Polycyclic Aromatic Hydrocarbons inthe Air in the St. Lawrence Basin(Quebec)B A R N A B EÄ N G A B EÄ A N DL A U R I E R P O I S S A N T *

Meteorological Service of Canada, Atmospheric Toxic ProcessesSection, Environment Canada, 105 McGill Street,Montreal, Quebec, Canada H2Y 2E7

High-volume air samples were collected from 1993 to1996 in rural areas of Quebec to investigate on the levelsand the vapor-particle partitioning of polycyclic aromatichydrocarbons (PAHs). Ranges for the mean concentrationsof total PAHs (ng m-3) were as follows: Villeroy, 3.31-18.92; St. Anicet, 7.57-22.84; and Mingan, 0.50-0.53. Lowermolecular weight PAHs predominated at all locations.Particle-gas partition coefficients (Kp) were in goodcorrelation with their vapor pressures (r 2 ) 0.79-0.97)with slopes deviating from the expected value of -1. Thecurve of the fractions of PAHs on particles in St. Anicetin 1995 fell on that of Lake Superior. In St. Anicet, fractionsof phenanthrene, fluoranthene, pyrene, and chrysene onparticles were close to those calculated from the soot-air partition coefficient (KSA). At all sites the mean ratios ofparticulate PAH of the same molecular weight but ofvery different reactivities were similar during the samesampling days, suggesting that particle-bearing PAHs inVilleroy and Mingan were of the same nature as those foundin St. Anicet where adsorption onto soot particles wasthe major mechanism. Furthermore, the enthalpies ofdesorption for the predominant PAHs were close at allsites.

IntroductionPolycyclic aromatic hydrocarbons (PAHs) are at the presenttime one of the most documented class of organic pollutantsin the environment. Yet because of their toxicity, carcino-genicity, and mutagenicity, PAHs continue to attract thecuriosity of researchers (1, 2). There are reports of toxic effectson organisms due to PAH exposure (3-5). PAHs oxidize toform nitro- and hydroxinitro-PAHs which may enhance theirenvironmental effects (6). PAHs are ubiquitous in theenvironment. In some reports, concentrations have surpassedthe guidelines set for the protection of the environment(7, 8).

PAHs are present in crude and refined oil. In Canada, themain sources of PAHs in the air include aluminum plants,open fires, home wood burning, diesel engine emissions,and other combustion sources. The St. Lawrence Basin ishome to aluminum plants using carbon paste electrodes (9),which when heated emit PAHs (10, 11). Also, wood isfrequently used in the St. Lawrence Basin for home heating(9).

Once released into the atmosphere, PAHs partitionbetween the gas and the particulate phases. Recently, it has

been shown that the distribution of PAHs between the particleand the gas phases can also be described by the soot-airpartition coefficient (12). Despite their fast photodegradationrates, PAHs present in particles can be transported longdistances from their emission sources. For example, PAHssorbed to fly ash particles did not photodegrade (13). Theabsence of benzo[a]pyrene (BaP) degradation during long-range transport was also observed in the winter in the HighArctic where the ratio of BaP (a photolytically unstable PAH)to benzo[e]pyrene (BeP) (a more stable isomer) was ap-proximately equal to 1 (14, 15).

Within the framework of the action plan St. LawrenceVision 2000 (SLV 2000; the scopes of which were to maintain,protect, and restore the St. Lawrence ecosystem), a programwas established to determine more accurately the levels ofpollutants in the air in the St. Lawrence Basin. The samplingsites coverered downwind to upwind of the main wind patternin the St. Lawrence Basin. The objectives of this work wereto determine levels and study the particle-gas partitioningof PAHs in the air in the St. Lawrence Basin.

Materials, Locations, and MethodsAir samples (280-400 m3) were collected from 1993 to 1996at Villeroy, St. Anicet, and Mingan. Villeroy is a rural areasituated on the south side of the St. Lawrence River betweenTrois-Rivieres and Quebec City at 46°26′ N latitude and 71°56′W longitude. Farms and woodlands dominate land use inthe region. St. Anicet is located in a rural area on the southborder of the St. Francis Lake between Cornwall and Montrealat 45°07′ N latitude and 74°17′ W longitude. A small town of1000 inhabitants is located 3 km from the site. Mingan islocated in a forested region at 50°16′ N latitude and 64°14′W longitude between Sept-Iles and Havre Saint-Pierre onthe north border of the St. Lawrence River.

