphthenea

a Observed products are boxed and reported in Table 2. The phase of the observed products (gas = g; particle = p) is mentioned in the left lower corner.

115 (C9H7)+, 127 (C10H7)+, and 139 (C11H7)+.38 As for naphthalene SOA, these smaller ions are not useful for diagnostic purposes since they could be attributed to any of the products. However, some of the peaks at higher masses can be attributed to specific oxidation products. The peaks centered at m/z 168, 170, and 198 are assigned to acenaphthenone, acenaphthenol, and 1,8-naphthalic anhydride, respectively. All three compounds were also observed as particle-phase products resulting from the OH-initiated oxidation of acenaphthene.14 The observation of a very large peak at m/z 182 is especially interesting since a product with this molecular mass was not identified by PTR-TOF-MS. Several possible explanations exist. First, it is possible that this peak is due to a major fragment ion from a higher MW product, that is, naphthalene-1,8dicarbaldehyde or 1,8-naphthalic anhydride. Both compounds

will produce the same radicalregardless of oxidant typeand is thus expected to yield the same products. This is confirmed for both OH- and NO3-initiated oxidation where the product distributions are very similar to those observed in this work: acenaphthenone and acenaphthenol are identified as major and minor gas-phase products, respectively, along with trace levels of naphthalene-1,8-dicarbaldehyde.14 However, note that other products (e.g., dialdehyde or epoxide),14 attributed to OH addition on the aromatic ring and/or reactions with NOx, were not detected in Cl atom oxidation experiments. ATOFMS data for the (Cl + acenaphthene) reaction is presented in Figure 4. The average mass spectrum (positive mode) obtained using a laser pulse energy of 0.2 mJ exhibits characteristic aromatic fragment ions at m/z 39 (C3H3)+, 51 (C4H3)+, 63 (C5H3)+, 74 (C6H2)+, 77 (C6H5)+, 89 (C7H5)+, G


Article


species formed by chemical reactions occurring in or on the surface of the SOA. Acenaphthylene. Five oxidation products from the reaction of Cl with acenaphthylene were identified by PTRTOF-MS. Temporal profiles for the products are shown in Figure 5. The most abundant product peaks at m/z 169 and 199 are assigned to protonated molecular ions of 1,8-naphthalic anhydride and acenaphthenone, respectively, which were also oxidation products of the Cl + acenaphthene reaction. The peak at m/z 183 is attributed to acenaphthenequinone (C12H6O2, 182 g mol−1), which along with the two main products was also previously observed in laboratory studies of the OH- and NO3-initiated oxidation of acenaphthylene.15,39 The peak at m/z 203 is accompanied by another with approximately one-third the intensity at m/z 205, and these are therefore assigned to chloroacenaphthenone. Finally, the weak signal at m/z 187 (not shown in Figure 5) is assigned to chloroacenaphthylene. A summary of the proposed products and their identifying ions is listed in Table 3. As for acenaphthene, reaction of Cl with acenaphthylene is expected to occur predominantly via initial attack at the cyclopenta-fused ring. However, the rate coefficient for reaction of Cl with acenaphthylene is higher than its saturated counterpart, indicating that, as for other alkenes, addition of Cl to the carbon−carbon double bond is expected to be the dominant reaction pathway.16 As shown in Scheme 3, Cl addition to the cyclopenta-fused ring, followed by oxidation, produces a peroxy radical, which can react further via the molecular channel to yield chloroacenaphthenone and chloroacenaphthenol. It is possible that the peak observed at m/z 205 by PTR-TOF-MS, and identified as the 37Cl isotopologue of chloroacenaphthenone, could also correspond to chloroacenaphthenol containing 35Cl. However, as noted above, the relative ion intensity of the peaks at m/z 203 and 205 is perfectly consistent with their assignment to chloroacenaphthenone.32 The molecular channel is thus deemed to be minor, and the preferred route for the peroxy radical reaction is thus via the radical channel to produce a chlorinated oxy radical. Reaction of this oxy radical with O2 yields chloroacenaphthenone as the sole product, while C−C

