Exceeding of Henry's Law by Hydrogen Peroxide Associated with

by gas-particle partitioning (Henry's law), which averages. 0.1 mM. This indicates .... described by Hasson and Paulson (20) into a helical coil detec...
0 downloads 0 Views 199KB Size
Download PDF
Research Exceeding of Henry’s Law by Hydrogen Peroxide Associated with Urban Aerosols CHUAUTEMOC ARELLANES,† S U Z A N N E E . P A U L S O N , * ,† PHILIP M. FINE,‡ AND CONSTANTINOS SIOUTAS‡ Department of Atmospheric and Oceanic Sciences, University of California, Los Angeles, Los Angeles, California 90025, and Department of Civil and Environmental Engineering, University of Southern California, Los Angeles, California 90089

Simultaneous measurements of gas- and aerosol-phase hydrogen peroxide (H2O2) have been made at two sites in Los Angeles, one near the Pacific coast at the University of California at Los Angeles (UCLA), and the other in downtown Los Angeles with close proximity to a heavily traveled freeway (freeway site). At both the freeway and UCLA sites, gas-phase H2O2 levels were similar, averaging 1.17 ( 1.0 and 1.05 ( 0.6 ppb, respectively. The particle-associated H2O2 in both fine (PM2.5) and coarse (>PM2.5) modes was higher at the freeway site, as compared to UCLA, by a factor of 2. However, when aerosol-phase H2O2 is normalized to particle mass loadings, the fine-mode H2O2 levels are very similar at the two sites: 0.42 ( 0.3 and 0.58 ( 0.3 ng H2O2/µg particle mass at the freeway and UCLA sites, respectively. The normalized coarse-mode H2O2 levels were significantly higher at the freeway site than at UCLA, 1.05 ( 0.3 and 0.51 ( 0.3 ng/µg, respectively. Estimating aerosol liquid water content on the basis of relative humidity and aerosol mass, a calculated equivalent H2O2 in aerosol liquid water averages 70 mM, more than 2 orders of magnitude higher than concentrations predicted by gas-particle partitioning (Henry’s law), which averages 0.1 mM. This indicates that the sampled particles are capable of generating H2O2 in aqueous solution. These corresponding aqueous-phase H2O2 concentrations in aerosol liquid water exceed levels that have been observed to produce cellular damage to lung epithelial cells in laboratory experiments by at least 3 orders of magnitude. Although most measurements of H2O2 in particles were made using an extraction solution adjusted to pH 3.5, a set of measurements indicates that H2O2 from fine-mode particles extracted in the physiologically relevant pH range 5-7.5 also generate H2O2 with only slightly lowered efficiency; coarse-mode H2O2 production dropped by 75% at the upper end of this range. Finally, a small set of measurements was performed to investigate the degree to which the recently developed Versatile Aerosol Concentrator Enrichment System (VACES) affects H2O2 levels in concentrated * Corresponding author phone: (310)206-4442; fax: (310)206-5219; e-mail: [email protected]. † University of California, Los Angeles, Los Angeles. ‡ University of Southern California. 10.1021/es0513786 CCC: $33.50 Published on Web 07/14/2006

 2006 American Chemical Society

ambient aerosols. The VACES appeared to a have minimal impact on particulate H2O2.

