Environ. Sci. Technol. 2001, 35, 1294-1298
Haloacetates in Fog and Rain ANDREAS RO ¨ M P P , †,‡ O T T O K L E M M , § WOLFGANG FRICKE,| AND H A R T M U T F R A N K * ,† Environmental Chemistry and Ecotoxicology, and Bayreuth Institute for Terrestrial Ecosystem Research (BITO ¨ K), University of Bayreuth, D-95440 Bayreuth, Germany, and Meteorological Observatory, German Weather Service, D-82383 Hohenpeissenberg, Germany
Atmospheric haloacetates can arise from photochemical degradation of halogenated hydrocarbons and from direct anthropogenic emissions. Furthermore, there is also evidence of natural sources although these are quantitatively uncertain. As haloacetates are highly soluble in water, hydrometeors are most significant for their deposition. Fogwater (96 samples) and rainwater samples (over 100 samples) were collected from July 1998 to March 1999 at an ecological research site in northeastern Bavaria, Germany. They were analyzed for monofluoroacetate (MFA), difluoroacetate (DFA), trifluoroacetate (TFA), monochloroacetate (MCA), dichloroacetate (DCA), trichloroacetate (TCA), monobromoacetate (MBA), and dibromoacetate (DBA). The major inorganic ions were also determined. High concentrations of up to 11 µg/L MCA, 5 µg/L DCA, 2 µg/L TCA, and 2 µg/L TFA were found in fogwater associated with westerly winds. Backward trajectories were calculated to determine the origin of the air masses. MBA and DBA have highest concentrations in fogwater advected with air originating from the Atlantic, suggesting the marine origin of these two compounds. All analyzed substances show higher average concentrations in fog than in rain. Estimates of the deposition of haloacetates suggest that the contribution of fog may be more important than rain for the total burden of a forest ecosystem.
Introduction Haloacetates are polar, water-soluble, airborne micropollutants commonly found in the global atmosphere and hydrosphere, the most abundant being trifluoro- (TFA), monochloro- (MCA), dichloro- (DCA), trichloro- (TCA), and monobromoacetate (MBA). Monofluoro- (MFA), chlorodifluoro- (CDFA), and dibromoacetate (DBA) have also been detected (1, 2). Up to now, it is uncertain which of the haloacetates are anthropogenic or whether biogenic or geogenic sources contribute to their abundance in the environment. MFA is found in plants of the Southern Hemisphere at toxic levels and is present in many other plants in trace amounts (3). TFA has been believed to be predominantly an atmospheric oxidation product of the chlorofluo* Corresponding author phone: +49 921 552373; fax: +49 921 552334; e-mail:
[email protected]. † Environmental Chemistry and Ecotoxicology, University of Bayreuth. ‡ Present address: Department of Atmospheric Chemistry, MaxPlanck-Institut fu ¨ r Chemie, D-55020 Mainz, Germany. § Bayreuth Institute for Terrestrial Ecosystem Research, University of Bayreuth. | German Weather Service. 1294
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 7, 2001
rocarbon (CFC) substitutes 1,1-dichloro-2,2,2-trifluoroethane (HCFC-123), 1-chloro-1,2,2,2-tetrafluoroethane (HCFC-124), and 1,1,1,2-tetrafluoroethane (HFC-134a) (4-6). However, TFA turned out to occur at such high levels in the global environment that other sources must also be taken into consideration (1, 7). CDFA may be a product of the OHinitiated oxidation of 1-chloro-1,1-difluoroethane (HCFC142b) (8). Chloroacetates can be formed via atmospheric breakdown of the airborne C2-chlorocarbons, trichloroethene, tetrachloroethene, and 1,1,1-trichloroethane that are used as solvents for degreasing and dry-cleaning (9); however, industrial production of 1,1,1-trichloroethane has been discontinued because of its ozone-depleting potential. The annual production volumes of tri- and tetrachloroethene have progressively decreased over the past decade. Other possible sources of haloacetates are waste incineration (10), breakdown of fluorinated pesticides (11, 12), or water chlorination (e.g., refs 13 and 14). For TCA, biological processes have also been suggested to contribute (15); the strengths of these sources are difficult to assess however. Bromoacetates are suggested to be atmospheric degradation products of brominated hydrocarbons released to the atmosphere by marine organisms (16). Some of the haloacetates are known for their relatively high phytotoxicological potential (17-20). There is evidence that forest trees at medium elevation of mountain ranges in Central Europe exposed to air masses advectively transporting water in the condensed form as clouds or fog exhibit pronounced phytotoxicological symptoms, although numerous other primary and secondary air pollutants may