Figure 4. Average mass spectra (in the positive ion mode) for SOA formed from the reaction of Cl atoms with acenaphthene using a laser pulse energy of 0.2 mJ.

previously exhibited a very high degree of partitioning to the particle phase in the OH-initiated oxidation of acenaphthene,14 and the relatively low signals observed by PTR-TOF-MS suggests that an extensive amount of partitioning also occurred in this study. Although the fragmentation routes required to form a species with m/z 182 (loss of 2 amu from naphthalene1,8-dicarbaldehyde and 16 amu from 1,8-naphthalic anhydride) were not observed using electron impact and chemical ionization,14 it is possible that the laser desorption ionization process used in the ATOFMS may result in different fragmentation. A second possibility is that the peak at m/z 182 is due to a product formed in the gas phase that undergoes extensive partitioning to the particle phase. One such compound is acenapthenequinone (C12H6O2, 182 g mol−1), which could be formed as a secondary product from the Clinitiated oxidation of acenaphthenone. As shown in Figure 3, acenaphthenone signal decreases considerably after ca. 2000 s of reaction time, providing some support for this hypothesis. A third possibility is that the peak at m/z 182 is due to some

Figure 5. Temporal profiles of major ions identified by PTR-TOF-MS in the gas-phase oxidation of acenaphthylene initiated by Cl atoms. H


Article


Table 3. Proposed Structures for Ions Identified by PTR-TOF-MS during the Gas-Phase Oxidation of Acenaphthylene Initiated by Cl Atoms

attributed to specific oxidation products. The peaks centered around m/z 168, 182, and 198 are assigned to acenaphthenone, acenaphthenequinone, and 1,8-naphthalic anhydride, respectively. All three of these compounds have previously been detected as particle-phase products in the OH- and NO3initiated oxidation of acenaphthylene.15 Secondary Organic Aerosol Formation Yields. Using a semiempirical model for SOA formation based on the gasparticle partitioning equilibrium of semi-volatile products, the SOA yields (Y) were determined to be (0.91 ± 0.05), (0.85 ± 0.07), and (0.98 ± 0.09) for naphthalene, acenaphthylene, and acenaphthene, respectively. The indicated uncertainties (2σ) correspond to scatter in particle volume measurements. It should be pointed out that the presence of benzene as OH radical scavenger during the oxidation of naphthalene is highly unlikely to impact on the SOA yield since the reported SOA yield under low NOx conditions is low (1.6 to 2.9%),42 and in this study, only ∼3% of the benzene was depleted during the oxidation of naphthalene. The aerosol mass was calculated using volume concentrations measured by SMPS and assuming a particle density of 1.4 g cm−3.11 The high concentrations used in this work will probably enhance partitioning of semi-volatile species to the particle phase resulting in SOA yields that are larger than might be expected under more realistic atmospheric conditions. Nevertheless, the high yields determined in this work further illustrate the important potential of PAHs to form SOA from their oxidation by Cl, as well as from OH radicals. Indeed, the SOA yields obtained in this work appear to be up to three times higher than those determined from OH initiated oxidation.11 However, additional studies are needed to better characterize SOA yields under atmospherically realistic conditions.