Introduction Aerosols have long been known to obscure visibility, damage materials, and adversely affect human health. Recent epidemiological studies indicate that increases in human morbidity and mortality due to lung cancer, cardiopulmonary disease, and asthma are associated with significantly lower concentrations of fine particles (PM2.5) than previously thought to affect human health (1-3). Aerosols in the ultrafine mode (2.5 µm) are not free of toxic effects, however; several recent studies indicate their toxicity may exceed fine-mode particles (7, 8). One suite of compounds suspected to be causative agents in particle toxicity are reactive oxygen species (ROS). These oxygen-centered species, such as H2O2, the hydroxyl radical, and superoxide anion, have been implicated in respiratory distress diseases such as asthma and in carcinogenesis (9-13). Thus, aerosol-borne compounds that elicit formation of ROS as well as ROS themselves have been the subject of intense investigation (9-16). Numerous in vitro studies of H2O2 at levels below those expected for particles in equilibrium with ambient air have been shown to damage lung epithelial cells. The results of these studies form the basic framework supporting the hypothesis that elevated levels of ROS induces oxidative stress within macrophages (9-14, 16, 17). Exposure of respiratory tract cells, extracted from laboratory rats, to aqueous H2O2 solutions with concentrations ranging from 20 pM to 1 µM result in significant cell damage (9-13, 18). Damage is characterized by inhibition of the ADP to the ATP conversion cycle; destruction of the alveolar epithelium, including DNA strand breaking; a decrease in lung surfactant biosynthesis; and inhibition of alveolar epithelial wound repair (10-12, 18). To date, one in vivo study has been performed, in which Sprague-Dawley rats were exposed to ammonium sulfate aerosols (215 or 430 µg/m3) and H2O2 (10, 20, 100 ppb) (19). Exposure to either (NH4)2SO4 aerosols or gas-phase H2O2 generated very limited changes; however, when the aerosols were combined with gas-phase H2O2, notable cellular damage occurred. This included an increase in the number of neutrophils in the pulmonary capillaries found adhered to the vascular endothelium and an increase in tumor necrosis factor-R production by alveolar macrophages. A reversible transient increase in the production of superoxide anion by alveolar macrophages was also noted, indicative of induced oxidative stress (19). In addition to the physiological relevance of ROS, generation of oxidants in particles is likely to affect the chemistry of the particles themselves; for example, by facilitating some of the polymerization reactions that lead to the formation of secondary organic aerosol and by affecting redox reactions in the particles. Hydrogen peroxide is formed in the atmosphere predominantly via the self-reaction of the HO2 radical and to a lesser degree by water reacting with Criegee intermediates (20). Numerous campaigns over the last three decades have shown that gas-phase H2O2 concentrations typically lie between 0.5 and 5 ppb worldwide (21). Hydrogen peroxide VOL. 40, NO. 16, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4859

partitions strongly into the aqueous phase with a Henry’s law coefficient of 1.0 × 105 M atm-1 (22). Henry’s law is given as

X a ) Pa × H a where Ha ) Henry’s law coefficient for species “a” (M atm-1), Xa ) the aqueous concentration (M) of species “a”, and Pa ) partial pressure of species “a” (atm). This relation predicts that the aqueous concentration of H2O2 in liquid water that is in equilibrium with 1 ppb of H2O2 will be 0.1 mM. For an aerosol mass of 50 µg m-3, of which 50% is water, this translates into 0.085 ng m-3 of H2O2 in the aerosol phase. To date, four studies have reported ROS measurements in aerosols. Hewitt and Kok (23), using the same detection method employed here, briefly mention measurements of aerosol-phase H2O2 and report levels of 2.5 µm. At the end of each channel is a virtual impactor operating at a 2.5-µm size-cut which removes 95% of the gas, thus concentrating aerosols. The fine-mode aerosols investigated in this study were then passed through a diffusion drier (model 3062, TSI Inc.), which returns aerosols to their original size. Concentrated finemode aerosols were collected on Teflon filters (37 mm, 2-µm pore, Pall Corporation). These filters were extracted and analyzed for H2O2 using the same method as that used for virtual impactor-collected ambient aerosols.

Results H2O2 Mass Loadings: a Site Comparison. Figure 1 shows gas and aerosol associated H2O2, the latter segregated into fine and coarse modes for the freeway and UCLA sites. Samples were collected at the freeway site prior to May 6, 2004, and at UCLA after this date. Data collected prior to March 2004 used a coarse/fine size cut of 1.8 µm; thereafter, the size cut was set at 2.5 µm. All samples were collected during daytime hours. As a whole, the aerosol-associated H2O2 levels were high and well above levels predicted by Henry’s law, as discussed below. Ambient gas-phase H2O2 at both sites ranged from 0.2 to 3.0 ppbv, and averages were similar at the two sites: 1.2 ( 1 and 1.0 ( 0.6 ppb for the freeway and UCLA sites, respectively. In contrast, other pollutant levels were generally higher at the freeway site. Two factors that could act to reduce H2O2 relative to other pollutants at the freeway site are (a) that gas-phase H2O2, which is photochemically rather than source-derived, is typically lower in winter months, when samples were collected at the freeway site; and (b) that the freeway site experiences higher levels of NOx, which suppresses H2O2 formation. The peroxide content in both fine- and coarse-mode particles at the freeway site were roughly double those observed at UCLA. Fine-mode H2O2 was 12 ( 9 ng m-3 and 5.4 ( 6 ng m-3 at the freeway and UCLA sites, respectively. Coarse-mode H2O2 was 20 ( 9 ng m-3 and 10 ( 7 ng m-3 at the freeway and UCLA sites, respectively. Differences in particle masses for the two sites appear to account for most of the observed differences for the fine-mode particles: finemode particle mass loadings were 1.7 times larger at the freeway site as compared to UCLA: 23 ( 8 µg m-3 and 13 ( 10 µg m-3, respectively. In contrast, the average coarsemode particle masses are similar at the two sites: 27 ( 33 VOL. 40, NO. 16, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4861