be involved in the induction (21-23). Medium to weak organic acids are highly mobile in vascular plants and are easily taken up, either directly via the canopy or the roots. Secondary air pollutants that have water solubilities and octanol/water partition coefficients (24, 25) that enable their equilibration into the atmospheric condensed water phase are also wellsuited for uptake by plant cells (26). Many compounds relevant in a phytotoxicological sense are predominantly present in fog and cloud droplets (27). In mountain ranges of about 600-800 m above sea level (asl), fog may contribute significantly to the total deposition of water to forests (28, 29) and, consequently, to the xenobiotic load of these ecosystems. So far, no data are available for concentration of haloacetates in fogwater; only TFA has been determined in California fog (30, 31). The present study was undertaken (i) to gain insight into the anthropogenic or natural origin of the major haloacetates and (ii) to estimate the significance of fog and/or rain for the total deposition of these compounds.
Experimental Section Fogwater and rain samples were collected at an ecological research site located in a spruce forest (Picea abies (L.) Karst, average canopy height 18 m) in northeast Bavaria (50°08′40′′ N, 11°51′55′′ E) at an elevation of 776 m asl (32). The fogwater collectors were installed on a meteorological tower 24 m above ground. At this site, fog most often represents clouds advected from the west crossing France and Western Germany with their highly diverse land-use patterns of agricultural areas, industrial regions, and forest patches. The average wind speed during fog events was 3.2 m/s (0.5-8.1 m/s). Fogwater samples have been collected during each fog event between July 1998 and March 1999, lasting 1270 h in total (defined as visibility being below 500 m). Precipitation of 855 mm in the form of rain and snow was registered during this time. Between July and November 1998, fogwater was 10.1021/es0012220 CCC: $20.00
2001 American Chemical Society Published on Web 02/20/2001
sampled using a Caltech Active Strand Cloudwater Collector (CASCC) (33, 34) with which fog droplets are collected from an airstream by impaction on Teflon strands. From December 1998 through March 1999, a Caltech Heatable Rods Cloudwater Collector (CHRCC) was used (34, 35); the latter features heatable steel rods to allow sampling at subzero temperatures. The collectors were triggered automatically by means of a visibility detector (Present Weather Detector PWD 11, Vaisala, Helsinki, Finland). Ninety-six fogwater and more than 100 rainwater samples were collected. The samples were stored frozen in 250-mL polypropylene (PP) bottles (Nalgene, Rochester, NY) at -20 °C. Sample preparation was performed in PP test tubes (25.5 × 92 mm, Nalgene, Rochester, NY) as described previously (36). After addition of internal standards (pentafluoropropionic acid (PFP) and 2,2-dichloropropionic acid (DCP)) the haloacetates were extracted with tert-butyl methyl ether, derivatized with 1-(pentafluorophenyl)diazoethane, and determined by gas chromatography (HP 5890 Series II, Hewlett-Packard, Waldbronn, Germany) coupled to a mass spectrometer (HP 5989 A, MS-Engine, Hewlett-Packard, Waldbronn, Germany) in the negative chemical ionization and selected ion monitoring (SIM) modes (36). The ions monitored were m/z 77 (MFA), 93 (MCA), 95 (DFA), 113 (TFA), 127 (DCA), 129 (CDFA), 139 (MBA), 141 (DCP), 161 (TCA), 163 (PFP), and 217 (DBA). Limits of detection and quantification are defined as 3- and 7-fold of the standard deviations above the respective analytical system blank values (deionized water). These were determined for each sample series separately because of high day-to-day variability. Values of sampling system blanks (rinsing of the fog collectors with deionized water) were in the range of the analytical blank values. There was no TFA contamination from the Teflon parts of the fog collector. Concentrations of sodium, potassium, magnesium, and calcium were determined by inductively coupled plasmaatomic emission spectroscopy (32); those of ammonium, chloride, nitrate, and sulfate were determined by ion chromatography. Conductivity and pH of the samples were also determined. The origin of the air masses associated with the fog events in October 1998 were analyzed using 72-h backward trajectories for three different pressure levels (1000, 900, and 850 hPa) utilizing the European model of the German Weather Service (37, 38). As sampling took place at an elevation of 800 m asl, the two lower trajectories (1000 and 900 hPa) were used for characterization of the origin of air masses.
