Sterols and Anionic Surfactants in Urban Aerosol - ACS Publications

May 11, 2005 - pre-aeration tank of a municipal WWTP and at two urban locations ... wastewater irrigation facility was a minor source of aerosol- boun...
0 downloads 0 Views 631KB Size
Download PDF
Environ. Sci. Technol. 2005, 39, 4391-4397

Sterols and Anionic Surfactants in Urban Aerosol: Emissions from Wastewater Treatment Plants in Relation to Background Concentrations† MICHAEL RADKE* Department of Hydrology, University of Bayreuth, 95440 Bayreuth, Germany

Aerosol particles that are emitted from aeration tanks of wastewater treatment plants (WWTPs) can be enriched with environmentally relevant wastewater constituents. In this study, aerosol particles were sampled simultaneously at the pre-aeration tank of a municipal WWTP and at two urban locations approximately 1 km away from the WWTP to evaluate the significance of these aerosol emissions. Moreover, aerosol particles were sampled at a small wastewater irrigation facility and at a rural site. In aerosol particles and wastewater, six sterols (cholesterol, coprostanol, campesterol, β-sitosterol, stigmasterol, stigmastanol) and anionic surfactants (expressed in terms of methylene blue active substances, MBAS) were quantified. The results showed significantly higher concentrations of sterols and MBAS at the WWTP than at the urban locations. At the WWTP, average concentrations of cholesterol (848 ( 321 pg m-3), coprostanol (1132 ( 565 pg m-3), and MBAS (132 ( 43 ng m-3) in aerosol were approximately twice as high. This can be attributed to emissions from the treatment tank. Coprostanol, a unique tracer for wastewater, was detected only occasionally at the urban locations. The results of this study show that the aeration of wastewater is a continuously operating local source for organic compounds in aerosol. The wastewater irrigation facility was a minor source of aerosolbound sterols (coprostanol, 287 ( 218 pg m-3) and anionic surfactants (64 ( 32 ng m-3). Except for coprostanol, all compounds were also present in samples of rural aerosol.

Introduction Organic trace compounds can be emitted from aeration tanks of wastewater treatment plants (WWTPs) via the formation of aerosol particles by bursting air bubbles (1, 2). Due to their enrichment at the air-water interface (3, 4), nonpolar or surface-active compounds will be preferentially transferred into aerosol particles compared to the bulk water. This process could lead to an input of environmentally relevant compounds into the urban atmosphere and thus to an exposure of humans. So far, however, the concentrations of such compounds in aerosol at WWTPs have not been compared to background data, and therefore, the impact of * Phone: +49/921/552297; fax: +49/921/552366; e-mail: [email protected]. † In memory of Prof. Dr. Reimer Herrmann (1938-2003). 10.1021/es048084p CCC: $30.25 Published on Web 05/11/2005

 2005 American Chemical Society

emissions from wastewater treatment on urban aerosol could not be properly assessed. The aim of this study was to evaluate the significance of emissions of wastewater-born organic compounds, especially sterols and anionic surfactants, during wastewater treatment by relating concentrations of these compounds in aerosol sampled at a WWTP to aerosol sampled at urban and rural locations. The concentrations of anionic surfactants and six sterols in aerosol were determined simultaneously at an aeration tank of a WWTP and two adjacent urban sites; the corresponding concentrations in the wastewater were measured as well. Additionally, aerosol samples from a small wastewater irrigation facility and a rural site were analyzed. Sterols are constituents of wastewater. Through excretions of humans and animals, these compounds reach the sewer system. Coprostanol (5β-cholestan-3β-ol) is quite unique for wastewater, since it is only formed by bacterial reduction of cholesterol and other sterols in the digestive tract of animals and humans (5). It is widely used as a marker for the influence of wastewater on surface water bodies (6-12), although there exist some studies on nonsewage sources of coprostanol such as feces of marine mammals (13) or biodegradation of cholesterol in anaerobic sediments (14). In urban aerosol, coprostanol has not been determined so far, but Radke and Herrmann (1) demonstrated the occurrence of coprostanol in aerosol particles at a WWTP. Sheesley et al. (15) have recently reported the combustion of cow dung as another potential source of coprostanol and other fecal stanols in aerosol of South Asia. In contrast to coprostanol, cholesterol (cholest-5-en-3β-ol) has been determined in atmospheric aerosol particles before, originating from plants (16, 17), wood smoke (18), and cooking processes (19, 20), for example. The other sterols analyzed in this study were campesterol (24methylcholest-5-en-3β-ol), β-sitosterol (24-ethylcholest-5en-3β-ol), stigmasterol (24-ethylcholest-5,22-dien-3β-ol), and stigmastanol (24-ethyl-5R-cholestan-3β-ol); these compounds which are present in plants (17, 21) or animal feces (12) have already been determined in aerosol particles, originating from terrestrial sources such as erosion of vegetation detritus or biomass burning (22). Another class of compounds analyzed in this study were anionic surfactants, expressed in terms of methylene blue active substances (MBAS). Previous studies on the formation of sea spray have shown that surface-active compounds significantly influence the formation of aerosol particles (23) and that such compounds can be enriched in water-born aerosol particles (24). Anionic surfactants are constituents of household detergents and are present in wastewater in concentrations up to several milligrams per liter (25). Data on MBAS in aerosol particles are rare: Sukhapan and Brimblecombe (26) recently reported concentrations of MBAS in atmospheric aerosols sampled in the U.K. in the range of 6-170 pmol m-3. For marine aerosols, a few studies showing the presence of MBAS are available (27, 28).