FIGURE 2. H2O2 in fine- (2.5 mm) aerosols, normalized to particle mass: (a) freeway samples and (b) UCLA samples. Data that are not labeled with a date correspond to additional samples collected on the same day. Normalized H2O2 levels for the fine-mode particles are similar at the two sites, whereas coarse-mode levels are higher at the freeway site. µg m-3 and 26 ( 15 µg m-3 at the freeway and UCLA, respectively. Our measurements are in reasonable agreement with two of the other measurements of H2O2 in urban air; for a sampling of particles at a sidewalk in Taipei, Hung and Wang (24) observed an average total H2O2 of 21 ng m-3 for PM10, which is between our (total suspended particulate) values of 15 ng m-3 and 32 ng m-3 for the UCLA and freeway sites, respectively. The earlier study by our group (25) found somewhat lower total particulate-associated H2O2, averaging 3.2 ng m-3; this can be compared to the average, 11 ng m-3, prior to the construction activity at the UCLA site. It is not clear if this difference is due to the relatively small number of samples in the earlier study, intrinsic differences in the samples, or something else. All of these results are much lower than those of Venkatachari et. al. (26), who reported an average of 243 ng m-3 in Rubidoux, a heavily polluted area of Los Angeles. It is possible that particle mass loadings at the Rubidoux site were extraordinarily high (masses were not reported) or that sonication or another aspect of the analytical technique employed, or another factor, is responsible for the large difference. Aerosol H2O2 is most likely a function of the aerosol intrinsic properties, scaled by the quantity of aerosol sampled. Particle mass concentration appears to be a weak indicator of aerosol-phase H2O2. For all data combined (from both the freeway site and UCLA), the coarse-mode particles exhibit a positive correlation (r ) 0.53, P ) 0.1%), whereas the finemode particles overall do not show any correlation. Subsets of the data exhibit much larger correlation coefficients, although they are based on smaller numbers of observations and, thus, do not reveal clearly if particle mass is a better indicator for sets of presumably similar particles. For example, fine-mode particle mass at the freeway site is well-correlated with particulate H2O2 (r ) 0.64, P ) 3%), as are coarse-mode particles at the UCLA site (r ) 0.82, P ) 0.3%). Figure 2 shows aerosol-phase H2O2, normalized to particle mass for both sites in units of nanograms of H2O2 per microgram of particle mass. The average fine-mode H2O2 loadings were similar at the two sites: 0.42 ( 0.3 and 0.58 ( 0.3 ng µg-1 at the freeway and UCLA sites, respectively. The coarse-mode H2O2 loadings were higher by a factor of ∼2 at the freeway site, as compared to UCLA, 1.05 ( 0.3 ng µg-1 and 0.51 ( 0.3 ng µg-1, 4862

