Environ. Sci. Technol. 1986, 20, 1033-1038
Ionic Alkylleads in Herring Gulls from the Great Lakes Region Donald S. Forsytht and Wllllam D. Marshall"
Department of Food Science and Agricultural Chemistry, Macdonald College, Ste. Anne d e Bellevue, Quebec, Canada H9X IC0 Herring gull (Larus argentatus) tissues, collected from various breeding colonies in the Great Lakes, were examined to determine alkyllead levels and possible alkyllead sources into the Great Lakes region. Ionic trialkyl- and dialkyllead species (R,Pb+, R2Pb2+;R = Me, Et) were quantitated by gas chromatography-atomic absorption spectrometry. The extraction procedure was tested at trace levels (3-4 ppb as Pb) with four domestic chicken (Gallus domesticus) tissues. Trimethyllead was found in two avian species and all examined tissues. Methyllead levels frequently exceeded ethyllead levels with no direct automotive source. Correlation between alkyllead levels in herring gull tissue and lake sediment lead levels suggests possible methylation but not ethylation of inorganic lead. The methyllead concentration trend in gull tissues, Lake Ontario > Lake Huron N Lake Erie > Lake Superior, was opposite to the ethyllead concentration trend, Lake Superior > Lake Huron > Lake Erie N Lake Ontario. 4
Introduction Organolead compounds are ubiquitous in the environment, having been detected in air (1-3), street dust (4), rainwater (5), cod, lobster, and mackerel (6), freshwater fish (7,8), pigeons (9),and human brains (4-84% of the total lead) (10). Investigations of the environmental fate of tetralkylleads (used as gasoline antiknock additives) have been hampered by the lack of suitable methodology for determining ionic alkyllead species (R3Pb+and R,Pb2+; R = Me, Et) in biological tissue. Ionic alkylleads occur from the hydrolysis (11-13), photolysis (11, 141, and metabolic dealkylation (15-1 7) of tetraalkylleads. Although alkylleads may constitute only a small percentage of the total lead burden of an organism, they are likely to appreciably enhance toxicity. Mammalian acute toxicity studies indicate that trialkyllead salts are 10-100 times more toxic and dialkylleads are 1-10 times more toxic than inorganic lead salts (18,19). Acute organolead poisoning in humans resulted in hyperactivity, tremors, periodic convulsions, and aggressive behavior (20), indicating involvement of the central nervous system. Environmental methylation of trialkyllead salts is well established (21-24) although it remains unclear whether the methylation is predominately biologically (21,22) or chemically (23,24) mediated. The biomethylation of Pb(11) salts remains controversial (21,24-26). Chemical alkylation of Pb(I1) salts in the presence of aqueous carbonium ion reagents produced Me,Pb (27, 28) and ionic methyllead compounds (28). With the presence of methyl iodide in the environment (29), this could represent an important pathway for organolead into the environment. North American avian populations have not been previously studied for the presence of organolead compounds. The herring gull (Larus argentatus) is used as the monitor species for toxic contaminants in the Great Lakes (30). Great Lakes breeding populations tend to remain in the region of the colony and therefore are considered a valuable Present address: Bureau of Chemical Safety, Food Directorate, Health Protection Branch, Health and Welfare Canada, Ottawa, Ontario, Canada KIA OL2. 0013-936X/86/0920-1033$01.50/0
regional contamination indicator (31). The objectives of this study were to (a) determine the relative ionic alkyllead levels in various regions of the Great Lakes by analyzing herring gull tissue collected from different breeding colonies and (b) examine possible alkyllead sources into the Great Lakes region.
