Environ. Sci. Technol. 2007, 41, 6725-6731
Deposition versus Photochemical Removal of PBDEs from Lake Superior Air JONATHAN D. RAFF AND RONALD A. HITES* School of Public and Environmental Affairs, Indiana University, Bloomington, Indiana 47405
Analysis of a sediment core collected from Siskiwit Lake, located on a remote island in Lake Superior, provides evidence that polybrominated diphenyl ethers (PBDEs) are removed effectively from the atmosphere via deposition processes during long-range transport. A mass balance model based on photochemical rate constants and data from atmospheric samples was created to understand the relative importance of various photochemical and deposition processes in removing PBDEs from the atmosphere. Photolysis rate constants were derived from UV absorption spectra of 25 PBDEs recorded in isooctane over the range of 280-350 nm at 298 K. Photolysis decays measured for BDE-3 and -7 in the gas phase were substantial compared to a well-defined chemical actinometer, indicating that their photolysis quantum yields are significant. Dibenzofuran production was observed when PBDE congeners containing ortho-bromines were photolyzed in helium. From estimates of removal rates of PBDEs from the lower troposphere, we find that wet and dry deposition account for >95% of the removal of BDE-209, while photolysis accounts for ∼90% of the removal of gas-phase congeners such as BDE-47. These results help explain the deposition patterns of PBDEs found in lake and river sediments and have important implications concerning the inclusion of photolysis as a fate process in multimedia models.
Introduction Polybrominated diphenyl ether (PBDE) flame retardants are persistent organic pollutants that have been found globally in the environment. These high production volume chemicals (∼70 000 metric tons were sold in 2003 (1)) are polymer additives that have between 2 and 10 bromine substituents per molecule. Since they are not chemically incorporated into polymers, PBDEs are easily released to the atmosphere during their production, use, and disposal. Over the past decade, a mounting body of data has shown that PBDEs are prone to undergo long-range atmospheric transport to regions where they were never used, for example, the Arctic (2-6). A full understanding of the long-range transport potential of PBDEs, however, has been limited by large uncertainties in their emission inventories, atmospheric chemistry, and deposition processes. During their dispersion through the atmosphere, PBDEs may be destructively removed by photolysis and reactions with hydroxyl (OH) radicals. Structure-activity relationships for aromatic compounds are well established (7, 8) and have been shown to reasonably predict PBDE + OH reaction rate * Corresponding author e-mail:
[email protected]. 10.1021/es070789e CCC: $37.00 Published on Web 08/29/2007
2007 American Chemical Society
constants (9). Multimedia box models have been used to assess the long-range transport potential of PBDEs by taking into consideration, among other factors, PBDE + OH reaction rate constants derived from structure-activity relationships (10, 11). However, a critical uncertainty in these models is the rate of direct photolysis for gas- and particle-phase PBDEs in the atmosphere. The photolytic degradation of PBDEs has been studied extensively in solution and mixed-phase systems with photolysis half-lives ranging from less than 1 h to more than a year, depending on the solvent or substrate (12-15). There are no experimental data available on the rate constants for the gas-phase photolysis of PBDEs, although these experiments suggest that photolysis should be an important fate for PBDEs in the atmosphere. The goals of the present study are (a) to evaluate the importance of photolysis, OH radical reactions, and deposition processes in determining the atmospheric fate of PBDEs and (b) to understand why congener profiles of PBDEs found in lake sediment are enriched in BDE-209 but depleted in congeners with 3-8 bromines, relative to their levels in air. We analyzed a sediment core from Siskiwit Lake on Isle Royale, a pristine island in Lake Superior, to provide a deposition pattern of PBDEs that was free of anthropogenic sources. Gas-phase photolysis frequencies were derived from a combination of measured and modeled data. The UV-vis spectra of 25 PBDE congeners in solution provided absorption cross-section data. The photolysis of 2,4-dibromodiphenyl ether in the gas phase was studied at a wavelength that is relevant to tropospheric chemistry to provide an estimate of the photolysis quantum yield. The wavelength dependent solar actinic flux was given by a radiative transfer model (16). Loss rates of PBDEs from the air above Lake Superior caused by photolysis are compared to lifetimes resulting from OH radical reactions and wet and dry deposition processes using a steady-state mass balance model that incorporates PBDE air concentrations measured at an air-sampling station located near Lake Superior (6).