bond cleavage followed by further oxidation and elimination of HCl could lead to the formation of 1,8-naphthalic anhydride. Thus, Cl addition to the cyclopenta-fused ring can explain the formation of two of the observed products. As for the other PAHs, it is proposed that the minor product chloroacenaphthylene is generated via Cl addition to the aromatic ring. However, possible mechanisms for the formation of the remaining two products (acenaphthenone and acenaphthenequinone) are not so straightforward. Acenaphthenone retains all of the original H atoms, and the Cl addition pathway therefore seems the most plausible formation route. However, Cl atom elimination and H atom transfer reactions are both required, and considering the structures of the radicals shown in Scheme 3, this seems unlikely. Interestingly, acenaphthenone was observed as a product in previous studies of the OH- and NO3-initiated oxidation of acenaphthylene,15,39 as well as during the photolysis of the PAH in liquid water,41 but suitable formation mechanisms have yet to be proposed. It is also difficult to understand how the Cl addition pathway can lead to acenaphthenequinone. As a result, it is proposed that this product is formed via H atom abstraction from the cyclopentafused ring, Scheme 3. It is possible that this pathway may also contribute to the formation of the major product, 1,8naphthalic anhydride. Although H atom abstraction is not normally considered to be a major pathway for reaction of Cl with alkenes, the fact that acenaphthene is almost as reactive toward Cl atoms as acenaphthylene provides some support for the existence of this pathway.16 ATOFMS data for the (Cl + acenaphthylene) reaction is shown in Figure 6. The average mass spectrum (positive mode) obtained using a laser pulse energy of 0.2 mJ is virtually identical to that measured for the (Cl + acenaphthene) reaction. This is not surprising given that these reactions yield products that are either identical or very similar in structure. As for the other PAHs, some of the peaks at higher masses can be I


Article

The Journal of Physical Chemistry A Scheme 3. Proposed Mechanism for the Cl Atom-Initiated Oxidation of Acenaphthylenea

a

Observed products are boxed and reported in Table 3. The phase of the observed products (gas = g; particle = p) is mentioned in the left lower corner.

■

CONCLUSION In this work, major products from the reaction of Cl atoms with naphthalene, acenaphthene, and acenaphthylene have been determined, and reactions mechanisms are proposed. The formation of large amounts of phthalic anhydride and chloronaphthalene during the oxidation of naphthalene indicates that H atom abstraction and Cl addition reaction pathways are both important. Interestingly, this behavior is different than other oxidation processes (e.g., OH or NO3 radicals) where the abstraction pathway is considered as negligible. Acenaphthenone was the principal product arising from reaction of Cl with acenaphthene, indicating that (Cl + acenaphthene) reaction is governed by H-abstraction from the saturated cyclopenta-fused ring, which is consistent with measured rate coefficient for this reaction.16 The reaction with Cl atoms with acenaphthylene yielded chloroacenaph-

thenone, attributed to addition of Cl to the carbon−carbon double bond in the unsaturated cyclopenta-fused ring. However, identification of 1,8-naphthalic anhydride, acenaphthenone, and acenaphthenequinone as products suggests that H atom abstraction is also important, which is not normally considered to be a major pathway for reaction of Cl with alkenes. Reaction mechanisms proposed in this work confirm that the cylcopenta-fused ring controls the reactivity of acenaphthene and acenaphthylene toward Cl atoms, which is consistent with other oxidation processes. The oxidation of naphthalene, acenaphthene, and acenaphthylene formed large yields of SOA suggesting that reaction of Cl atoms with PAHs could contribute to anthropogenic SOA formation. This atmospheric process might be important in specific atmospheres, such as the marine boundary layer or industrial areas. For instance, using a maximum Cl concenJ