TABLE 1. Haloacetate Concentrations in Fog median (µg/L) MFA TFA CDFA MCA DCA TCA MBA DBA
0.23 0.54 0.60 0.28 0.30 0.14
min/max (µg/L)
na
0.05/0.32 0.02/1.9 0.02/0.05 0.28/11 0.12/5.0 0.02/2.0 0.02/1.0 0.02/0.45
2 85 10 44 64 81 24 27
toxicity (µg/L) 55b 120c 7d 10e 40f
a n, number of samples in which the respective haloacetate was found above blank level. b NOEC, cell multiplication inhibition test, Scenedesmus quadricauda (17). c NOEC, cell multiplication inhibition test, Scenedesmus capricornutum (18). d EC10, cell multiplication inhibition test, Scenedesmus subspicatus (19). e LOEC, Somatochlora cingulata (20). f EC10, cell multiplication inhibition test, Scenedesmus subspicatus (19).
FIGURE 1. Dependence of MBA concentrations (µg/L) on wind direction.
Results The concentrations of the haloacetates in fogwater are given in Table 1. MCA, DCA, and TCA were present at highest levels, followed by TFA and MBA. Concentrations were higher in fogwater advectively transported by westerly winds, but the west/east ratios of the haloacetate concentrations differed considerably. For example, Figure 1 shows the dependence of MBA upon wind direction. Of a total of 87 fogwater samples, 76 were collected during westerly winds; in these latter samples, the 90th percentile of MBA concentration was 0.44 µg/L, whereas for those associated with easterly winds (11 samples) it was 0.05 µg/L. For TFA (Figure 2), the corresponding concentrations were 0.65 and 0.43 µg/L, respectively. A backward trajectory typical for marine air masses is shown in Figure 3. The fog events of October 1998 were divided as being of “marine” or “nonmarine” nature according to their backward trajectories (Table 2). Those which had been over the open sea (North Atlantic or North Sea) for more than one-fourth of the last 72 h before arrival at the sampling site were regarded as marine, although they had been influenced by different terrestrial sources on their way
FIGURE 2. Dependence of TFA concentrations (µg/L) on wind direction. to the Waldstein site. Besides the marine tracers sodium and chloride, MBA, and DBA predominated in fog associated with marine air, while in nonmarine air masses the median VOL. 35, NO. 7, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1295
levels than fog. For TCA, the average fog/rain ratio was 4.4. Similar behavior was observed for MCA, DCA, MBA, and DBA.