Experimental Section Photolysis Experiments. UV-vis absorption spectra of 25 PBDE congeners containing 1-10 bromines (Table S1, Supporting Information) were collected at 298 K using a Varian Cary 100 Bio UV-vis double-beam spectrophotometer. Absorption spectra of a 50 ( 2 µg mL-1 solution of each PBDE congener were measured in a 1 cm path length quartz cell over the range of 250-400 nm. Absorption cross-sections, σ(λ), to the base e were derived from the measured absorbance, log[I0(λ)/I(λ)] from the Beer-Lambert law
( )
2.303 log
I0(λ) I(λ)
) σ(λ)[PBDE] l
(1)
where l is the path length. The linearity of log[I0(λ)/I(λ)] versus [PBDE] over a concentration range of 0-50 µg mL-1 was verified in the case of BDE-47 and assumed to apply for all other PBDE solutions measured. No significant (99%); see Table S1 (Supporting Information) for a complete list of the congeners used and their abbreviations. Ultrahigh-purity (99.999%) helium and ultrazeroambient monitoring air were purchased from Praxair, Inc. (Indianapolis, IN). Chlorine gas (5% in ultrahigh-purity helium) was from Matheson Tri-gas (Joliet, IL). All other reagents and solvents were obtained from Aldrich Chemical Co. (Milwaukee, WI) and were used without further purification. 2-Bromodibenzofuran was obtained from the SigmaAldrich Library of Rare Chemicals.
Results and Discussion Observations of PBDEs in the Atmosphere and in Sediment. Sediment cores provide a valuable record of the atmospheric deposition of semivolatile organic pollutants to remote lakes. The PBDEs found in the sediment of Siskiwit Lake are assumed to be entirely the result of atmospheric deposition because there are no significant anthropogenic sources within Siskiwit Lake’s drainage basin and the water level of this lake is higher than that of Lake Superior. Figure 1 (right) shows the PBDE congener profile found in the sediment core’s first 2.5 cm, a depth that corresponds to deposition from 1995 to 2005, based on a sedimentation rate of 0.25 cm year-1 (19). Decabromodiphenyl ether (BDE-209) at 2600 pg g-1 sediment (dry weight) is the most abundant congener, comprising 95% of the total PBDEs in the sample; tetra- through hexabrominated diphenyl ethers comprise most of the remaining fraction at 57, 69, 20, 9, and 9 pg g-1 sediment for BDE-47, -99, -100, -153, and -154, respectively. Analogous deposition profiles have been observed in sediment cores from Lake Superior (20) and from the other Great Lakes that are more heavily impacted by industrial and urban centers along their shores (18, 20, 21), suggesting that atmospheric deposition is the primary source of PBDEs to the Great Lakes. In all these cases, BDE-209 comprises over 90% of the PBDEs in the sediment. On the basis of the sedimentation rate, 6726
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 19, 2007
FIGURE 1. PBDE congener profiles (given as percent of total) in sediment from Siskiwit Lake and in air samples collected at Eagle Harbor, MI (6), near the shore of Lake Superior. concentrations found in the sediment core, and the dimension of the core, we derive sedimentation fluxes of 7, 10, 3, 2, 1, and 470 ng m-2 year-1 for BDE-47, -99, -100, -153, -154, and -209, respectively, to Siskiwit Lake. These fluxes are comparable to those derived from air samples (6) but lower (by a factor of ∼30 for the less-brominated PBDEs and by a factor of ∼3 for BDE-209) than surface fluxes obtained from a Lake Superior sediment core (20). Figure 1 also includes the average congener profiles of PBDEs found in the air (vapor and particles) sampled at Eagle Harbor, MI, on the southern shore of Lake Superior between 2003 and 2004 (6). Unlike the congener profile from Siskiwit Lake sediment, the air-sample profiles are enriched in the tetra- and pentabrominated diphenyl ethers. This discrepancy in congener abundances may arise from processes such as photolysis, reactions with atmospheric oxidants (e.g., OH), and wet and dry deposition. The relative importance of each of these processes is the subject of the following discussion. Estimation of PBDE Photolysis Lifetimes. The importance of photolysis in the removal of PBDEs from the atmosphere was evaluated by comparison of the photolysis frequencies of PBDE congeners to the first-order removalrate constants for oxidant reactions and dry and wet deposition. The photolysis frequencies (J) of PBDEs containing 1-10 bromines were calculated using the expression