Materials and Methods Reagents and Standards. Alkyllead chlorides (R,PbCl, R,PbCl,; R = Me, Et) and alkyllead butylates (R,BuPb, R2Bu,Pb; R = Me, Et) were prepared as previously described (32,33). All chemicals were ACS reagent grade or better, and chromatographic support gases were prepurified grade. Enzyme preparations were obtained from the Sigma Chemical Co., St. Louis, MO. The ammoniacal buffer solution consisted of diammonium citrate (22.6 g), potassium cyanide (4.0 g), and sodium sulfite (24.0 g) made up to 250 mL with distilled water. The pH was adjusted to 10.0 with concentrated ammonium hydroxide. Sample Collection. Herring gull tissues were collected from various breeding colonies (Figure 1)and homogenized by the Canadian Wildlife Service. Eggs were collected fresh and stored at 4 OC until homogenized. Egg homogenate (minus shell) was prepared by using a Sorvall Omni Mixer (Du Pont Instruments). The egg homogenates were stored at -40 "C. Adult tissues were initially stored at -20 "C, thawed, homogenized by using a Sorvall Omni Mixer, and then refrozen at -40 "C. Herring gull liver and kidney samples from each colony consisted of tissue pools of 10 mixed sex mature or immature birds from each colony with the exception of Middle Island mature bird liver, which consisted of seven pooled livers. Samples were collected during the summer of 1983. Herring gull whole egg homogenate samples were from tissue pools of 9-13 eggs, collected in the spring of 1982 and 1983. Domestic chicken (Gallus domesticus) tissue for recovery experiments was collected from mixed sex mature birds donated by the Poultry Unit at Macdonald College in May 1984. Tissue pools, consisting of 15 brains, 3 kidneys, or 2 livers, were used for analyses. Whole chicken egg (minus shell) was prepared with a Polytron homogenizer (Brinkmann Instruments) at low speed. The other chicken tissues were homogenized with a Teflon pestle tissue grinder. All tissue homogenates were stored at -20 "C until needed. Enzymatic Hydrolysis. Samples ( N = 3) of tissue homogenate (2.50 g) were incubated in 50-mL Nalgene screw-cap centrifuge tubes at 37 "C for 24 h in 20 mL of 5 % ethanoll0.5 M sodium dihydrogen phosphate buffer (pH 7.5) containing 40 mg each of lipase (type VII) and protease (type XIV). Extraction. Ammoniacal buffer (5 mL) was added to the hydrolysate. The diluted hydrolysate was then extracted 3 times with 0.01% (w/v) dithizone (10 mL) in 50 % (v/v) benzenelhexane. The pooled dithizone extracts were centrifuged at 4400 rpm for 10 min (5 "C, IEC Model PR-1centrifuge, rotor 845) to hasten phase separation. The combined organic extracts were back-extracted 3 times with 10 mL of 0.15 M "0,. The combined acidic extracts were neutralized with 1 M NaOH (4.5 mL) and
0 1986 American Chemical Society
Environ. Sci. Technol., Vol. 20, No. 10, 1986 1033
Table I. Absolute Retention Times and Retention Indices of Mixed Methylethylleads (Based on Retention Times) Relative to Alkylbutyllead Standards analyte
N
absolute retention time (RT), min
retentiona index (I)
Me3BuPb Me2EtBuPb MeEt2BuPb Me,Bu2Pb EtsBuPb EtMeBu,Pb EtzBuzPb BudPb
7 8 9 10 10 11 12 16
9.21 11.22 13.04 14.23 14.65 15.77 17.15 21.09
700 800 (805)b 900 (906)b 1000 1000 1100 (1105)b 1200 1600
+
Figure 1. Location of herring gull breeding colonies. 1, Snake Island: 2, Scotch Bonnet Island: 3, Muggs Island; 4, Hamilton Harbour: 5, Niagara River: 6, Middle Island: 7, Fighting Island: 8, Channel Shelter Islands; 9, Double Island; 10, Agawa Rock.
further basified with ammoniacal buffer (5 mL). The alkyllead salts were extracted 3 times from the basified aqueous extract with 5 mL of 0.01% (w/v) dithizone solution. The pooled organolead dithizonates were concentrated to 1.0 mL in precalibrated tubes (equipped with screw-cap tops and Teflon liners) under a gentle stream of nitrogen a t 30 "C. Derivatization. n-Butylmagnesium chloride (0.5 mL, 2.27 M; Alfa Products, Ventron Corp., Danvers, MA) was added to the dithizone extract under nitrogen. The solution was vortexed 10 s, magnetically stirred for 10 min a t ambient temperature, and then cooled in an ice bath. Excess Grignard reagent was destroyed by dropwise addition of double-distilled water. A total of 10 mL of water was added, and the mixture was shaken for 30 s and then centrifuged for 5 min at 1550 rpm (Precision Clinical Centrifuge, Precision Scientific Co.). The aqueous layer was removed, and the organic phase was extracted with another 10 mL of water. The organic phase was removed, dried over sodium sulfate, and placed in a sample vial, which was capped for immediate analysis. Sample Analysis. A gas chromatograph (GC)-silica furnace-atomic absorption spectrometer (QTAAS) system previously described (33) was