Article


secondary organic aerosol: current and emerging issues. Atmos. Chem. Phys. 2009, 9, 5155−5236. (4) Elder, A.; Oberdorster, G. Translocation and effects of ultrafine particles outside of the lung. Clin. Occup. Environ. Med. 2005, 5, 785− 796. (5) Tkacik, D. S.; Presto, A. A.; Donahue, N. M.; Robinson, A. L. Secondary organic aerosol formation from intermediate-volatility organic compounds: cyclic, linear, and branched alkanes. Environ. Sci. Technol. 2012, 46, 8773−8781. (6) Lammel, G.; Klanova, J.; Ilic, P.; Kohoutek, J.; Gasic, B.; Kovacic, I.; Skrdlikova, L. Polycyclic aromatic hydrocarbons in air on small spatial and temporal scales - II. Mass size distributions and gas-particle partitioning. Atmos. Environ. 2010, 44, 5022−5027. (7) Masih, J.; Singhvi, R.; Taneja, A.; Kumar, K.; Masih, H. Gaseous/ particulate bound polycyclic aromatic hydrocarbons (PAHs), seasonal variation in North central part of rural India. Sustain. Cities Soc. 2012, 3, 30−36. (8) Lee, J. Y.; Lane, D. A. Unique products from the reaction of naphthalene with the hydroxyl radical. Atmos. Environ. 2009, 43, 4886−4893. (9) Kautzman, K. E.; Surratt, J. D.; Chan, M. N.; Chan, A. W. H.; Hersey, S. P.; Chhabra, P. S.; Dalleska, N. F.; Wennberg, P. O.; Flagan, R. C.; Seinfeld, J. H. Chemical composition of gas- and aerosol-phase products from the photooxidation of naphthalene. J. Phys. Chem. A 2010, 114, 913−934. (10) Keyte, I. J.; Harrison, R. M.; Lammel, G. Chemical reactivity and long-range transport potential of polycyclic aromatic hydrocarbons − a review. Chem. Soc. Rev. 2013, 42, 9333−9391. (11) Chan, A. W. H.; Kautzman, K. E.; Chhabra, P. S.; Surratt, J. D.; Chan, M. N.; Crounse, J. D.; Kürten, A.; Wennberg, P. O.; Flagan, R. C.; Seinfeld, J. H. Secondary organic aerosol formation from photooxidation of naphthalene and alkylnaphthalenes: implications for oxidation of intermediate volatility organic compounds (IVOCs). Atmos. Chem. Phys. 2009, 9, 3049−3060. (12) Shakya, K. M.; Griffin, R. J. Secondary organic aerosol from photooxidation of polycyclic aromatic hydrocarbons. Environ. Sci. Technol. 2010, 44, 8134−8139. (13) Kleindienst, T. E.; Jaoui, M.; Lewandowski, M.; Offenberg, J. H.; Docherty, K. S. The formation of SOA and chemical tracer compounds from the photooxidation of naphthalene and its methyl analogs in the presence and absence of nitrogen oxides. Atmos. Chem. Phys. 2012, 12, 8711−8726. (14) Zhou, S.; Wenger, J. C. Kinetics and products of the gas-phase reactions of acenaphthene with hydroxyl radicals, nitrate radicals and ozone. Atmos. Environ. 2013, 72, 97−104. (15) Zhou, S.; Wenger, J. C. Kinetics and products of the gas-phase reactions of acenaphthylene with hydroxyl radicals, nitrate radicals and ozone. Atmos. Environ. 2013, 75, 103−112. (16) Riva, M.; Healy, R. M.; Flaud, P.-M.; Perraudin, E.; Wenger, J. C.; Villenave, E. Kinetics of the gas-phase reactions of chlorine atoms with naphthalene, acenaphthene and acenaphthylene. J. Phys. Chem. A 2014, 118, 3535−3540. (17) Faxon, C. B.; Allen, D. T. Chlorine chemistry in urban atmosphere: a review. Environ. Chem. 2013, 10, 221−233. (18) Cai, X.; Griffin, R. J. Secondary aerosol formation from the oxidation of biogenic hydrocarbons by chlorine atoms. J. Geophys. Res. 2006, 111, D14206. (19) Cai, X.; Ziemba, L. D.; Griffin, R. J. Secondary aerosol formation from the oxidation of toluene by chlorine atoms. Atmos. Environ. 2008, 42, 7348−7359. (20) Huang, M.; Zhang, W.; Gu, X.; Hu, C.; Zhao, W.; Wang, Z.; Fang, L. Size distribution and chemical composition of secondary organic aerosol formed from Cl-initiated oxidation of toluene. J. Environ. Sci. 2012, 24, 860−864. (21) Thüner, L. P.; Bardini, P.; Rea, G. J.; Wenger, J. C. Kinetics of the gas-phase reactions of OH and NO3 radicals with dimethylphenols. J. Phys. Chem. A 2004, 108, 11019−11025.