J)
∫
350
280
σ(λ)Φ(λ)F(λ)dλ
(2)
where σ(λ) is the wavelength dependent absorption crosssection, Φ(λ) is the wavelength-dependent photolysis quantum yield of a given PBDE, and F(λ) is the solar actinic flux. Measurements of absorption cross-sections of PBDEs in the gas phase are limited by their relatively low vapor pressures (from 10-2 to 270 nm (23) and did not decay in our system (see Figure 2). Thus, the observed decays of PBDEs are likely not influenced by reactive species that may be photochemically generated on the walls of the reactor. Noticeable increases in signal caused by dibenzofuran (m/z 168) and 2-bromodibenzofuran (m/z 246) were observed during the photolysis of BDE-3 and 7, respectively. These observations, which are discussed in the Supporting Information, suggest that dibenzofuran formation occurs during gas-phase photolysis of PBDEs with ortho-bromines. The quantum yield for photolysis of BDE-7 (ΦBDE) at 307 nm and 323 K was determined relative to Cl2 (at 298 K) using
ΦBDE )
kBDE ΦCl2 σCl2 kCl2 σBDE
(3)
where kBDE and kCl2 are the pseudo-first-order photolysis decay rates of BDE-7 and Cl2, ΦCl2 is the photolysis quantum yield of chlorine that is assumed to be unity and independent of wavelength (31), and σCl2 ) 1.66 × 10-19 cm2 molecule-1 at 307 nm (24). Linear least-squares fits to the data in Figure 2 yield kBDE and kCl2 values of 0.056 ( 0.004 and 0.095 ( 0.005 min-1 for BDE-7 and Cl2, respectively, where the errors represent 2σ of the mean from repeated experiments. Using the absorption cross-section (σBDE) measured at 307 nm in the present work, we derived a quantum yield of 0.8 for BDE7. The error associated with this calculation is large because the absorption cross-section of BDE-7 at 307 nm is in the tail region of the lowest-energy absorption band, close to the detection limit of the UV-vis spectrometer. However, the rapid decay of BDE-7 relative to Cl2 shows that the quantum yield of BDE-7 is significant and perhaps even close to unity. This is higher than what has been measured in solution phase
FIGURE 3. Atmospheric lifetimes of PBDEs as a function of bromine substitution with respect to OH radical reactions (τOH) and photolysis (τphoto). The values of τOH were estimated from a structure-activity relationship (7, 8) [log τOH ) 0.2333(#Br) + 1.472; R 2 ) 0.941, N ) 48]. The τphoto values for 25 PBDE congeners (gray circles) were derived as discussed in the text and fitted with a straight line [log τphoto ) -0.4031(#Br) + 2.527; R 2 ) 0.838, N ) 25]. (13), which is expected because of the higher probability for collisional deactivation of photoinduced excited states in solution. On the basis of this measurement, we will assume that a wavelength-independent quantum yield of 0.5 applies for all PBDEs. The photolysis frequencies (noon values) were calculated using the tropospheric UV-vis radiative transfer model (TUV 4.4) (16) at 48° latitude for the summer and winter solstices and the fall and spring equinoxes. The results we report here are averaged over these four seasons and over an assumed mixing height of 2.5 km. The assumptions used in the model were as follows: cloudless skies, a surface albedo of 0.1, and an ozone column of 300 Dobson units. Calculations were made using the PBDE absorption spectra measured in this work, assuming a quantum yield of 0.5, based on the BDE-7 photolysis experiments described above. Figure 3 shows the PBDE photolysis lifetimes (τphoto ) J-1) as a function of the number of bromine substituents; these data are included in Table S3. The lifetimes range from >20 h for PBDE congeners with 1-2 bromines to 200 14 ∞ 0.89 0.84
6.8 220 1.2 8.4 11 250
1.2 170 1.3 3.0 13 190
0 0.5 8.3 0 130 140
a Photolysis and OH reaction rate constants are calculated with eqs 5 and 6 and account for particle partitioning; dry, vapor, and wet deposition rate constants are calculated using eqs 7-9. b τ ) k-1; total lifetimes were calculated from eq 10. c Flows were calculated using eq 11.