Experimental Details Sample Collection and Sampling Sites. Aerosol particles (total suspended solids, TSP) were collected on quartz fiber filters (type QF20, Schleicher und Schu ¨ ll, Dassel, Germany; diameter 150 mm) using high-volume samplers with an air flow of approximately 30 m3 h-1 that was controlled and recorded by a microprocessor. Prior to use, the filters were baked at 280 °C for 12 h to remove organic residues. Aerosol particles were sampled for 48 h each. For the determination of collected particle mass, filters were equilibrated in a VOL. 39, NO. 12, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4391

FIGURE 1. Topographic map of the sampling area at the WWTP Bayreuth. Locations marked A, B, and C: sampling sites. Sites A and C are located at the edge of the shallow valley of the river Roter Main River; site B is located at the pre-aeration tank of the WWTP Bayreuth. desiccator and weighed before and after sampling. After sampling, filters were stored frozen (-20 °C) until analysis. Wastewater was sampled with an automated sampler (3700 standard, ISCO, Franklin). Every 80 min, 100 mL of wastewater was collected in glass bottles; usually, these samples were combined to composite samples for 1 d. Wastewater samples were filtered with glass fiber filters (GF6, Schleicher und Schu ¨ ll) and stored frozen (-20 °C) until extraction. Aerosol particles were sampled at three different sites between January and June 2003: (1) at a conventional WWTP and its surroundings, (2) a wastewater irrigation facility, and (3) a rural site. First, samples were taken simultaneously at the pre-aeration tank of the WWTP of Bayreuth, Germany, with a typical inflow of 23000 m3 d-1 and at two urban locations approximately 1 km from the WWTP (see Figure 1). In contrast to a previous study where the aeration was operating continuously (1), the pre-aeration tank was aerated for intervals of 20 min with a 60 min intermission between each interval. Wastewater was collected directly from the tank as described above. Data on wind speed and direction measured 17 m above ground were recorded by a meteorological station operated by the Department of Micrometeorology at the ecological-botanical garden at the University of Bayreuth located approximately 4.5 km southeast of location B (29). A second sampling site was established at the WWTP of the town Hiltpoltstein, Germany, a small plant (inflow approximately 130 m3 d-1) in a rural area where wastewater is irrigated by sprinklers. Directly next to the irrigation field, aerosol samples were taken as described above. Wastewater was sampled as individual grab samples. The third aerosol sampling site was established at a forested rural site at the meteorological and air chemical station Waldstein situated in northeastern Bavaria, Germany (30). There, no direct influence of residential areas or WWTPs on the aerosol composition was assumed. Aerosol samples were taken with a high-volume sampler as described above. 4392

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 12, 2005

Sample Extraction and Analysis. Half of each quartz fiber filter was analyzed for sterols, one-quarter of each filter for MBAS. Sterols were determined as previously described (1), but with the following modifications. Each filter was spiked with a solution containing an exactly known amount (approximately 1 µg) of cholesterol-d6 (Cambridge Isotope Laboratories, Andover, MA) and 5R-cholestane (SigmaAldrich, Munich, Germany) used as internal standards. The filter was cut into small pieces and extracted with hexane/ acetone (1:1, v/v) using a fluidized bed extractor (five extraction cycles; IKA, Staufen, Germany). After extraction, the solvent was evaporated to a volume of 1 mL with a rotary evaporator. Cleanup of the extract was done with column chromatography as previously described (1, 31), with the modification that the second and third fractions were combined as one fraction that was analyzed. After evaporation and silylation with Sylon BFT (Sigma-Aldrich), samples were redissolved in an internal standard solution, and sterols were measured by GC-MS (for details see ref 1). Wastewater samples were extracted with solid-phase extraction (SPE) on 200 mg of Chromabond-NH2 over 300 mg of Chromabond-C18ec (both Macherey & Nagel, Dueren, Germany). The method is described in detail by Beck and Radke (32). Prior to extraction, the SPE columns were rinsed with dichloromethane, methanol, and pure water, and 4 vol % methanol and a solution containing cholesterol-d6 as internal standard were added to 50 mL of the water sample. The sample was extracted, and the SPE columns were freezedried. Subsequently, the sterols were eluted with dichloromethane/methanol (1:1, v/v), and the samples were evaporated to dryness under a stream of nitrogen. Derivatization and analysis were done as described for aerosol samples. In recovery experiments with spiked wastewater, recovery rates for cholesterol and coprostanol were 46 ( 13% and 84 ( 18% (n ) 34 samples), respectively; the overall recovery of cholesterol-d6 in all samples (n ) 120) was 58 ( 13%. In recovery experiments conducted with pure water as