Discussion The chloroacetates MCA, DCA, and TCA were all present at fairly high levels in fogwater collected at the Waldstein Mountain (Table 1). MBA and DBA were mainly found in samples of marine nature. The persistent TFA was found in all samples. Differentiation between marine and nonmarine events (Table 2) allows us to speculate about the sources of the fogwater constituents. Obviously, air masses characterized as marine have passed over heavily populated areas before arrival at the sampling site, so some influence of anthropogenic sources is to be expected. Nevertheless, the fact that MBA and DBA concentrations in samples with 72-h backward trajectories originating over the Atlantic were much higher than in those originating in continental Europe suggest that these two bromoacetates or their precursors are of marine origin. Another indication is the strong dependence of MBA concentrations in fogwater on wind direction (Figure 1). These observations are consistent with analyses of ambient air samples collected at Mace Head at the Atlantic Coast of Ireland (16). Further information can be derived from the correlation analysis (Table 3). The precursors of sulfate and nitrate are emitted predominantly by power plants, industrial facilities, and motor vehicles. Positive correlations of TFA, DCA, and TCA to these compounds suggest that the former or their respective precursors are of similar, i.e., anthropogenic origin. This is in accordance with observations of Scott et al. (39) that these three haloacetates are higher in areas of high population density. For TCA, a significant correlation was also found to Cl- but not to Na+, indicating the involvement of chloride in the atmospheric formation of trichloroacetate. It is known that the conversion of tetrachloroethene to TCA requires chlorine radicals (40, 41) that are generated by the reaction of chloride with OH radicals (42). MBA was the only substance significantly correlated to both marine indicators sodium and chloride (95% significance level). For DBA there was no correlation at the 95% significance level, but the mean concentration was 4-fold higher in marine-influenced fog as compared to nonmarine air masses. Taking into account the results of the trajectory study and the correlation analysis, the air masses that had been influenced by marine sources show a pattern different to that of nonmarine air masses and indicate that the bromoacetates are of marine origin. Possible precursors could be brominated hydrocarbons that are produced by marine algae (43). The high concentrations of TFA and TCA in the nonmarine air masses as well as their significant correlation to nitrate and sulfate suggest that these or their precursors are of terrestrial and/or anthropogenic origin. MCA and DCA also seem to be of terrestrial origin. Concentrations of haloacetates were higher in fog than in rain (Figures 4 and 5), similar to inorganic electrolytes
FIGURE 3. Trajectories of October 20, 1998 (6 p.m. UTC).
TABLE 2. Concentrations of Major Electrolytes and Haloacetates in Fogwater Associated with Marine (cm) and Nonmarine Air (cn) in October 1998a marine range
nonmarine
median n
range
median n
Milligrams per Liter (mg/L) ammonium sodium chloride nitrate sulfate
0.3-33 0.3-3.3 0.7-4.8 2.5-70 1.4-40
2.5 1.1 2.3 7.5 4.0
7 6 6 7 7
0.5-13 0.3-0.9 0.6-1.7 1.1-24 3.9-45
6.5 0.3 1.1 14 10
7 3 5 7 7
Micrograms per Liter (µg/L) trifluoroacetate 0.02-0.93 trichloroacetate nq-1.25 monobromoacetate 0.19-1.0 dibromoacetate 0.02-0.46
0.10 0.09 0.83 0.20
7 0.02-0.42 7 nq-0.29 3 0.03-0.05 5 0.03-0.10
0.23 0.16 0.04 0.05
7 7 5 4
a For each component, the median higher in marine or nonmarine air, respectively, is printed in bold letters. n, number of samples. nq, not quantified.