Figure 6. Average mass spectra (in the positive ion mode) for SOA formed from the reaction of Cl atoms with acenaphthylene using a laser pulse energy of 0.2 mJ.

tration of 4 × 105 molecules per cubic centimeter and a global tropospheric 12 h average daytime OH radical concentration of 2 × 106 molecules per cubic centimeter, 4%, 35%, and 45% of naphthalene, acenaphthene, and acenaphthylene may react, respectively, with Cl atoms under these conditions and could produce a significant amount of SOA. Some of the oxidation products identified in the SOA, for example, chloronaphthalene and chloroacenaphthenone, are not observed in competing gasphase oxidation processes and are thus potential markers for identifying the occurrence of the Cl atom-initiated oxidation of PAHs in the ambient atmosphere.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. Phone: 33-5-40006350. (E.V.) *E-mail: [email protected]. Phone: 353-2-1490-2454. (J.C.W.) Present Address #

Department of Environmental Sciences and Engineering, Gillings School of Global Public Health, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Research at Univ. College of Cork was supported by the EUFP7 European Simulation Chambers for Investigating Atmospheric Processes (EUROCHAMP-2, Grant No. 228335). The authors wish to thank the French Agency for Environment and Energy Management (ADEME) and the Aquitaine Region for their financial support.

■

REFERENCES

(1) Kroll, J. H.; Seinfeld, J. H. Chemistry of secondary organic aerosol: formation and evolution of low-volatility organics in the atmosphere. Atmos. Environ. 2008, 42, 3593−3624. (2) Stevens, B.; Boucher, O. The aerosol effect. Nature 2012, 490, 40−41. (3) Hallquist, M.; Wenger, J. C.; Baltensperger, U.; Rudich, Y.; Simpson, D.; Claeys, M.; Dommen, J.; Donahue, N. M.; George, C.; et al. Goldstein et al. The formation, properties and impact of K