FIGURE 2. Concentrations of (a) cholesterol, (b) coprostanol, (c) stigmastanol, and (d) campesterol in aerosol collected at the sampling sites marked in Figure 1. The sample Jan 21 at location C was excluded from presentation due to a sampling error. For Feb 6, no data for campesterol are available. matrix, higher recovery rates and lower standard deviations were obtained (32). For the analysis of MBAS in aerosol particles, the German standard method for the analysis of MBAS in wastewater (DIN EN 903 (33)), which is based on the method originally developed by Longwell and Maniece (34), was modified. Aerosol particles were extracted two times with 25 mL of ultrapure water in an ultrasonic bath. After sequential extraction with neutral and acidic solutions containing methylene blue, the absorption of the final solution was measured photometrically at a wavelength of 650 nm. Details on the preparation of solutions and the analytical protocol are described in the literature (33, 34). Calibration was performed using aqueous solutions with known concentrations of sodium dodecylbenzenesulfonate (SDBS; Merck, Darmstadt, Germany) as it is common for wastewater analyses; thus, results are expressed in terms of SDBS equivalents. This compound is a mixture of several homologues with an average molecular mass of 348.5 g mol-1. For the determination of blank values, new quartz fiber filters and ultrapure water were analyzed. Concentrations in blank values were usually below the limit of detection. The method yielded reproducible and high recovery rates (102 ( 8%; n ) 6) independent of the amounts spiked on the filter. For the determination of MBAS in wastewater, filtered wastewater was used, following the same analytical protocol as for aerosol particles, starting after the ultrasonic extraction. This method is well established for the analysis of wastewater (33), so no further tests were conducted. Statistical analyses were carried out using WinSTAT (version 1999.1, R. Fitch Software, Staufen, Germany). Data were tested for normal distribution with the KolmogorovSmirnov test; a paired t test was used for the analysis of differences between concentrations simultaneously deter-

mined at several sampling locations. Pearson correlation was applied for correlation analysis.

Results WWTP Bayreuth. The concentrations of four sterols in aerosol from the WWTP Bayreuth and the two nearby locations are shown in Figure 2. The concentrations of cholesterol in aerosol were significantly higher at location B than at locations A and C (p < 0.05); no significant difference was determined between locations A and C. The average concentration of cholesterol at location B was 848 ( 321 pg m-3; at locations A and C cholesterol was present at 409 ( 138 and 395 ( 148 pg m-3, respectively (Table 1). The highest concentrations of coprostanol were also determined at location B (1132 ( 565 pg m-3); it was present only occasionally at locations A and C at concentrations up to 566 pg m-3. Statistical analyses of differences in the coprostanol concentrations between the three locations were not meaningful due to the low number of samples in which coprostanol was determined at locations A and C. If coprostanol was present, its concentrations at locations A and C were of the same order of magnitude as cholesterol concentrations. For all other sterols, no significant elevation of their aerosol concentration at location B was determined. The concentrations of all sterols in aerosol are shown in Table 1. All compounds showed a high variability with relative standard deviations of approximately 20-50%. At the WWTP (location B), coprostanol was the compound with the highest average concentration in aerosol, followed by β-sitosterol and cholesterol. This was different from the composition of the wastewater, where the concentrations of coprostanol and cholesterol were highest (91 ( 37 and 84 ( 40 µg L-1, respectively) and the concentration of β-sitosterol was much lower (16 ( 9 µg L-1). Other sterols were present VOL. 39, NO. 12, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4393

TABLE 1. Average Aerosol Concentration of Sterols and MBAS in Samples Taken at the Three Locations in Bayreutha coprostanol (pg m-3)

mean min-max mean min-max mean min-max mean min-max mean min-max mean min-max mean min-max

cholesterol (pg m-3) β-sitosterol (pg m-3) campesterol (pg m-3) stigmasterol (pg m-3) stigmastanol (pg m-3) MBAS (ng m-3)