Air was pulled at approximately 0.3 m3 min-1 for 24 hevery 6 days through a glass fiber filter followed by a 6 cmdiameter × 8 cm length polyurethane foam plug (PUF)(density 0.0022 g cm-3) (weight ) 4.98 g) using a high-volumesampler (model PS-1 Thermal Anderson). The glass fiber filterwas used to collect particles, and the PUF plug was used tocollect the gas-phase PAHs. Only one PUF plug was used.The breakthrough was corrected using a relation developpedby (16)

where Vg (m3 g-1) is the compound-related volume that canbe sampled at 293 K with little breakthrough using a porousPUF and PL

s is the subcooled vapor pressure of the pollutant.In consideration of the chromatographic theory and assumingthat the enthalpy of desorption (∆H) remains approximatelyconstant over the temperature range of interest, Pankow (16)derived the following equation:

where Vg,T2 is the predicted sampling volume at a giventemperature T2. The resulting concentrations have beencorrected assuming that the measured sorbed mass can bemultiplied by the ratio of the sampling volume V to Vg,T2 toobtain the actual mass that was in the gas phase (17). Thetotal suspended particles (TSP) for each sample weredetermined by weighing the filters before and after sampling.

Prior to extraction, 100 pg mL-1 solutions of acenaph-thene-d8, fluoranthene-d10, and benzo[a]pyrene-d12 were

* Corresponding author fax: (514)283-8869; e-mail:laurier.poissant@ec.gc.ca.

log (Vg, 293) ) -1.059 log PL,293s -1.764 (1)

log (Vg,T2/T2)/(Vg, 293) ) (∆H/2.303R)(293 -

T2/293T2) (2)

Environ. Sci. Technol. 2003, 37, 2094-2099

2094 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 10, 2003 10.1021/es0210345 CCC: $25.00 Published 2003 by the Am. Chem. Soc.Published on Web 04/18/2003

added to the samples as surrogates. PUF plugs were Soxhletextracted with distilled dichloromethane for 24 h. Hexane(3-4 mL) was added to the extract, which was furtherconcentrated to about 2 mL using a rotary evaporator. Theresulting organic fraction was transferred to a centrifuge tube.The same procedure was applied to the filter.

The extract was separated on a silver-loaded column (18).The silver support is made of an alumina silica gel modifiedwith silver (18). Two fractions were obtained: fraction 1contained organochlorine pesticides and PCBs and fraction2 contained PAHs. The fractions were concentrated to about0.5 mL with a stream of nitrogen. Each fraction wastransferred to a 1-mL vial. Triphenylamine [(C6H5)3N] (100pg) was added as an internal standard for analytical control.PAHs were determined by capillary gas chromatography-mass spectrometry (GC-MS) in selected ion monitoring(SIM) mode using a MD-800 (VG-Fisons, Manchester, U.K.)mass selective detector under the following conditions:capillary 30 m × 0.25 mm i.d. fused silica, DB-5, 0.25 µm filmthickness. Samples were injected splitless (split time ) 55 s).The temperature program was as follows: inject at 100 °C,hold 1 min, program to 270 °C at 8 °C min-1, then to 300 °C,hold 7 min. Carrier gas: He 1.8 mL min-1. Detectortemperature: 300 °C. Ions used for SIM of PAHs wereacenaphthene (ACE), 162; ACE-d8 and fluorene (Fle), 83 and166; phenanthrene (PH) and anthracene (AN), 89, 178, and188; fluoranthene (FLA) and pyrene (PY), 101 and 202;FLA-d10, 212; benz[a]anthracene (BaA) and chrysene (CHRY),114 and 229; benzo[b]fluoranthene (BbF) and benzo[k]-fluoranthene (BkF), 126 and 253; benzo[e]pyrene (BeP) andbenzo[a]pyrene (BaP), 126 and 253; BaP-d12, 264; indeno-[1,2,3-cd]pyrene (IcdP) and benzo[ghi]perylene (BghiP), 138and 277; and dibenz[a,h]anthracene (DBA), 138 and 278. PAHconcentrations were corrected using the surrogate recoveries.Percentage surrogate recoveries were (%) as follows: acenaph-thene-d8, 55 ( 25; fluoranthene-d10, 95 ( 31; and benzo[a]-pyrene-d12, 97 ( 32. The blanks from the laboratory and thefields were also analyzed. The levels of PAH found in blankswere in all cases smaller than 5% of the PAH concentrationsin samples.