Article

The Journal of Physical Chemistry A (22) Baklanov, A. V.; Krasnoperov, L. N. Oxalyl chloride - A clean source of chlorine atoms for kinetic studies. J. Phys. Chem. A 2001, 105, 97−103. (23) Cappellin, L.; Karl, T.; Probst, M.; Ismailova, O.; Winkler, P. M.; Soukoulis, C.; Aprea, E.; Märk, T. D.; Gasperi, F.; Biasioli, F. On quantitative determination of volatile organic compound concentration using proton transfer reaction time-of-flight mass spectrometry. Environ. Sci. Technol. 2012, 46, 2283−2290. (24) Ahmed, M.; Blunt, D.; Chen, D.; Suits, A. UV photodissociation of oxalyl chloride yields for fragments from one photon absorption. J. Chem. Phys. 1997, 106, 7617. (25) Barmet, P.; Dommen, J.; DeCarlo, P. F.; Tritscher, T.; Praplan, A. P.; Platt, S. M.; Prévôt, A. S. H.; Donahue, N. M.; Baltensperger, U. OH clock determination by proton transfer reaction mass spectrometry at an environmental chamber. Atmos. Meas. Tech. 2012, 5, 647− 656. (26) Wang, L.; Arey, J.; Atkinson, R. Reactions of chlorine atoms with a series of aromatic hydrocarbons. Environ. Sci. Technol. 2005, 39, 5302−5310. (27) Gard, E.; Mayer, J. E.; Morrical, B. D.; Dienes, T.; Fergenson, D. P.; Prather, K. A. Real-time analysis of individual atmospheric aerosol particles: Design and performance of a portable ATOFMS. Anal. Chem. 1997, 69, 4083−4091. (28) Silva, P. J.; Prather, K. A. Interpretation of mass spectra from organic compounds in aerosol time-of-flight mass spectrometry. Anal. Chem. 2000, 72, 3553−3562. (29) Zimmermann, R.; Ferge, T.; Galli, M.; Karlsson, R. Application of single-particle laser desorption/ionization time-of-flight mass spectrometry for detection of polycyclic aromatic hydrocarbons from soot particles originating from an industrial combustion process. Rapid Commun. Mass Spectrom. 2003, 17, 851−859. (30) Sasaki, J.; Aschmann, S. M.; Kwok, E. S. C.; Atkinson, R.; Arey, J. Products of the gas-phase OH and NO3 radical-initiated reactions of naphthalene. Environ. Sci. Technol. 1997, 31, 3173−3179. (31) Wang, L.; Atkinson, R.; Arey, J. Dicarbonyl products of the OH radical-initiated reactions of naphthalene and the C1- and C2alkylnaphthalenes. Environ. Sci. Technol. 2007, 41, 2803−2810. (32) Wei, H.-Z.; Jiang, S.-Y.; Xiao, Y.-K.; Wang, J.; Lu, H.; Wu, B.; Wu, H.-P.; Li, Q.; Luo, C.-G. Precise determination of the absolute isotopic abundance ratio and the atomic weight of chlorine in three international reference materials by the positive thermal ionization mass spectrometer -Cs2Cl+- graphite method. Anal. Chem. 2012, 84, 10350−10358. (33) Wibaut, J. P.; Bloem, G. P. The chlorination of naphthalene in the vapour phase. Recl. Trav. Chim. Pays-Ba. 1950, 69, 586−592. (34) Sokolov, O.; Hurley, M. D.; Wallington, T. J.; Kaiser, E. W.; Platz, J.; Nielsen, O. J.; Berho, F.; Rayez, M.-T.; Lesclaux, R. Kinetics and mechanism of the gas-phase reaction of Cl atoms with benzene. J. Phys. Chem. A 1998, 102, 10671−10681. (35) Zhang, Z.; Lin, L.; Wang, L. Atmospheric oxidation mechanism of naphthalene initiated by OH radical. A theoretical study. Phys. Chem. Chem. Phys. 2012, 14, 2645−2650. (36) Calvert, J. G.; Atkinson, R.; Becker, K. H.; Kamens, R. M.; Seinfeld, J. H.; Wallington, T. J.; Yarwood, G. The Mechanisms of Atmospheric Oxidation of Aromatic Hydrocarbons; Oxford University Press, 2002; p 556. (37) Orlando, J. J.; Tyndall, G. S.; Wallington, T. J. The atmospheric chemistry of alkoxy radicals. Chem. Rev. 2003, 103, 4657−4689. (38) Healy, R. M.; Chen, Y.; Kourtchev, I.; Kalberer, M.; O’Shea, D.; Wenger, J. C. Rapid formation of secondary organic aerosol from the photolysis of 1-nitronaphthalene: role of naphthoxy radical selfreaction. Environ. Sci. Technol. 2012, 46, 11813−11820. (39) Reisen, F.; Arey, J. Reactions of hydroxyl radicals and ozone with acenaphthene and acenaphthylene. Environ. Sci. Technol. 2002, 36, 4302−4311. (40) Sauret-Szczepanski, N.; Lane, D. A. Smog chamber study of acenaphthene: gas-particle partition measurements of the products formed by reaction with the OH radical. Polycyclic Aromat. Compd. 2004, 24, 161−172.

(41) Sigman, M. E.; Chevis, E. A.; Brown, A.; Barbas, J. T.; Dabestani, R.; Burch, E. L. Enhanced photoreactivity of acenaphthylene in water: A product and mechanism study. J. Photochem. Photobiol., A 1996, 94, 149−155. (42) Borrás, E.; Tortajada-Genaro, L. A. Secondary organic aerosol formation from the photo-oxidation of benzene. Atmos. Environ. 2012, 47, 154−163.

L


Gas- and Particle-Phase Products from the Chlorine-Initiated

Recommend Documents