" Adapted from Fraser et al. (1998). h Reactivity classification (Nielsen, 1984; see Table 10.30). ' September 8-9 (48 h), 1993. Sampling at Claremont occurred during a major photochemical smog episode (peak Oj = 293 ppb).

" Adapted from Fraser et al. (1998). h Reactivity classification (Nielsen, 1984; see Table 10.30). ' September 8-9 (48 h), 1993. Sampling at Claremont occurred during a major photochemical smog episode (peak Oj = 293 ppb).

observed to accumulate during downwind transport due to their formation in atmospheric reactions.

These spatial and temporal distributions of biologically active PAHs and PACs during transport across an air basin, e.g., the rapid decay of reactive mutagens/carcinogens, the long lifetimes of low-reactivity PAH carcinogens, and the significant rate of formation of human cell nitro-PAH mutagens, have significant exposure and public health implications, e.g., in risk assessments of the health effects of benzo[a]pyrene and diesel emissions (e.g., see California Air Resources Board reports, 1994 and 1998, respectively; WHO, 1996; and Collins et al., 1998).

From the perspective of atmospheric chemistry and toxicology, recall that the human cell mutagenic potencies of fine ambient aerosols were essentially the same in samples collected concurrently at four sites across southern California (see Fig. fO.23 and Sections D.2 and D.3). However, by the time a heavily polluted air parcel from central Los Angeles reaches the downwind site of Claremont, the ambient concentrations of two of the most powerful mutagens, BaP and cyclopenta-[cd]pyrene, have dropped dramatically (Fig. 10.27). Key questions include: What are the chemical structures, ambient levels, and mechanisms of formation of the genotoxic PACs in fine ambient aerosols that make up this human cell "mutagenicity gap?" Are they biologically active O- and N-PAC reaction products of particle-phase PAHs, so that as the primary promutagens decay, biologically active reaction products are formed? Alternatively, are mutagens formed by reactions in air during transport of nonmutagenic PAH? Of course, both mechanisms may be occurring simultaneously.

4. Photochemical Reactions of Particle-Associated PAHs

The reviews "Toxicological Indications of the Organic Fraction of Aerosols: A Chemist's View" by Van Cauwenberghe and Van Vaeck (1983) and "Atmospheric Reactions of PAH" by Van Cauwenberghe (1985) provide critical assessments and extensive literature references of the status of research to 1985 in the complex area of heterogeneous photochemical reactions and of the interactions of PAHs on laboratory substrates, primary combustion particles, and ambient particulate matter with ozone and NOz in air. In the following sections, we briefly summarize results from this earlier era and address subsequent studies on heterogeneous atmospheric reactions of PAHs in simulated and real atmospheres.

ft is now generally agreed that photodegradation is the major chemical pathway for the loss of 4- to 6-ring PAHs in ambient aerosols. Reaction rates, mechanisms, and products of this overall phenomenon depend not only on the structures and UV-visible absorption spectra (on and in aerosol surfaces) of specific PAHs but also on the physical and chemical natures of the surfaces of the particles on/in which they are adsorbed/absorbed (i.e., sorbed). For example, the critical role played by substrate surfaces is reflected in the photolytic half-lives of ~ fOOO h for benz[a]anthracene (BaA) adsorbed on "black" coal fly ash and irradiated with a laboratory photoreactor (Behymer and Hites, 1988) compared to ~25 min for BaA associated with wood soot exposed midday outdoors in a Teflon chamber under moderate conditions of temperature and relative humidity (Kamens et al., 1988). For this atmospheric aerosol (as well as diesel soot and secondary organic aerosol, SOA), as mentioned earlier, the surface can be viewed as an organic liquid-like layer containing water and a wide range of different organic chemicals (e.g., see McDow et al., 1996, and references therein).