Results and DiscussionConcentrations. The number of samples, TSP, and medianand mean of total PAHs (gas + particle phases) are displayedin Table 1. The mean ∑PAH in Villeroy in 1993 equaled 9.37ng m-3 and was 2.83 times higher than in 1994 and aboutequal to that of 1995. This was half the mean total PAHconcentrations in 1996. In St. Anicet, the mean total PAH in1996 was 3 times greater than that recorded during 1994 and1.65 times higher than that of 1995. In Mingan, the morepristine site, the mean total PAH in 1994 was almost equal

to that found during the winter of 1995. This was 6.24-46times lower than those observed in St. Anicet and Villeroyduring all years. The medians for total PAHs, which are themore robust statistics, were also higher in St. Anicet andVilleroy than they were in Mingan. Variations in relativeproportions of PAHs can be seen among locations and duringall years (Figure 1). Phenanthrene was the most predominantPAH followed by fluoranthene and pyrene during all yearsand at all sites.

For the sake of comparability, ∑Ph + An + Fla + Py and∑Bep + BaP were compared with those reported from otherrural locations in North America (Table 2). The ∑Ph + An +Fla + Py in Villeroy and St. Anicet were 9-20 times higherthan those of Mingan. They were higher or equal to thosereported from the Great Lakes (19-21). The ∑BeP + BaP inVilleroy and St. Anicet were 5-46 times the levels reportedfrom Mingan and other rural areas.

Gas-Particle Partitioning. Assessment of atmosphericremoval mechanisms, reactivity, and health effects due toinhalation of organic pollutants present in the atmosphererequire investigations on gas-particle partitioning. Numer-ous past and more recent studies have lead to the conclusionthat the apparent particle to gas ratio and the vapor pressure

TABLE 1. Locations, Years, Number of Samples, TSP, andMean and Median ∑PAH Concentrationsa

location year NTSP

(µg m-3)mean ∑PAH

(ng m-3)median ∑PAH

(ng m-3) notes

Villeroy 1993 62 19.16 9.37 7.04 a1994 45 23 3.31 1.6 c1995 55 30.66 5.6 3.6 d1996 13 27.54 18.92 16.46 d

St. Anicet 1994 47 20 7.57 5.3 b1995 57 31.84 13.8 7.96 d1996 15 25.19 22.84 19.54 d

Mingan 1994 17 12.3 0.53 0.41 b1995 13 11.62 0.5 1.71 d

a Key: N ) number of samples. a ) Ph + An + Fla + Py + BaA +Chry + Bep + BaP + IcdP + BghiP. b ) Ph + An + Fla + Py + BaP.c ) Ph + An + Fla + Py. d ) a + BbF + BkF + Dba.

FIGURE 1. Variations in relative proportions of PAHs.

TABLE 2. PAH Concentrations in Villeroy, St. Anicet, Mingan,and Other Locations

location yearPh + An

+ Fla + PyBeP

+ BaP ref

Villeroy 1993 4.16 0.16 this work1994 4.61 this work1995 3.71 0.26 this work

St. Anicet 1994 7.61 .12 this work1995 7.43 1.41 this work1996 9.35 2.48 this work

Mingan 1994 0.49 0.011 this work1995 0.42 0.017 this work

Great Lakes 1989-1992 1.23-5.84 0.011-0.054 19-21

VOL. 37, NO. 10, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 2095

of the subcooled liquid are related through

where TSP is the total suspended particle concentrations (µgm-3), A and F are adsorbent- and filter-retained concentra-tions of organic compounds (ng m-3) (22-27), and PL

s is thevapor pressure of the compound at the sampling temperature.