NO3 and O3 are expected to be slow, on the basis of reaction rate constants measured for structurally similar compounds (e.g., dibenzofuran: kO3 < 8 × 10-20 cm3 molecule-1 s-1; kNO3 < 1.6 × 10-15 cm3 molecule-1 s-1 (26)). Photo-oxidation in the Particle Phase. PBDE congeners associated with the particle phase do not follow gas-phase reactivity patterns because of additional physical and chemical considerations that are unique to the particle phase. The reactive lifetimes of PBDEs buried in multicomponent particles are likely longer than their lifetimes in the gas phase. For such complex systems, the OH radical must diffuse into the particle, where it will more likely be scavenged by organic components that are more abundant and reactive than the PBDEs. There are several other factors that may cause the photolysis rate constants of PBDEs to be lower in the particle phase than in the gas phase. Carbonaceous aerosols from urban air masses, the type PBDEs are most associated with, have high absorption cross-sections in the actinic region (27, 28) and may effectively shield a buried PBDE molecule from solar radiation. If a sorbed PBDE molecule happens to absorb light, its electronic excited state may be quenched by neighboring molecules. If photodissociation does occur, the products may be prevented from diffusing away from each other and may recombine to form the parent PBDE within the particle (the “cage effect”). These processes may explain experimental evidence showing that polycyclic aromatic hydrocarbons and BDE-209 associated with organic carbonrich particles (e.g., clay, natural sediment, fly ash, and carbon black) are less susceptible to photodegradation than when they are absorbed to mineral substrates (29, 30). From these considerations we proceeded on the assumption that particle-bound PBDEs are not susceptible to photolysis or reactions with gas-phase atmospheric oxidants. Hence, the expressions for the photochemical removal rate constants with respect to photolysis (k′photo) and OH radical reactions (k′OH) that include particle-partitioning may be written
k′photo ) (1 - f ) J
(5)
k′OH ) (1 - f )kOH[OH]
(6)
where f is the fractional occurrence of a PBDE in the particlephase (f < 1), as determined from vapor- and particle-phase air samples collected with high-volume air samplers at five different sites in the east-central U.S (4, 6) (see Table S4 and Figure S4). For example, the fraction of PBDEs in the particle phase at 288 ( 1 K is 0.17 ( 0.09 for BDE-47 and 0.42 ( 0.12 for BDE-99. Since BDE-209 is not detected in the vapor phase, it is assumed that its fractional occurrence is 0.99999 for the purpose of our calculations. The photochemical rate constants k′photo and k′OH for three of the most environmentally prevalent congeners, BDE-47, -99, and -209, are shown in Table 1. The photolysis lifetime for BDE-47 and -99 may be as short as 0.4 and 0.3 day, while PBDE congeners associated 6728
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 19, 2007
mostly with particles are more resistant to photochemical removal. Under these conditions, BDE-209 is essentially unreactive toward gas-phase OH radicals, and its lifetime is >200 days with respect to photolysis. Mass Balance of PBDEs in Lake Superior Air. A more complete analysis of the atmospheric fate of PBDEs accounts for nondestructive removal processes such as wet and dry deposition. We compared these atmospheric removal processes in the context of a box model of PBDEs in the atmosphere above Lake Superior, focusing on the congeners BDE-47, -99, and -209. The rate constant for removal of PBDEs from the atmosphere resulting from particle dry deposition, kdry, is given by
kdry )
νp f Z
(7)
where νp is the deposition velocity for the particle-bound PBDEs and Z is the assumed tropospheric mixing height (2.5 km). Particle-bound deposition velocities of PBDEs to industrialized regions have been reported to be 0.4-49 (average of 6.8) cm s-1 for a site near a municipal solid waste incineration and electronics recycling plant in Sweden (31) and 2.8-7.8 (average of 4.5) cm s-1 at a site near Izmir, Turkey (32). Recently, Su et al. reported particle-bound deposition velocities to a forest canopy located in Southern Ontario, Canada, in the range of 0.1-2.5 (average of 0.8) cm s-1 (33). Deposition velocities are strongly influenced by particle size and will be high at urban sites where gravitational settling of course particles is important on a local scale. Forests are efficient scavengers of gas- and particle-phase semivolatile organic pollutants (33). Therefore, deposition velocities to vegetation are expected to be high in forested regions relative to open-water sites. There are no reports of dry deposition velocities to the Great Lakes in the literature. Because