concentrations of ammonium, nitrate, sulfate, TCA, and TFA were higher. The fogwater samples in October 1998 were further analyzed for correlation of the haloacetates to inorganic electrolytes, as shown in Table 3. Significance levels were calculated using the Spearman Rank Correlation test. In Table 4, concentration ratios of the major electrolytes in fogwater versus rainwater are shown. For most ions, average concentrations were considerably higher in fogwater than in rain. This was particularly pronounced for ammonium (median fog/rain ratio ca. 13) while potassium exhibited only marginal differences (median fog/rain ratio 1.6). High variability between individual days was observed for protons, but there was no obvious tendency. Concentrations of TFA and TCA in fog and rain are presented in Figures 4 and 5. On average, TFA concentrations in fog were 6.6-fold those of rain. Only on three occasions (August 28, October 2, October 9) out of 14, rain had similar
TABLE 3. Results of Correlation Analysis of Haloacetate and Inorganic Ions in Fogwater Samples of October 1998a Na+
TFA TCA DCA MCA DBA MBA a
Cl-
NO3-
SO42-
NH4+
(n)
P (%)
(n)
P (%)
(n)
P (%)
( n)
P (%)
( n)
P (%)
(n)
ns ns ns ns ns 95
(16) (15) (12) (8) (11) (9)
ns 99 ns ns ns 95
(17) (16) (13) (9) (12) (10)
99 99 95 ns ns ns
(23) (20) (17) (11) (13) (11)
99 99 99 95 ns ns
(23) (20) (17) (11) (13) (11)
99 99 99 95 ns ns
(40) (38) (31) (24) (23) (21)
99 99 99 95 ns ns
(22) (19) (16) (10) (13) (11)
Figures are the significance levels for the rank correlation of two given species. ns, not significant.
1296
9
conductivity
P (%)
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 7, 2001
FIGURE 4. TFA concentrations in fogwater and rain samples.
FIGURE 5. TCA concentrations in fogwater and rain samples.
TABLE 4: Concentration Ratios (Fog/Rain) of Inorganic Ions, TCA, and TFA concentration ratios H+
NH4+ Ca2+ K+ Mg2+ Na+ ClNO3SO42TFA TCA conductivity
median
min
max
1.2 13 2.1 1.6 2.5 4.0 2.5 5.3 6.1 6.6 4.4 5.8
0.1 1.1 1.0 1.0 1.6 1.7 1.2 0.5 1.2 0.6 1.4 0.7
41 90 14 3.8 10 17 7.1 18 20 18 9.7 25
(e.g., refs 45 and 46); this was also reflected by the conductivity ratios of fog and rain. The difference may be explained by the cloud/fog formation process during which cloud condensation nuclei (CCN) deliquesce, i.e., take up water in the condensed form. When a critical radius for an individual particle is reached, its growth depends only on the availability
of gas phase water. If condensable water vapor is limited, droplet growth will stop and a steady-state droplet size distribution will be established. Under such conditions, water-soluble substances will remain at relatively high concentrations. When rain forms, this takes place at higher altitudes and the droplets grow larger by condensation, leading to decreased concentrations of water-soluble substances. Because of their high water solubility, haloacetates will be predominantly present in the liquid phase even during the early stages of droplet formation. Using published Henry constants (24, 25) to calculate the partitioning between gaseous and condensed water phase during the fog events, the latter was found to contain 94-99.9% and 99.4-99.9% of the total TFA and TCA, respectively. Therefore, as droplets grow in an atmosphere already denuded of haloacetates, the deliquescence process entails dilution. TFA and TCA exhibit fog/rain ratios similar to those of nitrate and sulfate (Table 4), suggesting that they might arise from similar sources as precursors rather than as direct emissions. The concentrations of TFA were in the same range as reported for California fog by Wujcik et al. (30, 31). A decrease of TFA concentrations with continued collection as reported by these authors (30) was not observed in our study, neither VOL. 35, NO. 7, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1297
for fog nor rain. Dependence of fogwater concentrations on sample volume was not observed. In contrast to the static radiation fog collected in California, the fog events at the Waldstein site represents clouds that are transported to the sampling site. In Table 1, ecotoxicological data reported in the literature are given for several of the haloacetates. MCA is the ecotoxicologically most relevant haloacetate, the concentrations in fog being in the range of the EC10 value for growth inhibition of an algal species; so far no systematic studies have been performed to assess its phytotoxicity to higher plants although MCA is known to be a nonselective herbicide. Also relevant is TCA, with an algal LOEC only about 5-fold higher than the highest concentration observed in fog. For terrestrial plants, i.e., pine and spruce, the EC10 value has been determined as 140 mg/kg of soil (20). Applying a security factor of 100 as typical for risk assessment procedures, the concentrations of TFA, MCA, and TCA all exceed the projected no-effect concentration (PNEC). Thus, haloacetates in fog must be regarded as ecotoxicologically problematic in the sequence MCA > TCA > TFA. Deposition of fogwater has been estimated to account for 25-50% of the total atmospheric water input at the Waldstein site (32). Based upon the average rain and fogwater concentrations, fog is estimated to contribute between 40 and 70% of the total deposition of TCA and even 70-90% for that of TFA. This indicates that fog may be more important than rain for the input of these acids into forest ecosystems. However, these are preliminary data that need to be supported by additional experimental work. In summary, it becomes apparent that some haloacetates occur in ecotoxicologically relevant levels in hydrometeors, particularly in fog. While TFA and the chloroacetates are likely to be of continental, i.e., possibly anthropogenic, origin the brominated acetates are likely to be of marine, i.e., natural, origin.