Log F/A(TSP) for PAHs were plotted against log PLs as

shown in Figure 2. Average temperatures (K) were 264.35and 263.21 in Villeroy in 1995 and 1996, 278.64 and 266.52in St. Anicet in 1995 and 1996, and 268.84 in Mingan. Valuesfor PL

s were from refs 28 and 29. Positive linear relationshipswere obtained (r 2 ) 0.79-0.99). Slopes and intercepts for allsampling locations and from literature are summarized inTable 3. The values for mr in eq 1 in Villeroy and St. Anicetwere higher than -1 and comparable with those reportedfrom other rural locations in North America. In Mingan, themr value was also higher than the expected value of -1 andhigher than the ones found in Villeroy, St. Anicet, and otherrural areas (19, 27, 30-32). The deviations from the expectedvalue of -1 have been attributed to the kinetic restraints,sampling artifacts, and thermodynamic factors (33). But theslope equaling -1 does not always define equilibriumbetween the particulates and the gas phases (34, 35). It hasbeen shown that these shallow slopes may be due to thenonexchangeable PAHs in the aerosols (16, 27, 36).

Among the few existing predictive models, there is theJunge-Pankow model. This is defined by

where θ ) 1.5 × 10-6 for average background air in Villeroyand St. Anicet and θ ) 0.42 × 10-6 for clean continentalbackground (37, 38) in Mingan and PL

s is the vapor pressureof the compound (Pa). Curves for all sites were derived usingdifferent Kp from the log Kp equations for each site. Theseare shown in Figure 3. At PL

s ) 1.26 × 10-5 Pa, which is knownto be an equipartitioning vapor-pressure of PAHs in ruralair, the field percentages of PAHs on particles were 74% inSt. Anicet in 1995, 84% in St. Anicet in 1996, 50% in Villeroyin 1995, 87% in Villeroy in 1996, and 52% in Mingan in 1995.In literature, the range for measured values in rural areas is55-85%. The curve for St. Anicet in 1995 fell on that of LakeSuperior 2 (Figure 4). In Lake Superior, most of the pollutionis due to long-range transport (39). Other studies includedin Figure 4 were from refs 19, 27, 30, and 31. The particle-

FIGURE 2. Plots of log Kp against log P Ls (Pa).

TABLE 3. Slopes and Intercepts for Log Kp versus Log PLs from

This Work and Literature

slopes intercepts ref

St. Anicet, 1995 -0.545 -3.71 this workSt. Anicet, 1996 -0.514 -3.59Villeroy, 1995 -0.467 -3.77Villeroy, 1996 -0.568 -3.41Mingan, 1995 -0.312 -2.55Coastal Oregon -0.724 -4.94 27Lake Superior 1 -0.614 -4.25 19Lake Superior 2 -0.586 -3.83 30Lake Erie & Ontario -0.688 -4.14 31Green Bay -1.00 -5.47 32

log F/A(TSP) ) log Kp ) mr log PLs + br (3)

φ ) cθ/PLs + cθ (4)

2096 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 10, 2003

gas expression is also related to the organic matter absorptionand soot carbon adsorption through (12)

where fec ) 0.051 is the fraction of elemental carbon in theatmospheric particles, aac ) 1000 m3 g-1 is the surface areaof elemental carbon, fom ) 0.15 is the fraction of organicmatter in the particules, γoct and γom are the activity of the

absorbing substance in octanol and particulate organicmatter, MWoct and MWom are molecular weights of octanoland organic matter, Foct is the density of octanol and equals0.820 kg m-3, and Koa is the octanol-air coefficient calculatedfrom (40)

The Koa for chrysene was calculated by assuming that thetemperature slope for Koa is close to that for vapor pressure

FIGURE 3. Plots of fraction of PAH on particulates against log P Ls (Pa) in Villeroy, St. Anicet, and Mingan.

FIGURE 4. Plots of fraction of PAH on particulates against log P Ls (Pa) from this work and other rural locations.

Kp ) ( fomMWoctγ/FoctMWomγom × 1012Koa) +( fecaec/aac × 1012)KSA (5)

log Koa ) -1.04 log PLs + 6.441 (6)