Lake Superior is relatively remote and is open water, particle-bound deposition velocities of PBDEs are expected to be lower than those derived in refs 31-33. Thus, for the purpose of the estimation of kdry, νp was assumed to be 0.2 cm s-1, based on particulate dry deposition velocities of polychlorinated biphenyls (PCBs) and polycyclic aromatic hydrocarbons to the Great Lakes (34, 35). Dry deposition rate constants calculated for BDE47, -99, and -209 correspond to lifetimes of 83, 34, and 14 days, respectively (see Table 1). Gaseous dry deposition will be an important consideration for the more volatile congeners. Indeed, analysis of water samples collected recently from Lake Michigan showed that dissolved-phase BDE-47 and -99 comprised up to 90% of the total water concentration (dissolved + particle) (36). The rate constant for removal of vapor-phase BDE-47 and -99 from the atmosphere resulting from gaseous dry deposition, kvap, was estimated using the two-film model and is given by
kvap )
KL{(1 - f )[PBDE]air - [PBDE]waterH′} (1 - f )[PBDE]airZ
(8)
where KL is the total liquid-phase mass transfer coefficient for a wind speed of 3 m s-1 (0.32 and 0.29 cm s-1 for BDE-47 and -99, respectively) (37), and H′ is the dimensionless Henry’s law constant at 15 °C (2.3 × 10-4 and 7.1 × 10-5 for BDE-47 and -99, respectively) (37). Dissolved-phase concentrations of BDE-47 and -99 in Lake Superior were estimated to be 1.0 ( 0.8 and 1.7 ( 1.5 pg L-1 (36). The two-film model predicts a net deposition of BDE-47 and -99 to Lake Superior with vapor deposition rate constants that correspond to lifetimes of 12 and 15 days, respectively (see Table 1). The first-order rate constant for wet removal of PBDEs from the atmosphere kwet is given by
kwet )
p˘ W Z
(9)
where p˘ is the annual precipitation rate for Lake Superior (0.775 m yr-1) and W is the washout ratio of individual PBDE congeners. The washout ratios used to calculate kwet are 1.3 × 105, 3.4 × 105, and 12.8 × 105 for BDE-47, -99, and -209, respectively, as calculated from air and precipitation samples taken from a remote region of Sweden (3). Estimates of wet deposition rate constants for BDE-47, -99, and -209 range from 1.3 to 13 × 10-6 s-1 (see Table 1). These correspond to wet removal lifetimes of 9, 3, and 1 days for BDE-47, -99, and -209, respectively. The overall lifetime (τtot) of PBDEs in the atmosphere is estimated from the reciprocal of the sum of all the rate constants
τtot ) (k′OH + k′photo + kdry + kvap + kwet)-1
(10)
Overall lifetimes for BDE-47, -99, and -209 are 0.4, 0.2, and 0.8 days, respectively. According to the calculations indicated in Table 1, the lifetime of BDE-209 is controlled by deposition processes; in this case, wet deposition is the most important removal pathway for particle-bound congeners from the atmosphere. Gas-phase congeners that absorb significant portions of the Sun’s light, such as BDE-47 and -99, are primarily removed by photolysis. A mass balance model was used to estimate the loss rates of PBDEs from the atmosphere above Lake Superior. In this approach, a steady state is assumed, and the flow of each compound from the atmosphere resulting from the various loss processes (Floss) is calculated from
Floss ) kloss[PBDE]airAZ
(11)
where kloss is any of the rate constants, k′OH, k′photo, kdry, kvap, or kwet derived in the discussion above, and A is the surface area of Lake Superior (82 100 km2). Average PBDE air concentrations, [PBDE]air, for combined vapor and particle phase are from air samples collected from the shores of Lake Superior at Eagle Harbor, MI, from 2003-2004 (6); they are 1.3 ( 0.4, 0.6 ( 0.1, and 1.6 ( 0.8 pg m-3 for BDE-47, -99, -209, respectively. The estimated loss rates for these congeners are given in Table 1 and are depicted in the mass balance diagram in Figure 4. The combined deposition flows (Fdry + Fvap + Fwet) to the surface of the lake (Table 1 and Figure 4) are comparable to the flows to the sediment (Fsed) as given by Song et al., who indicate that these deposition rates are ∼4, ∼3, and 100 kg year-1 for BDE-47, -99, and -209, respectively (20). This agreement with our estimates justifies our choice of parameters used to derive the deposition rates in this work. Particle-water partitioning of the less-brominated congeners