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 16, 2006

respectively. Our normalized values are similar to, but somewhat higher than, those of Hung and Wang (24), which averaged in the range 0.15-0.35 ng µg-1, depending on the aerosol size fraction. VACES Intercomparison Results. The VACES, which first grows aerosols by condensing water vapor on them, concentrates them by virtual impaction, and then dries them back to roughly their original size distribution (27, 28), has the potential to alter the aerosol content of highly soluble species, such as H2O2. Results of the intercomparison of finemode aerosol associated H2O2 in particles sampled by the VACES and the pair of UCLA virtual impactors are shown in Figure 3. The H2O2 content of particles sampled using these two methods is in reasonable agreement, averaging 12 ( 4 and 14 ( 6 ng m-3 for UCLA virtual impactors averaged and VACES, respectively. In four of the five sampling events, the fine-mode H2O2 content was slightly higher in the VACES. There are several possible sources of differences, including flow rate errors and enhancement of aerosol H2O2 via absorption from the gas phase when particles are grown via condensation of water. The average ratio of the VACES to the virtual impactor results is 1.2 ( 0.2, well within the variability and combined sampling and measurement errors of the VACES and virtual impactors. Although the data set is too small to differentiate between a slight enhancement or none at all, it appears that the VACES processing induces a slight enhancement. The results are broadly consistent with the notion that H2O2 partitioning from the gas phase does not substantially control particle peroxide content. Does Aerosol Phase H2O2 Obey Henry’s Law?. Henry’s law, which describes gas-particle phase partitioning, might be expected to predict the aerosol H2O2 levels from the gasphase concentration of H2O2. It can be shown that the time to establish Henry’s law equilibrium is on the order of 10-3 seconds; thus, the aerosol liquid phase should, in principle, be in equilibrium with the gas-phase H2O2 concentration at the time filters are transferred from filter holders to the extraction solution. The data collected in this study can be used in several ways to assess the ability of Henry’s law to predict H2O2 in aerosols. Figure 4 shows hydrogen peroxide concentrations in aerosol liquid water calculated variously on the basis of Henrys law, the measured gas and particleassociated H2O2 levels, relative humidity, and the aerosol

FIGURE 3. Intercomparison of the hydrogen peroxide content in PM2.5 processed with the VACES and sampled with two UCLA virtual impactors. Samples were collected during the week of May 4-7, 2004, at the freeway site. mass for each sample for which all of the necessary parameters was measured. To derive the Henry’s law prediction for aqueous H2O2 levels, the measured gas-phase H2O2 is multiplied by HA. To calculate the concentration of H2O2 in aerosol liquid water from measured aerosol peroxide levels, we considered two cases: (a) that water makes up 50% of particle mass regardless of the relative humidity (RH), or (b) that the aerosol liquid water content (LWC) is determined by the RH in a manner that accounts for the nonlinear uptake of water by mixed salts, estimated according to the method described by Slone and Wolff (32). In these calculations, we assumed that a particle density of 1 g cm-3, which is likely only a slight underestimation, given that we measured wet particle mass. Since the RH was usually below 55%, the point at which LWC accounts for ∼50% of particle mass according to Slone and Wolff’s model, the calculation that accounts for RH results in higher calculated concentrations for H2O2 in aerosol liquid water. In all cases, measured particle-borne H2O2 greatly exceeds the Henry’s law value. Using the assumption that the LWC is 50% of the measured aerosol mass, the measured concentrations exceed Henry’s law by a factor of 450; assuming the LWC calculated on the basis of measured aerosol mass and RH, the measured concentrations exceed Henry’s law, on average, by a factor of 700 (the last data point in Figure 4, which exceeds other data by a factor of 10 or more was omitted from these averages). The concentrations of H2O2 calculated for the aerosol liquid water are very high, up to 100 mM or more. It should be emphasized that because calculated H2O2 partitions rapidly between the gas and liquid phases, we do not believe that aerosol liquid water contains this calculated H2O2; rather, that H2O2 is generated continuously by particles in the extraction solution. Most likely, particles also generate H2O2 while they are suspended in the atmosphere, as well, although in this case, it should result in a continuous flux of H2O2 from the particles. Further supporting the notion that partitioning of H2O2 into aerosol liquid water does not control particulate peroxide levels, no correlation is observed between the measured particulatephase H2O2 and the RH-dependent aerosol LWC in the coarse mode. A weak negative correlation is observed for the fine mode (not shown). Hasson and Paulson (25) concluded that calculated aerosol liquid water H2O2 concentrations were “several times larger” than their HA equilibrium values, finding an enhancement factor of around 7, much lower than the present