VOL. 37, NO. 10, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 2097

in absolute value. Since the vapor pressure slope for chryseneis -5200 (29), we assumed that the slope for Koa was 5200withlog Koa ) 5200/T + b ) 10.525 at 293.5; b ) -7.213. Theequation for chrysene is

and the soot-air partition coefficient is

where Ksw is the soot-water coefficient (40) and H′ is thedimensionless Henry’s law constant at the sampling tem-perature expressed by (42)

where ∆H (kJ mol-1) and ∆S (kJ mol-1 K-1) are the enthalpyand entropy of volatilization. A comparison of measured andpredicted φ values for phenanthrene, fluoranthene, pyrene,and chrysene at St. Anicet is shown in Table 4. At St. Anicet,the measured PAH fraction on particles were close to thosepredicted from KSA, demonstrating that adsorption onto sootparticles was the predominant process. It has been shownthat adsorption onto aerosol soot carbon governs gas-particle partitioning of PAHs (12). For Mingan and Villeroy,Ksa were not calculated since the mean sampling temper-atures were less than 277 K. Equation 9 can be applied onlyfor temperatures ranging from 277 to 304 K (42), and the useof general equations relating the enthalpies to the temper-ature are sources of high uncertainties in predicting variationsin the enthalpies (42). However, ratios of particulate PAHsof the same molecular weight and very different reactivitieswere investigated at all sites during the same sampling days.The mean ratios of BeP/BaP were 3.62 ( 4.31, 4.08 ( 2.89,and 3.58 ( 3.78 in Mingan, St. Anicet, and Villeroy, respec-tively. Those of PH/AN were 15.29 ( 12.02, 15.86 ( 10.64,and 15.88 ( 9.43 in Mingan, St. Anicet, and Villeroy. Thesemean ratios, which are about equal at all sites, suggest thatlateral mixing of PAH particulates might have been fasterthan the photochemical reactions of the PAHs present onparticles. The major loss mechanism for gas-phase PAHs inthe atmosphere is the reaction with OH radicals. Reactionswith other oxidants (e.g., O3 and NOx) are minor as comparedwith those with OH radicals (43). Although OH radical reactionrate for gas-phase PAHs is fast, there are factors that stillallow the PAHs to travel long distances. OH radicals decreasefrom tropical to polar latitudes, demonstrating that atmo-spheric lifetimes of gas-phase PAHs become longer as theygo further north. Also, OH radicals are present in air duringthe day but not at night. Therefore, PAHs have longer lifetimeswhen they move during the night. Sorption of PAHs toaerosols may increase their lifetimes. PAHs sorbed onto flyash particles did not photodegrade (13). Half-lives of selectedPAHs were found to be 200 h higher on different substrates(44). A similarity in the ratios of particle-associated PAHs ofthe same molecular and very different reactivities, due to thehigher rate of lateral mixing of air particles as compared tothat of the PAH photodegradation, has been reported from

Chesapeake Bay (45). Therefore, it was surmised that theparticle-bearing PAHs at Mingan and Villeroy should be ofthe same nature as those found in St. Anicet where theadsorption onto soot particles was the major mechanism.When adsorption is the predominant mechanism, the slope(mp) in the relation of log Kp versus 1/T should be equal to(45, 46)

where Hdesorption ) the enthalpy of desorption from the surfacein J mol-1; R ) 8.314 J K-1 mol-1, the gas constant; and Tamb

) the center of the ambient temperature range (K) duringsampling time. In this work, log Kp were plotted against 1/Tfor phenanthrene and fluoranthene. The regression param-eters are displayed in Table 5. The heats of desorption forphenanthrene and fluoranthene at all sites were calculatedusing eq 8 and were in kilojoules (kJ). For phenanthrene,they were 62.33, 62.02, and 62.01 in Mingan, St. Anicet, andVilleroy, respectively. For fluoranthene, these were 67.42,63.08, and 65.07 in Mingan, St.Anicet, and Villeroy, respec-tively. Although Mingan is a more pristine location, the heatsof desorption for phenanthrene and fluoranthene werealmost equal to those estimated from St. Anicet and Villeroy.These were in the range of the values reported from LakeSuperior (47). The similarity in the energies of desorptionindicates that particle-bearing PAHs may be of the samenature and size and coming from the same source.

AcknowledgmentsWe thank Dr. M. Bertrand (University of Montreal) foroverseeing the analysis of the samples during the project.This project was funded by St. Lawrence Action Plan (Vision2000)-Environment Canada.

Literature Cited(1) Gigliotti, C. L.; Dachs, J.; Nelson, E. D.; Brunciak, A. P.;

Einsenreich, S. J. Environ. Sci. Technol. 2000, 34, 3547-3554.(2) Kavouras, I. G.; Koutrakis, P.; Tsapakis, M.; Lagoudaki, E.;

Stephanou, E. G.; Baer, V. D.; Oyola, P. Environ. Sci. Technol.2001, 35, 2288-2294.