FIGURE 4. Mass balance of PBDEs in the atmosphere above Lake Superior. Chemical removal rates: photolysis (Fphoto) and OH reactions (FOH). Depositional loss rates: dry, vapor, and wet deposition (Fdry, Fvap, and Fwet). Flows are listed in order of BDE-47, -99, and -209 next to each arrow; units are in kg year-1. Fem is the estimated atmospheric emission rate of PBDE from sources (see text). Fsed is the sedimentation loss rate as reported by Song et al. (20). could also occur while particles settle out of the water column. This would further reduce the fraction of BDE-47 and -99 deposited to the sediments relative to BDE-209. According to the model results, photolysis results in the highest loss (∼90% of the total losses) of BDE-47 and -99 from the atmosphere above Lake Superior. This is not the case for BDE-209, where >90% of the removal is the result of precipitation events. These calculations show that wet deposition is likely the most important source of BDE-209 to Lake Superior. This explains why the PBDE congener profile found in sediments from Siskiwit Lake and the Great Lakes appear to be enriched in BDE-209 compared to other congeners. These results are consistent with congener profiles obtained from air-particle samples collected in the Arctic that showed significant depletion of BDE-209 relative to tetraand pentabrominated PBDE congeners (5). Modeling work by Wania and Dugani (10) and Breivik et al. (11) also provided evidence that the long-range transport potential of BDE-209 is limited by deposition, rather than by reactions with atmospheric oxidants. Highly brominated PBDE congeners are most efficiently removed from the atmosphere by wet deposition, similar to what is observed for low-volatility polycyclic aromatic hydrocarbons (PAHs) and polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) (35, 38, 39). The sum of all loss processes listed in Table 1 can be used to estimate the BDE-47, -99, and -209 emission rates (Fem) to the Lake Superior airshed
Fem ) Fphoto + FOH + Fdry + Fvap + Fwet
(12)
According to this model, BDE-209 emissions amount to only one-third of the combined emissions of BDE-47 and -99, even though recent market demand for BDE-209 was ∼3 times greater than for the penta product (1). This difference may stem from the higher vapor pressures of BDE-47 and -99 and their greater propensity to be released during production and use. Differences between the relative emisVOL. 41, NO. 19, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6729
sions of BDE-47, -99, and -209 and the congener profiles for the air samples (Figure 1) could indicate that the losses of BDE-47 and -99 from the atmosphere are overestimated in our model. This would likely be caused by overestimated photolysis rates because they are derived from noon actinic flux estimates that represent the maximum light intensity that would reach the tropospheric boundary layer over Lake Superior during the course of a year. Estimates of PBDE emissions are scarce, and attempts to calculate emission inventories are limited to Europe and Japan (40, 41). According to these studies, between 22-31 t year-1 of BDE-47 were released in Europe and between 1-7 ton year-1 of BDE-209 were released in Japan. Scaling these estimates to the human population around Lake Superior suggests releases of ∼27 and ∼11 kg year-1 of BDE-47 and BDE-209, respectively. However, PBDE levels in the U.S. tend to be ∼20 times higher than elsewhere (1). Therefore, the emissions to Lake Superior’s atmosphere may be more on the order of 30-500 kg year-1 for BDE-47 and 10-200 kg year-1 for BDE-209. The emission rates derived from the mass balance shown in Figure 4 and listed in Table 1 fall within these estimate ranges. This work shows that the long-range transport of PBDEs from urban regions will be dependent on the frequency of precipitation events and the intensity of sunlight, factors that may vary significantly and unpredictably on the time scale of hours or less. Thus, episodic transport events during the night and at times when precipitation events are infrequent may play an important role in the transportation of PBDEs to remote areas. Predictions of atmospheric fluxes of PBDEs are limited by uncertainties in all of the parameters used for these calculations. Virtually none of these uncertainties and none of the statistical distribution functions are known well enough to carry out a quantitative error analysis. Our estimate is that the fluxes given in Table 1 and Figure 4 have errors of about a factor of 2 (see the Supporting Information). Clearly, additional data on vapor deposition (and evaporation) for the more volatile PBDE congeners are needed. This will require additional measurements of aqueous and particle concentrations of PBDEs in Great Lakes water, concurrently with air concentrations. However, inclusion of this information into more elaborate models is not likely to change the main conclusion of this work: Deposition processes control the loss of BDE-209 from the atmosphere and are responsible for the enhancement of BDE-209 found in sediment samples from the Great Lakes and other lakes and rivers around the world.