HV-A

HV-B

HV-C

370 ( 169 (7b) 154-566 409 ( 138 (7) 206-638 909 ( 364 (6) 464-1317 245 ( 48 (6) 171-310 328 ( 115 (6) 170-471 331 ( 91 (6) 185-444 76 ( 31 (12) 37-135

1132 ( 565 (7) 656-2252 838 ( 321 (7) 435-1293 945 ( 534 (6) 378-1765 301 ( 61 (6) 207-367 405 ( 81 (6) 325-518 371 ( 84 (6) 262-471 132 ( 43 (12) 57-210

254 (6c) 137-371 395 ( 148 (6) 124-525 664 ( 201 (5) 321-849 232 ( 44 (5) 171-293 266 ( 84 (5) 126-330 301 ( 87 (5) 163-386 75 ( 35 (12) 11-121

a Mean ( standard deviation (number of samples shown in parentheses) and concentration ranges (minimum-maximum). determined in four samples. c Compound was determined in two samples.

TABLE 2. Average Concentrations of Sterols (µg L-1) in Wastewater Sampled from the Pre-aeration Tank of the WWTP Bayreuth and at the Hiltpoltstein Irrigation Sitea WWTP Bayreuth (n ) 24) 91 ( 37 84 ( 40 16 ( 9 6.4 ( 2.0 5.9 ( 2.5 2.4 ( 0.8

cholesterol coprostanol β-sitosterol campesterol stigmasterol stigmastanol a

( standard deviation.

b

Hiltpoltstein (n ) 4) 134 ( 34 157 ( 60 14 ( 2 nab na na

b

Compound was

TABLE 3. Average Concentration of Cholesterol, Coprostanol, and MBAS in Aerosol Samples Taken at the Wastewater Irrigation Site (Hiltpoltstein) and at the Remote Site (Waldstein)a m-3)

coprostanol (pg cholesterol (pg m-3) MBAS (ng m-3)

Hiltpoltstein

Waldstein

278 ( 218 (n ) 5) 148 ( 79 (n ) 5) 64 ( 32 (n ) 3)

nd (n ) 3) 46-52 (n ) 2) 43 (n ) 2)

a Mean value ( standard deviation. n ) number of samples. nd ) not detected.

na ) not analyzed.

WWTP Bayreuth (Table 2). This was true for MBAS, too, with concentrations in the range of 20 mg L-1. The average concentrations of MBAS for aerosol collected at this site are shown in Table 3. Concentrations of cholesterol and coprostanol were 5-10-fold lower than at the WWTP in Bayreuth; MBAS concentrations were also substantially lower. Waldstein. In aerosol particles sampled at the rural site Waldstein, coprostanol was not detected (Table 3), and the concentration of cholesterol was much lower than at the two WWTPs and in urban aerosol (Table 1). MBAS concentrations were of the same order of magnitude as at the irrigation site and substantially lower than at the urban sampling sites.

Discussion

FIGURE 3. Concentrations of MBAS in aerosol collected at the sampling sites marked in Figure 1. The sample Jan 21 at location C was excluded from presentation due to a sampling error. in minor concentrations in wastewater (Table 2). A detailed description of temporal concentration trends in wastewater is given by Beck and Radke (32). The spatial distribution of MBAS (Figure 3) in aerosol particles was quite similar to that of cholesterol: MBAS were present in all aerosol samples, but concentrations determined at the WWTP (132 ( 43 ng m-3) were significantly higher than at locations A and C (p < 0.001; Table 1). The concentration of MBAS in wastewater was in the range of 1-10 mg L-1. Compared to those of cholesterol, the concentrations of MBAS in wastewater and in aerosol particles were about 3 orders of magnitude higher (Tables 1 and 2). WWTP Hiltpoltstein. To evaluate aerosol emissions of a wastewater irrigation facility, aerosol and wastewater samples from the WWTP Hiltpoltstein were analyzed. The average sterol concentrations in wastewater were higher than at the 4394

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 12, 2005

Compared to those in urban aerosol (locations A and C), concentrations of cholesterol and MBAS at the WWTP Bayreuth were significantly increased up to 300% on some days (Figures 2 and 3), which can be attributed to emissions from the WWTP. This conclusion is supported by the correlation of coprostanol concentrations in aerosol and wastewater (r ) 0.91, p < 0.01; Figure 4), which also points to wastewater as the dominant coprostanol source, even though this correlation should be regarded as tentative due to the limited number of samples. This interpretation is supported further by measurements done with a cascading impactor at the same treatment tank (32). For cholesterol and MBAS, no correlation between aerosol and wastewater was observed (not shown). The average MBAS concentrations in rural aerosol (Waldstein) were lower than in urban aerosol (locations A and C; Tables 1 and 3), which indicates a natural background concentration of MBAS that is elevated due to human activities. Regarding the nature of the surfactants or surfactant-like compounds responsible for this background concentration and possible sources, however, no detailed information can be given. Recently, Sukhapan and Brimblecombe (26) and Latif and Brimblecombe (35) reported on