(3) Jones, J. D.; Van Dollah, R. F.; Fulton, M. H. Abstracts of Papers,20th Annual Meeting of the Society of Environmental Toxicologyand Chemistry, Philadelphia, PA, 1999; PMPO17.

(4) Williamson, M. W.; Chandler, G. T.; Porter, D. E. Abstracts ofPapers, 20th Annual Meeting of the Society of EnvironmentalToxicology and chemistry, Philadelphia, PA, 1999; PMPO2.

(5) Wilson, J. Y.; Addison, R. F.; Martens, D.; Gordon, R.; Glickman,B. Can. J. Fish. Aquat. Sci. 2000, 57 (2), 405-413.

(6) Yaffe, D.; Cohen, Y.; Arey, J.; Atkinson, R.; Grosovsky, A. J. RiskAnal. 2001, 21, 275-294.

(7) Bishop, C. A.; Struger, J.; Shirose, L. J.; Dunn, L.; Campbell, G.D. Water Qual. Res. J. Can. 2000, 35, 437-474.

(8) Dimashki, M.; Lim, L. H.; Harrisson, R. M.; Harrad, S. Environ.Sci. Technol. 2001, 35, 2264-2267.

(9) Poissant, L.; Koprivnjak, J. F.; Fecteau, M. Substances toxiquesaeroportees dans la vallee du fleuve Saint-Laurent; Ministere del’Environnement: 2000; Catalogue No. En 56-137/1999F; ISBN0-662-83784-3.

(10) Cretney, W. J.; Wong, C. S.; McDonald, R. W.; Erickson, P. E.;Fowler, B. R. Can. Tech. Rep. Hydrogr. Ocean Sci. 1982, 18, 162-195.

(11) Mackay, D.; Hickie, B. Chemosphere 2000, 41, 681-692.(12) Dachs, J.; Einsenrich, S. J. Environ. Sci. Technol. 2000, 34, 3690-

3697.

TABLE 4. Comparison of Measured and Calculated ParticulateFractions in St. Anicet

O

location PAH measd Koa KSA Koa + KSA P Ls

St. Anicet PH 0.014 0.002 0.016 0.018 0.0072FLA 0.23 0.046 0.13 0.17 0.087PY 0.28 0.031 0.12 0.14 0.13CHRY 0.86 0.62 0.94 0.95 0.81

log Koa(CHRY) ) 5200/T - 7.213 (7)

KSA ) Ksw/H′ (8)

H′ ) exp(-∆H/RT + ∆S/R) (9)

TABLE 5. Regression Parameters for Log Kp versus 1/T (K)

phenanthrene fluoranthene

site slope intercept r 2 slope intercept r 2

Villeroy 3179 -14.286 0.665 3339 -14.042 0.725St. Anicet 3179 -14.345 0.609 3295 -14.033 0.612Mingan 3197 -14.027 0.518 3197 -14.027 0.518

mp ) Hdesorption/2.303R - Tamb/4.606 (10)

2098 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 10, 2003

(13) Kormacher, W. A.; Wehry, E. L.; Mamantov, G.; Natusch, D. F.S. Environ. Sci. Technol. 1980, 14, 1094-1099.

(14) Halsall, C. J.; Barrie, L. A.; Fellinn, P.; Muir, D. C. G.; Billeck, B.N.; Lockhart, L.; Rovinsky, F. Ya.; Pasthukhov, B. Environ. Sci.Technol. 1997, 31, 3593-3599.

(15) Hallsall, C. J.; Sweetman, A. J.; Barrie, L. A.; Jones, K. C. Atmos.Environ. 2001, 35, 255-267.

(16) Pankow, J. F. Environ. Sci. Technol. 1989, 23, 1107-1111.(17) Poissant, L.; Koprivnjak, J. F. Environ. Sci. Technol. 1996, 30,

845-851.(18) El Khoury, M.; Bertrand, M. J. Apport atmospherique de certains

composes organiques toxiques dans la composition chimique dufleuve Saint-Laurent et quelques uns des tributaires, Protocoled’echantillonnage et d’analyse; Departement de Chimie, Uni-versite de Montreal: Montreal, 1991; p 42.