In contrast to their photostability on a black coal fly ash substrate, reactive PAHs such as BaP and BaA photodecompose quite rapidly with half-lives in the range of ~l-4 h (depending on experimental conditions) at the solid-air interfaces of certain water-insoluble inorganic oxides such as Si02 and A1203 (e.g., Behymer and Hites, 1985; see also the reviews on PAH-Si02 systems by Sigman et al. (1997) and Dabestani (1997) and articles on specific PAHs discussed herein).

These photochemical reactions in liquids and on solid surfaces are discussed in the following sections.

a. Photooxidation in Solution and "Liquid-like" Surfaces of Organic Aerosols

In 1963, E. J. Bowen published his classic review "The Photochemistry of Aromatic Hydrocarbon Solutions," in which he described two major reaction pathways for PAHs irradiated in organic solvents: pho-todimerization and photooxidation mediated by the addition of singlet molecular oxygen, Ojf'A) (or simply '02), to a PAH (e.g., anthracene). For details on the spectroscopy and photochemistry of this lowest electronically excited singlet state of molecular oxygen, see Chapter 4.A, the monograph by Wayne (1988), and his review article (1994). For compilations of quantum yields of formation and of rate constants for the decay and reactions of 02('A), see Wilkinson et al., f 993 and 1995, respectively.

Khan and co-workers suggested in 1967 that in polluted urban atmospheres '02 might be generated by triplet energy transfer from "strongly absorbing polynuclear aromatic hydrocarbons to normal oxygen," i.e., that PAHs could act as photosensitizers. Furthermore, the absorbing molecule, e.g., a PAH, "could be in the solid, liquid, or gaseous state, or adsorbed on a solid" (Pitts et al., 1969). This suggestion was confirmed directly in experiments by Eisenberg and co-workers (1984) employing several model PAHs deposited on surfaces as sensitizers, interestingly, they suggested that the 'O, so formed could then react with a second molecule of the same PAH (in its ground electronic state) to form oxygenated products, some of which are direct-acting mutagens in the Ames assay. The overall process, shown in the following reaction scheme, is referred to as autooxidation.



The 02('Ag) can then react with another BaP molecule (or a different PAH receptor molecule in a POM mixture), e.g.,

Note: The symbols S(), S,, and T, represent the ground electronic state and first excited singlet and triplet states, respectively, of BaP. 02('A ) is 22.5 kcal mol"1 above the ground state of 02, while for BaP the S, and T, states are 7f and 42 kcal mol-1 above the S„ ground state, respectively.

Fox and Olive (1979) described the photooxidation by Oz('A) of anthracene associated with ambient particulate POM. McCoy and Rosenkranz (1980) reacted '02 with chrysene and 3-methylcholanthrene and reported products that are direct-acting mutagens. Furthermore, Seed and co-workers (1989) have shown that exposure to singlet molecular oxygen oxidizes several BaP metabolites, giving chemical products that are "mutagenic in the absence of mammalian metabolic action," i.e., that are direct bacterial mutagens in the Salmonella typhimurium forward mutation assay. Such phenomena are relevant to current interest in the possible formation during transport of directly mutagenic species in the heterogeneous degradation of certain PAHs associated with ambient aerosols.

The key role of 02('A) in the direct photooxidation of benzo[a]pyrene dissolved in aerated benzene solutions (10 ~3 M) and irradiated using a 500-W visible light source with a Pyrex filter is described by Lee-Ruff and co-workers (1986). They also carried out a sensitized photooxidation of BaP using such singlet oxygen sensitizers as tetraphenylporphyrin and methylene blue and found the same distribution of products as with the direct photooxidation, thus establishing the key role of 02('A) in both systems. Products included a mixture of 1,6-, 3,6-, and 6,12-benzo[a]pyrenedione, which had earlier been reported in ambient aerosols in Toronto, Canada (Pierce and Katz, 1976), Germany (König et al., 1983), and Scandinavia (Ramdahl, 1983b). Additionally, they isolated and identified a major product not previously reported, the 5a,6-secobenzo[a]pyrene derivative, the aldehyde XXXIII:


This product suggests a singlet oxygen mediated mechanism for quinone formation (Lee-Ruff et al., 1986).