FIGURE 4. Comparison of the time courses of coprostanol concentrations in wastewater and in aerosol particles sampled at the WWTP Bayreuth (location B). MBAS concentrations in mainly semiurban and urban aerosol particles sampled in Europe, Asia, and the United States. Since they used a different reference compound (sodium dodecyl sulfate; molecular mass 288.4 g mol-1), their data are not absolutely comparable to those of our study, but they determined concentrations between 12.8 and 285 pmol m-3, which is of the same order of magnitude as in our samples (100-350 pmol m-3). Obviously, compounds that behave like anionic surfactants are ubiquitously present in atmospheric aerosol. Giovannelli et al. (28) determined MBAS concentrations in marine aerosol that were 3 orders of magnitude higher than in this study and attributed the MBAS to anthropogenic pollution of seawater and subsequent emission into the atmosphere via sea spray. However, the assumption of marine sources for MBAS in aerosol, which has also been proposed by Oppo et al. (27), could not be confirmed by Latif and Brimblecombe (35), who in turn assumed that humic-like substances could be the origin for MBAS in aerosol particles. This could also be an explanation for the MBAS concentrations observed at the Waldstein site. To determine the proportion of anionic surfactants originating from wastewater treatment or other anthropogenic sources, compound-specific analytical methods, such as, for example, described by Holt et al. (36) or Matthijs et al. (37) for the analysis of surface or wastewater, should be applied. In contrast to MBAS, cholesterol concentrations in urban aerosol were much more elevated compared to those in rural aerosol (Tables 1 and 3). This points to substantial anthropogenic emissions of cholesterol in the urban environment; as a potential major cholesterol source, meat cooking has been reported (19, 20). Natural emissions such as the erosion of plant material (38) were obviously only responsible for a low background concentration at the rural site. Compared to other studies that reported concentrations of cholesterol in aerosol in the range 600-23000 pg m-3 for locations outside Europe (16, 18-20) and up to 50000 pg m-3 for a rural site in Portugal (38), the concentrations determined at urban sampling sites in this study were several orders of magnitude lower with a maximum concentration of cholesterol of 640 pg m-3 (Table 1). Even the highest concentration detected at the WWTP (1290 pg m-3) was near the lower end of the concentration range reported. This was also true for stigmasterol: Pio et al. (38) reported maximum concentrations of 34600 pg m-3, whereas in this study, the highest concentration determined in urban aerosol was 518 pg m-3 (Table 1). An explanation for these deviations cannot be given. The results of this study indicate that the only source of coprostanol in aerosol particles was wastewater treatment, since coprostanol was only occasionally detectable in urban aerosol (locations A and C; Figure 2) and it was not detectable

in rural aerosol at all. During a few sampling intervals, transport of coprostanol from the WWTP Bayreuth to locations A and C may have occurred (Figure 2). However, it was unexpected that coprostanol was present at both locations A and C simultaneously because these locations were in quite opposite directions. The prevailing wind direction was relatively constant during each of the sampling periods (data not shown), so we expected to detect aerosol particles emitted from the WWTP only at one of these locations at a time. The most plausible explanation for this finding is an inversion layer that was present during the sampling periods which covered the flat river valley where the samples were taken. Therefore, transport processes near the ground surface could have been uncoupled from the prevailing wind direction, and the meteorological measurements cannot be used for interpretation. Other sources near locations A and C which were temporarily emitting coprostanol, however, also cannot be absolutely excluded as explanations for these observations. A closer look at the results for cholesterol and coprostanol at the WWTP Bayreuth reveals some inconsistency between these two compounds. In wastewater the concentrations of both compounds were of the same order of magnitude (Table 2), whereas in the aerosol the average concentration of coprostanol was higher than that of cholesterol (Table 1). The physicochemical characteristics of both compounds are quite similar, so they should be emitted in relatively equal amounts from wastewater into the atmosphere. Therefore, cholesterolswhich was already present in remarkable concentrations in urban aerosol (120-640 pg m-3 at locations A and C)sshould have a much higher concentration in aerosol at the WWTP than coprostanol. An explanation for this observation cannot be given. In contrast, data from laboratory experiments confirmed similarities between the emission behaviors of cholesterol and coprostanol (31). The other sterols did not have significantly higher concentrations in aerosol at the WWTP Bayreuth than in urban aerosol (Figure 2). This can be explained by their concentrations in wastewater, which were 5-20-fold lower compared to that of cholesterol (Table 2), so the source strength of the WWTP was not high enough to affect the concentration of these compounds in aerosol substantially. The aerosol mass concentrations at the WWTP were not elevated in comparison to those at locations A and C (not shown). This indicates that the emitted particles mainly consisted of water that was evaporated during sampling. To determine the true aerosol mass at such WWTP sites, a different sampling technique that preserves aqueous aerosols should be applied, or aerosol mass or number concentrations should be determined by independent methods, for example, by an electrostatic low-pressure impactor (ELPI) or a scanning mobility particle sizer (SMPS). It was unexpected that the concentrations of sterols and MBAS in aerosol at the wastewater irrigation site were much lower than at the WWTP Bayreuth, since the irrigation of wastewater directly produces water and aerosol droplets and the composition of the wastewater was quite similar at both WWTPs (Table 2). These relatively low emissions of aerosolbound sterols and MBAS at the irrigation site were most probably due to the low volume of wastewater that was irrigated there, so the sprinklers were operating only a few hours per day. In conclusion, this site can be classified as a minor source of aerosol-bound sterols and surfactants compared to the conventional WWTP. However, when the aerosol emissions of both WWTPs are compared, it should be noted that the WWTP Bayreuth received approximately 180 times more wastewater than the WWTP Hiltpoltstein, whereas the coprostanol concentrations in the aerosol were only approximately 4 times higher at the WWTP Bayreuth. Given two WWTPs of comparable size, we would expect VOL. 39, NO. 12, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4395