Acknowledgments This research was supported by the German Federal Ministry for Education and Research (BMBF) through Grant PT BEO 51-0339476C. We gratefully acknowledge the assistance of T. Haas, T. Wrzesinsky, and J. Gerchau during the field experiments. Thanks are due to J. Collett (Colorado State University) for lending the heatable fog collector.
Literature Cited (1) Jordan, A.; Frank, H. Environ. Sci. Technol. 1999, 33, 522-527. (2) Reimann, S.; Grob, K.; Frank, H. Environ. Sci. Technol. 1996, 30, 2340-2344. (3) Vartiainen, T.; Takala, K.; Kauranen, P. In Selected and Edited Proceedings of the First Conference on Naturally Produced Organohalogens; Grimvall, A., de Leer, E. W. B., Eds.; Kluwer Academic: Dordrecth, 1995; pp 245-250. (4) Franklin, J. Chemosphere 1993, 27, 1565-1601. (5) Edney, E. O.; Gay, B. W.; Driscoll, D. J. J. Atmos. Chem. 1991, 12, 105-120. (6) Edney, E. O.; Driscoll, D. J. Int. J. Chem. Kinet. 1992, 24, 10671081. (7) Frank, H.; Klein, A.; Renschen, D. Nature 1996, 382, 34. (8) Martin, J. W.; Franklin, J.; Hanson, M. L.; Solomon, K. R.; Mabury, S. A.; Ellis, D. A.; Scott, B. F.; Muir, D. C. G. Environ. Sci. Technol. 2000, 34, 274-281. (9) Jordan, A.; Frank, H. Atmos. Environ. 1999, 33, 4525-4526. (10) Mowrer, J.; Nordin, J. Chemosphere 1987, 16, 1181-1192.