results. This difference arises from a combination of factors. The largest is due to Hasson and Paulson’s (25) very conservative assumption about particle mass (100 µg m-3), which they did not measure, and the related LWC, 50 µg m-3. In this study, we measured an average particle mass of 39 µg m-3 at the UCLA site prior to construction activity. The LWC is a strong function of the relative humidity, which averaged 33% for the measurements shown in Figure 5, corresponding to a liquid water content of 6 µg m-3, a factor of 8 lower than that assumed by Hasson and Paulson (25). The remaining difference likely results mostly from inherent sample variability; Hasson and Paulson’s relatively small data set of 16 points (25) had average gas-phase H2O2 that were higher by a factor of 2, and lower aerosol-associated H2O2 (above). Effect of Extraction Solution pH on Particulate-Associated H2O2. Figure 5 shows the effect of extraction solution pH on H2O2 extracted from particles. Twenty-two sets of sample filters, half coarse and half fine mode, were collected on eight different days. Filters were cut in half, and one filter half was analyzed at pH 3.5, and the other(s) of the set, at pH 5.0, 6.0, or 7.6. This data set indicates that pH has a modest effect on H2O2 generated by fine-mode particles at physiological pHs, which range from ∼5 (alveolar macrophages) to 7.4 (lung fluid); somewhat less H2O2 is generated as the pH is increased. The effect on coarse-mode particles is more significant, but substantial H2O2 is still observed at pH 7.6, ∼25% of that recorded at pH 3.5. Sample-to-sample variability was reasonably high, presumably due mostly to intrinsic variability in the particles. Possible Explanations for Observed Particle Phase H2O2. Aerosol hydroperoxide levels observed after extracting them into weakly acidic aqueous solutions are very high; levels are enhanced by a factor of ∼700 over that provided by Henry’s law partitioning from the gas phase. There are several possible sources of elevated H2O2 in aerosols, which fall into the following categories: (1) Decomposition of hydroxyhydroperoxides and S(IV) complexes to generate H2O2 in solution, (2) enhanced solubility of H2O2 due to the presence of cosolutes in the aerosol liquid water, (3) photochemically induced reactions directly yielding H2O2, and (4) redox chemistry of complexed transition metals and other redoxactive species. (1) Aqueous hydroxyhydroperoxides are known to decompose to H2O2 and a corresponding aldehyde (25). Our analytical technique is capable of detecting C5 and smaller VOL. 40, NO. 16, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4863

FIGURE 4. Particulate H2O2 in aerosol liquid water (top panel, fine mode; lower panel, coarse mode). Units are micromolar to allow comparison on a log scale. The first bar (red) indicates the concentration predicted by Henry’s Law, calculated from the gas-phase concentration. The second (yellow) and third (green) bars indicate the particulate H2O2 concentration derived from the measured particulate H2O2 loading. The middle bars result from assuming that the LWC equals 50% of the measured particle mass, regardless of the RH. The right-hand bars use the assumption that aerosol liquid water content (LWC) scales with measured particle mass and relative humidity as described in the text. Data that are not labeled with a date correspond to additional samples collected on the same day. hydroxyhydroperoxides, yet apart from the occasional detection of methylhydroperoxide, we do not observe any peroxides other than H2O2 in particles. In cloud- and fogwaters, H2O2 concentrations tend to be lowered from their Henry’s law equilibria due to the oxidation of S(IV) species by H2O2, the first step of which is reversible and could lead to the rerelease of H2O2 (33). This effect, however, is likely to be much less significant in aerosols due their lower pH, which reduces the solubility of SO2 (34). These two mechanisms could possibly account for a small fraction of the observed H2O2 enhancement, but likely not more than a factor or two. (2) Dissolved ammonium sulfate, and possibly other salts, are known to enhance somewhat the solubility of H2O2 (35). On the basis of a recent study by Chung et al. (36), it is likely that inorganic salts in particles enhance the solubility of H2O2, but not by more than a factor of 2 or 3. (3) Photochemical reactions are known to generate H2O2 in cloud- and fogwater (37, 38). Aerosols collected for this study, however, are shielded from sunlight on filters during aerosol sampling. This does not, however, rule out the possibility of H2O2 production via radical-based chemistry 4864

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 16, 2006

that is ultimately initiated by photochemistry, with a decay half-life that exceeds sampling and analysis times. (4) Particles contain a variety of redox-active species, including transition metals, such as iron or copper, and organic compounds, such as quinones (17), which may be capable of generating H2O2 under oxic conditions. Cho et al. (39) reported results of recent monitoring of quinones at sites around the Los Angeles basin, finding the sum of four of the most common quinones varied from 20 to 1060 pg/ m3. These observed concentrations were, in most cases, substantially below our observations of particle-associated H2O2, so unless there are electron donors available for redox cycling, quinones might only contribute a fraction of the observed H2O2 formation activity. Iron, in its dissolved and solid forms, tends to be the most abundant transition metal found in aerosols (40). We did not measure metals during this study; however, particulate iron measured in the Los Angeles air basin by Siefert et al. (43) suggests that there may be sufficient particulate iron (several hundred nanograms per cubic meter), and possibly other metals, as well, to account for the observed H2O2 generation, even if only a portion of the iron is available to mediate

(9)

(10)

(11)

(12)

(13)

FIGURE 5. The effect of extraction solution pH on observed H2O2. Twenty-two sets of samples collected on 8 days during April and July were analyzed, 11 each for fine and coarse modes. Averages represent 5, 2, and 4 samples extracted at pHs 5.0, 6.0, and 7.6, respectively. The error bars indicate the full range of measured values for each pH.