(19) Baker, J. E.; Eisenreich, S. J. Environ. Sci. Technol. 1990, 24,342-352.

(20) Cotham, W. E. Ph.D. Dissertation, University of South Carolina,1990.

(21) Gatz, D. F.; Sweet, C. W.; Basu, I.; Vermette, S.; Harlin, K.; Bauer,S. Great Lakes Integrated Atmospheric Deposition Network(IADN). Data report 1990-1992; Illinois State Water Survey:Champaign, IL, 1994.

(22) Yamasaki, H.; Kuwata, K.; Miyamoto, H. Environ. Sci. Technol.1982, 16, 189-194.

(23) Bidleman, T. F.; Billings, W. N.; Foreman, W. T. Environ. Sci.Technol. 1986, 20, 1038-1043.

(24) Foreman, W. T.; Bidleman, T. F. Environ. Sci. Technol. 1987, 21,869-875.

(25) Foreman, W. T.; Bidleman, T. F. Atmos. Environ. 1990, 24A,2405-2416.

(26) Pankow, J. F. Atmos. Environ. 1987, 21, 2275-2283.(27) Ligocki, M. P.; Pankow, J. F. Environ. Sci. Technol. 1989, 23,

75-83.(28) Hinckley, D. A.; Bidleman, T. F.; Foreman, W. T. J. Chem. Eng.

Data 1990, 35, 232-237.(29) Yamasaki, H.; Kuwata, K.; Kuge, Y. Nippon Kagaku Kaishi 1984,

8, 1324-1329.(30) McVeety, B. D.; Hites, R. A. Atmos. Environ. 1988, 22, 511-536.

(31) Hoff, R. M.; Strachan, W. M. J.; Sweet, C. W.; Chan, C. H.;Shackeleton, M.; Bidleman, T. F.; Brice, K. A.; Burniston, D. A.;Cussion, S.; Gatz, D. F.; Karlin, K.; Schroeder, W. H. Atmos.Environ. 1996, 30, 3505-3527.

(32) Cotham, W. E.; Bidleman, T. F. Environ. Sci. Technol. 1995,2782-2789.

(33) Pankow, J. F.; Bidleman, T. F. Atmos. Environ. 1992, 25A, 2241-2249.

(34) Goss, K. U.; Schwartzenbach, R. P. Environ. Sci. Technol. 1998,32, 2025-2032.

(35) Simcik, M. F.; Franz, T. P.; Zhang, H.; Eisenrich, S. J. Environ.Sci. Technol. 1998, 32, 251-257.

(36) Fernandez, P.; Grimalt, J. O.; Vilanova, R. M. Environ. Sci. Technol.2002, 36, 1162-1168.

(37) Bidleman, T. F. Environ. Sci. Technol. 1988, 22, 361-367.(38) Whitby, K. T. Atmos. Environ. 1978, 12, 135-159.(39) U.S. EPA. Identifying Sources Region of Selected Persistent Toxic

Substances in the U.S. Third Progress Report Prepared for theIAQAB (International Air Quality Advisory Board of the Inter-national Joint Commission). Draft, September 1999.

(40) Harner, T.; Bidleman, T. F. J. Chem. Eng. Data. 1998, 43, 40-45.(41) Walters, R. W.; Luthy, R. G. Environ. Sci. Technol. 1984, 18,

395-403.(42) Bamford, H. A.; Poster, D. L.; Baker, J. E. Environ. Toxicol. Chem.

1999, 18, 9, 1905-1912.(43) Arey, J. In PAHs and Related Compounds-Chemistry; Neilson,

A. H., Ed.; Springer-Verlag: Berlin, Heidelberg, New York, 1998;Vol. 3, Part 1.

(44) Behymer, T. D.; Hites, R. A. Environ. Sci. Technol. 1988, 22,1311-1319.

(45) Gustafson, K. E.; Dickut, R. M. Environ. Sci. Technol. 1997, 31,140-147.

(46) Pankow, J. F. Atmos. Environ. 1991, 25A, 2229-2239.(47) McVeety, B. D. Ph.D. Dissertation, Indiana University, 1986.

Received for review November 20, 2002. Revised manuscriptreceived March 12, 2003. Accepted March 12, 2003.

ES0210345

VOL. 37, NO. 10, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 2099