1298
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 7, 2001
(11) Ellis, D. A.; Marbury, S. A. Environ. Sci. Technol. 2000, 34, 632637. (12) Key, B. D.; Howell, R. D.; Criddle, C. S. Environ. Sci. Technol. 1997, 31, 2445-2454. (13) Williams, D. T.; LeBel, G. L.; Benoit, F. M. Chemosphere 1997, 34, 299-316. (14) Jolley, R. L.; Johnson, J. D. In Water chlorination: Chemistry, environmental impact and health effects; Condie, L. W., Jolley, R. L., Katz, S., Johnson, J. D., Eds.; Lewis: Chelsea, MI, 1990; pp 3-19. (15) Hoekstra, E. J.; de Leer, E. W. B.; Brinkmann, U. A. T. Chemosphere 1999, 38, 2875-2883. (16) Klein, A. Ph.D. Dissertation, University of Bayreuth, 1997. (17) Bringmann, G.; Kuehn, R. Water Res. 1980, 14, 231-241. (18) Berends, A. G.; Boutonnet, J. C.; de Rooij, C. G.; Thompson, R. S. Environ. Toxicol. Chem. 1999, 18, 1053-1059. (19) Ku ¨ hn, R.; Pattard, M. Water Res. 1990, 24, 31-38. (20) Organisation for Economic Cooperation and Development OECD Existing Chemicals Programme. Screening Information Data Sheet. Initial assessment profile: Trichloroacetic acid; 1997. (21) Frank, H. Eur. Photochem. Assoc. Newsl. 1985, April/June, 7-13. (22) Klo¨pffer, W.; Rippen, G. Environ. Sci. Technol. 1987, 21, 1141. (23) Natangelo, M.; Mangiapan, S.; Bagnati, R.; Benfenati, E.; Fanelli, R. Chemosphere 1999, 38, 1495-1503. (24) Bowden, D. J.; Clegg, S. L.; Brimblecombe, P. Chemosphere 1996, 32, 405-420. (25) Bowden, D. J.; Clegg, S. L.; Brimblecombe, P. J. Atmos. Chem. 1998, 29, 85-107. (26) Bromilow, R. H.; Chamberlain, K.; Evans, A. A. Weed Sci. 1990, 38, 305-314. (27) Glotfelty, D. E.; Seiber, J. N.; Liljedahl, L. A. Nature 1987, 325, 602-605. (28) Grunow, J. Forstwiss. Centralbl. 1955, 74, 21-36. (29) Baumgartner, A. Forstwiss. Centralbl. 1958, 77, 257-320. (30) Wujcik, C. E.; Zehavi, D.; Seiber, J. N. Chemosphere 1998, 36, 1233-1245. (31) Wujcik, C. E.; Cahill, T. M.; Seiber, J. N. Environ. Sci. Technol. 1999, 33,1747-1751. (32) Wrzensinsky, T.; Klemm, O. Atmos. Environ. 2000, 34, 14871496. (33) Daube, B. C., Jr.; Flagan, R. C.; Hoffmann, M. R. United States Patent No. 4697462, 1987. (34) Demoz, B. B.; Collett, J. L., Jr.; Daube, B. C., Jr. Atmos. Res. 1996, 41, 47-62. (35) Collett, J. L., Jr.; Daube, B. C., Jr.; Hoffmann, M. R. Atmos. Environ. 1990, 24A, 959-972. (36) Frank, H.; Renschen, D.; Klein, A.; Scholl, H. J. High Resolut. Chromatogr. 1995, 18, 83-88. (37) Kottmeier, C.; Fay, B. J. Geophys. Res. 1998, 103 (D9), 1094710959. (38) Stohl, A. Atmos. Environ. 1998, 32, 947-966. (39) Scott, B. F.; Mactavish, D.; Spencer, C.; Strachan, W. M. J.; Muir, D. C. G. Environ. Sci. Technol. 2000, 34, 4266-4272. (40) Tuazon, E. C.; Atkinson, R.; Aschmann, S. M.; Goodman, M. A.; Winer, A. M. Int. J. Chem. Kinet. 1988, 20, 241-265. (41) Itoh, N.; Kutsuna, S.; Ibusuki, T. Chemosphere 1994, 28, 20292040. (42) Zetzsch, C.; Pfahler, G.; Behnke, W. J. Aerosol Sci. 1988, 19, 12031206. (43) Gribble, G. W. Environ. Sci. Poll. Res. 1999, 7, 37-48. (44) Igawa, M.; Tsutsumi, Y.; Mori, T.; Okochi, H. Environ. Sci. Technol. 1998, 32, 1566-1572. (45) Lovett, G. M.; Kinsman, J. D. Atmos. Environ. 1990, 24A, 27672786. (46) Gordon, C. A.; Herrera, R.; Hutchinson, T. C. Atmos. Environ. 1994, 28, 323-337.
Received for review April 28, 2000. Revised manuscript received November 21, 2000. Accepted December 21, 2000. ES0012220