(14)

(15)

redox reactions. Dissolved Fe(II) and Fe(III) are typically complexed by organic material (41, 42); solid-phase iron typically takes the form of iron oxyhydroxides (43). Kieber (41) has demonstrated that levels of Fe(II) in rainwaters may be higher and have slower decay times than previously thought due to the ability of Fe(II) to form organic complexes, thus preserving iron in its reduced form for many hours. In the process of cycling back to Fe(III), H2O2 can be catalytically produced (42).

Acknowledgments The authors are grateful to Dr. Antonio A. Miguel (UCLA), for use of the weighing room, Dr. DeLing Liu (JPL) for the Henry’s law equilibration time scale calculation, and Prof. C. Anastasio (UC-Davis) for helpful discussions. The authors also thank Harish Phuleria for his valuable assistance in operating the VACES. The USC group was supported in part by the California Air Resources Board (Award 2155-G-AB115) and by the Asthma Consortium, funded by the South Coast Air Quality Management District-AQMD Contract No. 04062. Finally, the authors express appreciation for the insightful comments of an anonymous reviewer.

Literature Cited (1) Tolbert, M. A.; Klein, M.; Metzger, K. B.; Peel, J.; Flanders, W. D.; Todd, K.; Mulholland, J. A.; Ryan, P. B.; Frumkin, H. Interim results of the study of particulatyes and health in Atlanta (SOPHIA). J. Exposure Anal. Environ. Epidemiol. 2000, 10, 446460. (2) Pope, C. A.; Burnett, R. T.; Thun, M. J.; Calle, E. E.; Krewski, D.; Ito, K.; Thurston, G. D. Lung cancer, cardiopulmonary mortality and long term exposure to fine particulate air pollution. J. Am. Med. Assoc. 2002, 287, 1132-1141. (3) Samet, J. Deja-vu all over again. Epidemiology 2003, 13, 118119. (4) Salvi, S.; Holgate, S. T. Mechanisms of particulate matter toxicity. Clin. Exp. Allergy 1999, 29, 1187-1194. (5) Utell, M. J.; Frampton, M. W. Acute health effects of ambient air pollution: the ultrafine particle hypothesis. J. Aerosol Med. 2000, 13, 355-359. (6) Li, N.; Hoa, M.; Phalen, R. F.; Hinds, W. C.; Nel, A. E. Particulate air pollutants and asthma: a paradigm for the role of oxidative stress in PH-induced adverse heath effects. Clin. Immunol. 2003, 109, 250-265. (7) Kleinman, M. T.; Sioutas, C.; Chang, M. C.; Boere, A. J. F.; Cassee, F. R. Ambient fine and coarse particle suppression of alveolar macrophage functions. Toxicol. Lett. 2003, 137, 151-158. (8) Schins, R. P. F.; Lightbody, J. H.; Borm, P. J. A.; Shi, T. M.; Donaldson, K.; Stone, V. Inflammatory effects of coarse and

(16)

(17)

(18)

(19)

(20)

(21) (22)

(23) (24) (25) (26) (27)

(28)

(29)