Acknowledgments We are grateful to D. C. G. Muir and C. Marvin (Environment Canada) and G. Slater (McMaster University) for providing the sediment core and to Eunha Ho and the Integrated Atmospheric Deposition Network (IADN) for providing the air-sampling data. This research has been funded by the United States Environmental Protection Agency (EPA), under the Science to Achieve Results (STAR) Graduate Fellowship Program (FP-91663501). The EPA has not officially endorsed these results and the views expressed herein may not reflect those of the EPA. We also thank C. Parmenter (Indiana University) and J. Orlando (NCAR) for helpful discussions.
Supporting Information Available Table of PBDE nomenclature used, tabulated and plotted UV-vis spectra for 25 PBDE congeners, comparison of UV absorption spectra in isooctane and perfluorinated solvents, additional details of photochemistry experiments including a discussion of dibenzofuran formation, table and figure of the fraction of PBDEs in the particle phase, a list of PBDE photochemical rate constants and lifetimes, and a discussion 6730
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 19, 2007
of uncertainty in the lifetime estimates. This material is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Hites, R. A. Polybrominated diphenyl ethers in the environment and in people: A meta-analysis of concentrations. Environ. Sci. Technol. 2004, 38, 945-956. (2) de Wit, C. A.; Alaee, M.; Muir, D. C. G. Levels and trends of brominated flame retardants in the Arctic. Chemosphere 2006, 64, 209-233. (3) ter Schure, A. F. H.; Larsson, P.; Agrell, C.; Boon, J. P. Atmospheric transport of polybrominated diphenyl ethers and polychlorinated biphenyls to the Baltic Sea. Environ. Sci. Technol. 2004, 38, 1282-1287. (4) Hoh, E.; Hites, R. A. Brominated flame retardants in the atmosphere of the East-Central United States. Environ. Sci. Technol. 2005, 39, 7794-7802. (5) Wang, X.-M.; Ding, X.; Mai, B.-X.; Xie, Z.-Q.; Xiang, C.-H.; Sun, L.-G.; Sheng, G.-Y.; Fu, J.-M.; Zeng, E. Y. Polybrominated diphenyl ethers in airborne particles collected during a research expedition from the Bohai Sea to the Arctic. Environ. Sci. Technol. 2005, 39, 7803-7809. (6) Hoh, E. Investigation of organochlorine and organobromine contaminants in the atmosphere. Ph.D. Thesis, Indiana University, Bloomington, IN, 2006, pp. 245. (7) Zetzsch, C. In Predicting the Rate of OH-Addition to Aromatics using σ+-Electrophilic Substituent Constants for Mono- and Polysubstituted Benzene; Proceedings of the 15th Informal Conference on Photochemistry; Slanger, T., Golden, D. M., Eds.; Stanford, CA, 1982; pp. 29-32. (8) Kwok, E. S. C.; Atkinson, R. Estimation of hydroxyl radical reaction rate constants for gas-phase organic compounds using a structure-reactivity relationship: An update. Atmos. Environ. 1995, 29, 1685-1695. (9) Raff, J. D.; Hites R. A. Gas-phase reactions of brominated diphenyl ethers with OH radicals. J. Phys. Chem. A 2006, 110, 10783-10792. (10) Wania, F.; Dugani, C. B. Assessing the long-range transport potential of polybrominated diphenyl ethers: A comparison of four multimedia models. Environ. Toxicol. Chem. 2003, 22, 1252-1261. (11) Breivik, K.; Wania, F.; Muir, D. C. G.; Alaee, M.; Backus, S.; Pacepavicius, G. Empirical and modeling evidence of the longrange atmospheric transport of decabromodiphenyl ether. Environ. Sci. Technol. 2006, 40, 4612-4618. (12) Barcellos da Rosa, M.; Kru ¨ ger, H.-U.; Thomas, S.; Zetzsch, C. Photolytic debromination and degradation of decabromodiphenyl ether, an exploratory kinetic study in toluene. Fresenius Environ. Bull. 2003, 12, 940-945. (13) Eriksson, J.; Green, N.; Marsh, G.; Bergman, A° . Photochemical decomposition of 15 polybrominated diphenyl ether congeners in methanol/water. Environ. Sci. Technol. 2004, 38, 31193125. (14) Bezares-Cruz, J.; Jafvert, C. T.; Hua, I. Solar photodecomposition of decabromodiphenyl ether: Products and quantum yield. Environ. Sci. Technol. 2004, 38, 4149-4156. (15) Ahn, M.-Y.; Filley, T. R.; Jafvert, C. T.; Nies, L.; Hua, I.; BezaresCruz, J. Photodegradation of decabromodiphenyl ether adsorbed onto clay minerals, metal oxides, and sediment. Environ. Sci. Technol. 2006, 40, 215-220. (16) Madronich, S.; Flocke, S. In Handbook of Environmental Chemistry; Boule, P., Ed.; Springer: Heidelberg, Germany, 1998; pp 1-26. (17) Funk, D. J.; Oldenborg, R. C.; Dayton, D.-P.; Lacosse, J. P.; Draves, J. A.; Logan, T. J. Gas-phase absorption and laser-induced fluorescence measurements of representative polychlorinated dibenzo-p-dioxins, polychlorinated dibenzofurans, and a polycyclic aromatic hydrocarbon. Appl. Spectrosc. 1995, 49, 105114. (18) Zhu, L. Y.; Hites, R. A. Brominated flame retardants in sediment cores from Lakes Michigan and Erie. Environ. Sci. Technol. 2005, 39, 3488-3494. (19) Hites, R. A. Environmental behavior of chlorinated dioxins and furans. Acc. Chem. Res. 1990, 23, 194-201. (20) Song, W.; Ford, J. C.; Li, A.; Mills, W. J.; Buckley, D. R.; Rockne, K. Polybrominated diphenyl ethers in the sediments of the Great Lakes. 1. Lake Superior. Environ. Sci. Technol. 2004, 38, 32863293. (21) Li, A.; Rockne, K. J.; Sturchio, N.; Song, W.; Ford, J. C.; Buckley, D. R.; Mills, W. J. Polybrominated diphenyl ethers in the