extremely higher concentrations of sterols and MBAS in aerosol at the plant that applies wastewater irrigation. The sterol concentrations in the aerosol particles sampled at the WWTP Bayreuth are similar to those of Radke and Herrmann (1), who determined concentrations of coprostanol in the range of 1000-13500 pg m-3 and concentrations of cholesterol of 200-3200 pg m-3. The actual concentrations were lower, which was to be expected, since the aeration tank was reconstructed and switched from continuous to intermittent aeration between the two studies. The combustion of cow dung (15) and anaerobic formation of coprostanol (14) can be excluded as other potential sources for coprostanol in aerosol that were analyzed in this study. The results of this study show that aerosol emission from wastewater treatment leads to an increased concentration of nonpolar and amphiphilic organic wastewater constituents in aerosol at WWTPs. The aeration of untreated wastewater is a continuously operating source for organic compounds in urban aerosol particles. However, it is difficult to track and quantify these emissions even over relatively short distances of approximately 1 km, as could be shown for the wastewater marker coprostanol. Concentrations in aerosol are lowered by dispersion and dilution of the aerosol particles to concentration levels at which this marker is no longer detectable. Potential other wastewater markers such as MBAS were already present in relatively high concentrations in atmospheric aerosols, so this parameter could not be used for tracking wastewater-born aerosol particles. Nonetheless, the results of this study raise the question of to what extent other, from an environmental point of view, relevant wastewater constituents such as, for example, pharmaceuticals and personal care products or xenoestrogenic compounds such as nonylphenols are emitted into the urban atmosphere via the formation of aerosol particles during wastewater treatment. Emissions of nonylphenols from wastewater treatment plants have already been reported (2), and these compounds have also been detected in urban and coastal aerosol samples (39-41) in the lower nanogram per cubic meter range. To date, however, the function of WWTPs as sources for atmospheric nonylphenols cannot be properly evaluated. And even if wastewater-born aerosol particles are not transported over longer distances, workers at WWTPs, for example, could be continuously exposed to the emitted compounds by the aspiration of aerosol particles.

Acknowledgments This work was funded by the Bavarian State Ministry of the Environment, Public Health and Consumer Protection. The help of Ramona Stadelmann, Jutta Eckert, Isabel Seifert, and Melanie Beck during this project is gratefully acknowledged. The ISCO sampler was provided by Gunnar Lischeid, BayCEER, and meteorological data were provided by the Department of Micrometeorology, University of Bayreuth.