fine particulate matter in relation to chemical and biological constituents. Toxicol. Appl. Pharmacol. 2004, 195, 1-11. Gurtner, G. H.; Farrukh, L. S.; Adkinson, N. F.; Sciuto, A. M.; Jacobson, J. M.; Micheal, J. R. The role of arachidonate mediators in peroxide-induced lung injury. Am. Rev. Respir. Dis. 1987, 136, 2207-2213. Oosting, R. S.; Van Bree, L.; Van Iwaarden, J. F.; Van Golde, L. M. G.; Verheof, J. Impairment of phagocytic functions of alveolar macrophages by hydrogen peroxide. Am. J. Physiol. 1990, 259, L87-L94. Holm, B. A.; Hudak, B. B.; Keicher, L.; Cavanaugh, C.; Baker, R. R.; Hu, P.; Matalon, S. Mechanisms of H2O2-mediated injury to type II cell surfactant matabolism and protection with PEGcatalase. Am. J. Physiol. 1991, 261, C751-C757. Sporn, P. H. S.; Marshall, T. M.; Peters-Golden, M. Hydrogen peroxide increases the availability of arachidonic acid for oxidative metabolism by inhibiting acylation into phospholipids in the alveolar macrophage. Am. J. Respir. Cell Mol. Biol. 1992, 7, 307-316. Crim, C.; Longmore, W. J. Sublethal hydrogen peroxide inhibits alveolar type II cell surfactant phospholipid biosynthetic enzymes. Am. J. Physiol. 1995, 12, L129-L135. Li, N.; Wang, M.; Oberdly, T. D.; Sempf, J. M.; Nel, A. E. Comparison of the pro-oxidative and pro-inflammatory effects of organic diesel exhaust particle chemicals in bronchial epithelial cells and macrophages. J. Immunol. 2002, 169, 45314541. Sagai, M.; Lim, H. B.; Ichinose, T. Lung carcinogenesis by diesel exhaust particles and the carcinogenic mechanism via active oxygens. Inhalation Toxicol. 2000, 12, 215-223. Brown, D. M.; Donaldson, K.; Born, P. J.; Schins, R. P.; Dehnhardt, M.; Gilmor, P.; Jimenez, L. A.; Stone, V. Calcium and ROSmediated activation of transcription factors and TNF-R cytokine gene expression in macrophages exposed to ultrafine particles. Am. J. Physiol. Lung Cell. Mol. Physiol. 2004, 286. Dellinger, B.; Pryor, W. A.; Cueto, R.; Squadrito, G. L.; Hedge, V.; Deutsch, W. A. Role of free radicals in the toxicity of airborne fine particulate matter. Chem. Res. Toxicol. 2001, 14, 13711377. Geiser, T.; Ishigaki, M.; vanLeer, C.; Matthay, M. A.; Broaddus, V. C. H2O2 inhibits alveolar epithelial wound repair in vitro by introduction of apoptosis. Am. J. Physiol. Lung Cell. Mol. Physiol. 2004, 287, L448-L453. Morio, L. A.; Hooper, K. A.; Brittingham, J.; Li, T. H.; Gordon, R. E.; Turpin, B. J. Tissue injury following inhalation of fine particulate matter and hydrogen peroxide is associated with altered production of inflammatory mediators and antioxidants by alveolar macrophages. Toxicol. Appl. Pharmacol. 2001, 177, 188-199. Hasson, A. S.; Orzechowska, G. E.; Paulson, S. E. Production of stabilized criegee intermediates and peroxides in the gas phase ozonolusis of alkenes: 1 Ethene, trans-2-butene and 2,3dimethyl-2-butene. J. Geophys. Res. 2001, 106, 34, 131-134, 142. Jackson, A. V.; Hewitt, C. N. Atmospheric hydrogen peroxide and organic hydroperoxides: a review. Crit. Rev. Environ. Sci. Technol. 1999, 29, 175-228. Sander, R. Compilation of Henry’s Law Constants for Inorganic and Organic Species of Potential Importance in Environmental Chemistry, version 3; 1999; http://www.mpch-mainz.mpg.de/ sander/res/henry.html. Hewitt, C. N.; Kok, G. L. Formation and occurrence of organic hydroperoxides in the troposphere. Laboratory and field observations. J. Atmos. Chem. 1991, 12, 181-194. Hung, H. F.; Wang, C. S. Experimental determination of reactive oxygen species in Taipei aerosols. J. Aerosol Sci. 2001, 32, 12011211. Hasson, A. S.; Paulson, S. E. An investigation of the relationship between gas phase and aerosol borne hydroperoxides in urban air. J. Aerosol Sci. 2003, 34, 459-468. Venkatachari, P.; Hopke, P. K.; Grover, B. D.; Eatough, D. J. Measurement of particle bound reactive oxygen species in Rubidoux aerosols. J. Atmos. Chem. 2005, 50, 49-58. Kim, S.; Jaques, P.; Chang, M. C.; Froines, J. R.; Sioutas, C. A versatile aerosol concentrator for simultaneous in vivo and in vitro evaluation of toxic effects of coarse, fine and ultrafine particles. J. Aerosol Sci. 2001, 32, 1281-1297. Misra, C.; Geller, M. D.; Shah, P.; Sioutas, C.; Solomon, P. A. Development and evaluation of a continuous coarse (PM10PM2.5) particle monitor. J. Air Waste Manage. Assoc. 2001, 51, 1309-1317. Chung, M. Y.; Maris, C.; Krischke, U.; Ho, A. W.; Paulson, S. E. An investigation of the relationship between total non-methane