Subsequently, Lee-Ruff and Wang (1991) conducted a similar study of 6-methylbenzo[a]pyrene, a methyl isomer whose carcinogenicity is approximately equal to that of BaP. It photooxidized ~20 times faster than BaP and, in addition to quinones, formed as a major product a seco ketone analogous to XXXIII from BaP. Its formation (as with BaP) is ascribed to a '02-media-ted mechanism.

Since the mid-1980s, Kamens, McDow, and coworkers have carried out a number of studies on factors influencing the rates, mechanisms, and products of the photodegradation of PAHs. They employed both model laboratory systems in which solutions containing various organics found in ambient aerosols are irradiated in photoreactors and environmental chamber studies in which real organic combustion aerosols relevant to ambient atmospheres, e.g., diesel soot and wood smoke, are irradiated with natural or simulated sunlight under a variety of ambient conditions.

For example, Kamens and co-workers (1988) reported that daytime photodegradation rates of PAHs on wood smoke and gasoline combustion soot irradiated by natural sunlight in 25-m3 outdoor Teflon film chambers were clearly influenced by such variables as solar intensity, relative humidity, and ambient temperature. Thus, under "moderate conditions" (e.g., T = 20°C, "moderate" humidity, and a solar intensity of 1 cal cm-2 min-1), PAH half-lives for the most reactive PAHs (Class II, Table 10.30), cyclopenta[c<f]pyrene and BaP, were 0.3 and 0.5 h, respectively, compared to 0.8 and 1.3 h for the low-reactivity (Class V) benzofluo-ranthenes (k and b). At very low values of sunlight intensity, water vapor concentrations, and ambient temperatures (—10°C), half-lives increased significantly; e.g., for cyclopenta[c<i]pyrene and BaP, the half-lives increased to 2 and 6 h, respectively, and for the benzofluoranthenes to ~ 10 h. These PAH relative rates are generally consistent with the Nielsen reactivity scale (Nielsen, 1984; Table 10.30). Note that McDow and co-workers (1995) reported that the water content of combustion particles of wood smoke and gasoline soot increased with increasing relative humidity and that PAH photodegradation rates are likely to increase with increasing water content (BaP was an exception) (see also a recent modeling study of humidity effects by Jang and Kamens (1998)).

McDow, Kamens, and co-workers also conducted laboratory experiments on the effects of common organic constituents (e.g., methoxyphenols) on the rates, mechanisms, and products of the solution-phase photodegradation of PAHs associated with wood smoke and diesel soot (see, for example, Odum et al. (1994a), and McDow et al. (1994, 1995, 1996)). Figure 10.28, for example, shows the degradation of the reactive BaP (Class II reactivity) compared to BeP (Class V reactivity) in two solvents, hexadecane, taken as representative of aliphatics in diesel soot, or a mixture of 11 methoxyphenols found in particulate matter (McDow et al., 1994). As expected, BaP decays much more rapidly than BeP. In addition, the decay in the mixture of methoxyphenols is much faster than that in hexadecane.

a Hexadecane ■ Methoxyphenols

FIGURE 10.28 Rates of photodegradation of benzo[a]pyrene (Class II reactivity; see Table 10.30) and benzo[e]pyrene (Class V) in irradiated solutions of hexadecane or a mixture of methoxyphenols that are representative of important classes of organics present in particles of diesel soot and wood smoke, respectively (adapted from McDow et al., 1994).

a Hexadecane ■ Methoxyphenols

FIGURE 10.28 Rates of photodegradation of benzo[a]pyrene (Class II reactivity; see Table 10.30) and benzo[e]pyrene (Class V) in irradiated solutions of hexadecane or a mixture of methoxyphenols that are representative of important classes of organics present in particles of diesel soot and wood smoke, respectively (adapted from McDow et al., 1994).

0 0

Post a comment