(7)

(8) (9)

(10)

(11) (12) (13) (14) (15)

(16)

(17) (18)

(19) (20)

(21) (22) (23) (24) (25)

Literature Cited (1) Radke, M.; Herrmann, R. Aerosol-bound emissions of polycyclic aromatic hydrocarbons and sterols from aeration tanks of a municipal waste water treatment plant. Environ. Sci. Technol. 2003, 37, 2109-2113. (2) Lepri, L.; Del Bubba, M.; Masi, F.; Udisti, R.; Cini, R. Particle size distribution of organic compounds in aqueous aerosols collected from above sewage aeration tanks. Aerosol Sci. Technol. 2000, 32, 404-420. (3) Schneider, J. K.; Gagosian, R. B. Particle size distribution of lipids in aerosols off the coast of Peru. J. Geophys. Res. 1985, 90, 7889-7898. (4) Duce, R. A.; Hoffman, E. J. Chemical fractionation at the air-sea interface. Annu. Rev. Earth Planet. Sci. 1973, 4, 187-228. (5) Walker, R. W.; Wun, C. K.; Litsky, W. Coprostanol as an indicator of fecal pollution. Crit. Rev. Environ. Control 1982, 12, 91-112. (6) Elhmmali, M. M.; Roberts, D. J.; Evershed, R. P. Combined analysis of bile acids and sterols/stanols from riverine particu4396

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 12, 2005

(26) (27)

(28) (29)

lates to assess sewage discharges and other fecal sources. Environ. Sci. Technol. 2000, 34, 39-46. Du ¨ reth, S.; Herrmann, R.; Pecher, K. Tracing fecal pollution by coprostanol and intestinal bacteria in an ice-covered Finnish lake loaded with both industrial and domestic sewage. Water, Air, Soil Pollut. 1986, 28, 131-149. Brown, R. C.; Wade, T. L. Sedimentary coprostanol and hydrocarbon distribution adjacent to a sewage outfall. Water Res. 1984, 18, 621-632. O’Leary, T.; Leeming, R.; Nichols, P. D.; Volkman, J. K. Assessment of the sources, transport and fate of sewage-derived organic matter in Port Phillip Bay, Australia, using the signature lipid coprostanol. Mar. Freshwater Res. 1999, 50, 547-556. Leeming, R.; Nichols, P. D. Determination of the sources and distribution of sewage and pulp-fibre-derived pollution in the Derwent Estuary, Tasmania, using sterol biomarkers. Mar. Freshwater Res. 1998, 49, 7-17. Leeming, R.; Bate, N.; Hewlett, R.; Nichols, P. D. Discriminating faecal pollution: A case study of stormwater entering Port Phillip Bay, Australia. Water Sci. Technol. 1998, 38, 15-22. Leeming, R.; Ball, A.; Ashbolt, N.; Nichols, P. Using faecal sterols from humans and animals to distinguish faecal pollution in receiving waters. Water Res. 1996, 30, 2893-2900. Venkatesan, M. I.; Ruth, E.; Kaplan, I. R. Coprostanols in Antarctic marine-sedimentssa biomarker for marine mammals and not human pollution. Mar. Pollut. Bull. 1986, 17, 554-557. Taylor, C. D.; Smith, S. O.; Gagosian, R. B. Use of microbial enrichments for the study of the anaerobic degradation of cholesterol. Geochim. Cosmochim. Acta 1981, 45, 2161-2168. Sheesley, R. J.; Schauer, J. J.; Chowdhury, Z.; Cass, G. R.; Simoneit, B. R. T. Characterization of organic aerosols emitted from the combustion of biomass indigenous to South Asia. J. Geophys. Res. 108 (D9), 4285; doi: 10.1029/2002JD002981. Simoneit, B. R. T.; Mazurek, M. A.; Reed, W. E. Characterization of organic matter in aerosols over rural sites: phytosterols. In Advances in Organic Geochemistry 1981; Bjoroy, M. e. a., Ed.; Wiley: Chichester, England, 1983; pp 355-361. Nishimura, M.; Tadashiro, K. The occurrence of stanols in various living organisms and the behavior of sterols in contemporary sediments. Geochim. Cosmochim. Acta 1977, 41, 379-385. Nolte, C. G.; Schauer, J. J.; Cass, G. R.; Simoneit, B. R. T. Highly polar organic compounds present in wood smoke and in the ambient atmosphere. Environ. Sci. Technol. 2001, 35, 19121919. Nolte, C. G.; Schauer, J. J.; Cass, G. R.; Simoneit, B. R. T. Highly polar organic compounds present in meat smoke. Environ. Sci. Technol. 1999, 33, 3313-3316. Rogge, W. F.; Hildemann, L. M.; Mazurek, M. A.; Cass, G. R.; Simoneit, B. R. T. Sources of fine organic aerosol. 1. charbroilers and meat cooking operations. Environ. Sci. Technol. 1991, 25, 1112-1125. Volkman, J. K. A review of sterol markers for marine and terrigenous organic matter. Org. Geochem. 1986, 9, 83-99. Simoneit, B. R. T. A review of biomarker compounds as source indicators and tracers for air pollution. Environ. Sci. Pollut. Res. 1999, 6, 159-169. Paterson, M. P.; Spillane, K. T. Surface films and the production of sea-salt aerosol. Q. J. R. Meteorol. Soc. 1969, 95, 526-543. Hoffman, E. J.; Duce, R. A. Factors influencing the organic carbon content of marine aerosols: a laboratory study. J. Geophys. Res. 1976, 81, 3667-3670. Feijtela, T. C. J.; Matthijs, E.; Rottiers, A.; Rijs, G. B. J.; Kiewiet, A.; de Nijs, A. AIS/CESIO environmental surfactant monitoring programme. Part 1: LAS monitoring study in “de Meern” sewage treatment plant and receiving river “Leidsche Rijn”. Chemosphere 1995, 30, 1053-1066. Sukhapan, J.; Brimblecombe, P. Ionic surface active compounds in atmospheric aerosols. TheScientificWorldJOURNAL 2002, 2, 1138-1146. Oppo, C.; Bellandi, S.; Innocenti, N. D.; Stortini, A. M.; Loglio, G.; Schiavuta, E.; Cini, R. Surfactant components of marine organic matter as agents for biogeochemical fractionation and pollutant transport via marine aerosols. Mar. Chem. 1999, 63, 235-253. Giovannelli, G.; Bonasoni, P.; Loglio, G.; Ricci, C.; Tesei, U.; Cini, R. Evidence of anionic-surfactant enrichment in marine aerosol. Mar. Pollut. Bull. 1988, 19, 274-277. Department of Micrometeorology, University of Bayreuth. Description of the meteorological tower in the ecological-botanical garden, University of Bayreuth. http://www.bayceer.uni-bayreuth.de/bayceer/de/klima/5407/ BotGar/bginfo.php (accessed March 2005).