VOL. 40, NO. 16, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4865

(30) (31) (32)

(33)

(34) (35) (36)

(37)

organic carbon and the sum of speciated hydrocarbons and carbonyls measured by standard GC/FID. Atmos. Environ. 2003, 37, S159-S170. Hartkamp, H.; Bachhausen, P. A method for the determination of hydrogen peroxide in air. Atmos. Environ. 1987, 21, 22072213. Hellpointner, E.; Gab, S. Detection of methyl, hydroxymethyl and hydroxyethyl hydroperoxides in air and precipitation. J. Atmos. Chem. 1989, 12, 181-194. Sloane, C. S.; Wolff, G. T. Prediction of ambient light scattering using a physical model responsive to relative humidity: validation with measurements from Detroit. Atmos. Environ. 1985, 19, 669-680. Sauer, F.; Schuster, G.; Schafer, C.; Moortgat, G. K. Determination of H2O2 and organic peroxides in cloud- and rainwater on the Kleiner-Feldberg during FELDEX. Geophys. Res. Lett. 1996, 23, 2605-2608. Seinfeld, J. H.; Pandis, S. Atmospheric Chemistry and Physics; Wiley-Interscience: New York, 1997. Lind, J. A.; Kok, G. L.Henry’s law determination for aqueous solutions of hydrogen peroxide, methylhydroperoxide and peroxyacetic acid. J. Geophys. Res. 1986, 91, 7889-7895. Chung, M. Y.; Muthana, S.; Rodhelen, P. N.; Hasson, A. S. Measurements of effective Henry’s law constants for hydrogen peroxide in concentrated salt solutions. Atmos. Environ. 2005, 39, 2981-2989. Graedel, T. E.; Mandich, M. L.; Weshler, C. J. Speciation, photosensitivity and reactions of transition metals in atmospheric droplets. J. Geophys. Res. 1986, 91, 5189-5204.

4866

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 16, 2006

(38) Faust, B. C.; Zepp, R. G.; Hoigne, J. Hydroxyl radical formation in aqueous reactions (pH 3-8) of iron(II) with hydrogen peroxide: the photo Fenton reaction. Environ. Sci. Technol. 1992, 26, 313-319. (39) Cho, A. K.; DiStefano, E.; You, Y.; Rodriguez, C. E.; Shmidt, D. A.; Kumagai, Y.; Miguel, A. H.; Eiguren-Fernandez, A.; Kobayashi, E.; Avol, E.; Froines, J. R. Determination of four quinones in diesel exhaust particles, SRM 1649a, and atmospheric PM2.5. Aerosol. Sci. Technol. 2004, 38, 68-81. (40) Sedlak, D. L.; Hoigne, J.; David, M. M.; Colvile, R. N.; Seyffer, E.; Acker, K.; Wepercht, W.; Lind, J. A.; Fuzzi, S. The cloudwater chemistry of iron and copper at Great Dunn Fell, U.K. Atmos. Environ. 1997, 31, 2515-2526. (41) Kieber, R. J.; Skrabal, S. A.; Smith, B. J.; Wiley, J. D. Organic complexation of Fe(II) and its impact on the redox cycling of iron in rain. Environ. Sci. Technol. 2005, 39, 1576-1583. (42) Hoigne, J.; Zuo, Y. Formation of hydrogen peroxide and depletion of oxalic acid in atmospheric water by photolysis of iron(III) oxalato complexes. Environ. Sci. Technol. 1992, 26, 1014-1022. (43) Pehkonen, S. O.; Siefert, R.; Erel, Y.; Webb, S. M.; Hoffmann, M. R. Photoreduction of iron oxyhydroxides in the presence of important organic compounds. Environ. Sci. Technol. 1993, 27, 2056-2062.

Received for review July 15, 2005. Revised manuscript received April 3, 2006. Accepted June 12, 2006. ES0513786