(22)
(23) (24)
(25)
(26) (27)
(28) (29) (30)
(31)
(32)
sediments of the Great Lakes. 4. Influencing factors, trends, and implications. Environ. Sci. Technol. 2006, 40, 7528-7534. Scharping, H.; Zetzsch, C.; Dessouki, H. A. The UV absorption spectra of the trichlorobenzenes and the higher chlorinated benzenes in the gas phase and in n-hexane solution. J. Mol. Spectrosc. 1987, 123, 382-391. Berlman, I. B. Handbook of Fluorescence Spectra of Aromatic Molecules, 2nd ed.; Academic Press, Inc.: New York, 1971; p 473. Atkinson, R.; Baulch, D. L.; Cox, R. A.; Crowley, J. N.; Hampson, R. F.; Hynes, R. G.; Jenkin, M. E.; Rossi, M. J.; Troe, J. Evaluated kinetic and photochemical data for atmospheric chemistry: Volume III - gas phase reactions of inorganic halogens. Atmos. Chem. Phys., 2007, 7, 981-1191. Prinn, R. G.; Weiss, R. F.; Miller, B. R.; Huang, J.; Alyea, F. N.; Cunnold, d. M.; Fraser, P. J.; Hartley, H. E.; Simmonds, P. G. Atmospheric trends and lifetime of CH3CCl3 and global OH concentrations. Science 1995, 269, 187-192. Kwok, E. S. C.; Arey, J.; Atkinson, R. Gas-phase chemistry of dibenzo-p-dioxin and dibenzofuran. Environ. Sci. Technol. 1994, 28, 528-533. Liousse, C.; Cachier, H.; Jennings, S. G. Optical and thermal measurements of black carbon aerosol content in different environments: Variation of the specific attenuation crosssection, sigma (σ). Atmos. Environ. 1993, 27A, 1203-1211. Mikhailov, E. F.; Vlaskenko, S. S.; Podgorny, I. A.; Ramanathan, V.; Corrigan, C. E. Optical-properties of soot-water agglomerates: An experimental study. J. Geophys. Res. 2006, 111, DO7729. Behymer, T. D.; Hites R. A. Photolysis of polycyclic aromatic hydrocarbons adsorbed on fly ash. Environ. Sci. Technol. 1988, 22, 1311-1319. Dunstan, T. D. J.; Mauldin, R. F.; Jinxin, Z.; Hipps, A. D.; Wehry, E. L.; Mamantov, G. Adsorption and photodegradation of pyrene on magnetic, carbonaceous, and mineral subfractions of coal stack ash. Environ. Sci. Technol. 1989, 23, 303-308. ter Schure, A. F. H.; Agrell, C.; Bokenstrand, A.; Sveder, J.; Larsson, P.; Zegers, B. N. Polybrominated diphenyl ethers at a solid waste incineration plant II: Atmospheric deposition. Atmos. Environ. 2004, 38, 5149-5155. Cetin, B.; Odabasi, M. Air-water exchange and dry deposition of polybrominated diphenyl ethers at a coastal site in Izmir Bay, Turkey. Environ. Sci. Technol. 2007, 41, 785-791.
(33) Su, Y.; Wania, F.; Harner, T.; Lei, Y. D. Deposition of polybrominated diphenyl ethers, polychlorinated biphenyls, and polycyclic aromatic hydrocarbons to a boreal deciduous forest. Environ. Sci. Technol. 2007, 41, 534-540. (34) Swackhamer, D. L.; McVeety, B. D.; Hites, R. A. Deposition and evaporation of PCB congeners to and from Siskiwit Lake, Isle Royale, Lake Superior. Environ. Sci. Technol. 1988, 22, 664672. (35) Hoff, R. M.; Strachan, W. M. J.; Sweet, C. W.; Chan, C. H.; Shackleton, M.; Bidleman, T. F.; Brice, K. A.; Burniston, D. A.; Cussion, S.; Gatz, D. F.; Harlin, K.; Schroeder, W. H. Atmospheric deposition of toxic chemicals to the Great Lakes: A review of data through 1994. Atmos. Environ. 1996, 30, 35053527. (36) Streets, S. S.; Henderson, S. A.; Stoner, A. D.; Carlson, D. L.; Simcik, M. F.; Swackhamer, D. L. Partitioning and bioaccumulation of PBDEs and PCBs in Lake Michigan. Environ. Sci. Technol. 2006, 40, 7263-7269. (37) Cetin, B.; Odabasi, M. Measurement of Henry’s law constants of seven polybrominated diphenyl ether (PBDE) congeners as a function of temperature. Atmos. Environ. 2005, 39, 52735280. (38) Bidleman, T. F. Atmospheric processes: Wet and dry deposition of organic compounds are controlled by their vapor-particle partitioning. Environ. Sci. Technol. 1988, 22, 361-367. (39) Baker, J.; Hites R. A. Is combustion the major source of polychlorinated dibenzo-p-dioxins and dibenzofurans to the environment? A mass balance investigation. Environ. Sci. Technol. 2000, 34, 2879-2886. (40) Prevedouros, K.; Jones, K. C.; Sweetman, A. J. Estimation of the production, consumption, and atmospheric emissions of pentabrominated diphenyl ether in Europe between 1970 and 2000. Environ. Sci. Technol. 2004, 38, 3224-3231. (41) Yasuhiro, H.; Sakai, S. Atmospheric emissions of BDE-209 in Japan. Organohalogen Compds. 2004, 66, 3712-3717.
Received for review April 3, 2007. Revised manuscript received July 3, 2007. Accepted July 9, 2007. ES070789E
VOL. 41, NO. 19, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6731