(30) Klemm, O. Trace gases and particles in the atmospheric boundary layer at the Waldstein site: Present state and historic trends. In Biogeochemistry of Forested Catchments in a Changing EnvironmentsA German case study; Matzner, E., Ed.; Springer: Heidelberg, Germany, 2004. (31) Radke, M. Aerosolgebundene Emission organischer Umweltchemikalien aus druckbeluefteten Abwasserbehandlungsbecken; Bayreuther Forum O ¨ kologie Vol. 103; BITO ¨ K: Bayreuth, Germany, 2004. (32) Beck, M.; Radke, M. Determination of sterols, estrogens and inorganic ions in waste water and size-segregated waste water born aerosol particles. Chemosphere, submitted for publication. (33) Deutsches Institut fu ¨ r Normung. DIN EN 903 (H24): Bestimmung von anionischen oberfla¨chenaktiven Stoffen durch Messung des Methylenblau-Index MBAS; Deutsches Institut fu¨r Normung e.V.: Berlin, 1994. (34) Longwell, J.; Maniece, W. D. Determination of anionic detergents in sewage, sewage effluents and river waters. Analyst 1955, 80, 167-171. (35) Latif, M. T.; Brimblecombe, P. Surfactants in atmospheric aerosols. Environ. Sci. Technol. 2004, 38, 6501-6506. (36) Holt, M. S.; Waters, S.; Comber, M. H. I.; Armitage, R.; Morris, G.; Newbery, C. AIC/CESIO environmental surfactant monitoring programme. SDIA sewage treatment pilot study on linear alkylbenzene sulphonate (LAS). Water Res. 1995, 29, 2063-2070.

(37) Matthijs, E.; Holt, M. S.; Kiewiet, A.; Rijs, G. B. J. Environmental monitoring for linear alkylbenzene sulfonate, alcohol ethoxylate, alcohol ethoxy sulfate, alcohol sulfate, and soap. Environ. Toxicol. Chem. 1999, 18, 2634-2644. (38) Pio, C. A.; Alves, C. A.; Duarte, A. C. Identification, abundance and origin of atmospheric organic particulate matter in a Portuguese rural area. Atmos. Environ. 2001, 35, 1365-1375. (39) Dachs, J.; Van Ry, D. A.; Eisenreich, S. J. Occurrence of estrogenic nonylphenols in the urban and coastal atmosphere of the lower Hudson River estuary. Environ. Sci. Technol. 1999, 33, 26762679. (40) Van Ry, D. A.; Dachs, J.; Gigliotti, C. L.; Brunciak, P. A.; Nelson, E. D.; Eisenreich, S. J. Atmospheric seasonal trends and environmental fate of alkylphenols in the lower Hudson River Estuary. Environ. Sci. Technol. 2000, 34, 2410-2417. (41) Berkner, S.; Streck, G.; Herrmann, R. Development and validation of a method for determination of trace levels of alkylphenols and bisphenol A in atmospheric samples. Chemosphere 2004, 54, 575-584.

Received for review December 4, 2004. Revised manuscript received April 1, 2005. Accepted April 11, 2005. ES048084P

VOL. 39, NO. 12, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4397