used for quantitation of samples. The GC was fitted with a 1.8 m, 6 mm o.d., 2 mm i.d. glass column packed with 10% OV-101 on 80-100 mesh Supelcoport and equipped with an autoinjector. Optimized operating conditions were as follows: carrier gas helium, 35 mL min-'; injector temperature 200 "C; temperature program 50 "C (1 min) linear increase (8 "C min-l) to 250 "C (1min). The quartz T-tube furnace temperature was 900 "C with a hydrogen makeup gas flow rate of 50 mL min-l. Seven analytes were determined in each sample (four trialkyllead salts and three dialkyllead salts). Each butylated extract was quantitated 3 times by comparison with external standards containing Me,BuPb, MezBuzPb, Et3BuPb, and EtzBuzPb. Methylethyllead compounds were identified by prediction of retention time using Kovats retention index (34) from retention times of the alkylbutyllead standards (Table I). Actual retention times of methylethyllead compounds were determined from transalkylation reaction products. Quantitation of the methylethyllead compounds was achieved by comparison with a similar analyte for which standards were available. Thus, quantitation for MezEtPb+, MeEtzPb+, and MeEtPb2+was based on the instrumental response and recoveries for Me3Pb+,Et3Pb+,and EtzPb2+,respectively. 1034
Environ. Sci. Technol., Vol. 20,
No. 10, 1986
"Predicted retention index (34). bl= lOON lOOn[(RT(,,.RT,,,)/(RT,,,+,, - RT,(,)] where N = number of carbons in analyte, n = carbon number difference between R(N) and R(N + n), A = analyte for which retention index was calculated, R(N) = analyte containing N carbons, and R ( N + n) = analyte containing N + n carbons,
Table 11. Mean Recoveries of Ionic Alkyllead Compounds from Biological Tissue" tissue
meanb analyte % recoveryCf SD Me3PbCln8 Et,PbCln8 Me2PbC12* Et2PbC12*
egg liver kidney brain
105 f 9a 99 & 18a 94 f 15a 83 f 9a
82 f 4a 74 f 7a 74 f 9a 85 f 7a
94 f 29a 26 f 7b 25 f 7b 29 f l l b
95 f 12a 66 f 9b 71 f 10b 59 f l l b
" Spiked at 3-4 ppb (as Pb). Means with the same letter in column are not significantly different ( p = 0.05, Newman-Keuls multiple range test). Note, F test takes precedence over Newman-Keuls multiple range test. N = 4 replicate determinations. 'F test: asterisk, 0.05 significance; ne, not significant. Recovery Experiments. Four samples of each tested tissue were spiked a t a level of 3-4 ppb (as Pb) with a mixture of Me3PbC1,Me2PbC1,, Et3PbC1, and EtzPbC12. The percentage recovery of each analyte was determined by dividing the mean peak area of the recovered butylate by the mean peak area of a butylated spike solution diluted to the expected (assuming 100% recovery) concentration.
Results and Discussion Enzymatic Hydrolysis. A previous study (32) with chicken whole egg homogenate indicated that at 30-50 ppb (as Pb) methyllead salts but not ethyllead salts were strongly retained by the sample matrix. Enzymatic hydrolysis, with a mixture of protease and lipase, improved recoveries. This technique was 7 2 f 9% and 78 f 9% effective (relative to classical acid hydrolysis) at releasing amino nitrogen after 24 or 48 h of incubation, respectively. The present hydrolysis procedure was 88 f 5% ,72 A 18% , 48 f 8%,and 39 f 4% effective with liver, kidney, brain, and egg tissue, respectively, after 24 h of incubation. All hydrolyzed tissues were suitable for liquid-liquid extraction regardless of the extent of hydrolysis as most fibrillar material was solubilized and foaming (when shaken) was suppressed. Enzymatic hydrolysis for 24 h was therefore used in all subsequent studies as a standard pretreatment of environmental samples to minimize emulsification and organolead-sample tissue association. Recoveries. Trialkyllead recoveries were generally high and did not vary significantly (p = 0.05, Newman-Keuls multiple range test) among the four tissues (Table 11). However, the dialkyllead recoveries were suppressed and were significantly lower 03 = 0.05, Newman-Keuls multiple range test) from liver, brain, and kidney tissue than from egg*
Table 111. Ionic Alkyllead Levels (as Alkylbutylleads) i n Herring Gull Liver and Kidney Samples Collected from Various Great Lakes Colonies
source Lake Superior Agawa Rock mature immature Lake Huron Double Island mature immature Lake Erie Middle Island mature immature Lake Ontario Hamilton Harbour mature immature
Me3Pbt liver kidney
mean' alkyllead* concentrationc f SD, ng/g wet wt Me2EtPb' MezPbZt EtsPb' liver kidney liver kidney liver kidney
Et2Pb2' liver kidney
2.3 f 0.1 1.3 f 0.1
4.5 i 0.2 NDd 2.3 f 0.1 ND
ND ND
1.7 f 0.3 1.2 f 0.5 7.3 f 0.2 8.8 f 0.4 2.2 f 0.6 1.6 f 0.2 1.6 f 0.4 1.4 f 1.1 2.2 f 0.3 6.3 f 0.6 0.8 f 0.1 1.4 f 0.2
3.0 f 0.1 3.3 f 0.1
5.5 i 0.3 ND 5.5 f 0.3 ND
ND ND
ND 2.4 i 0.2 3.1 f 0.4 4.5 i 0.1 0.9 f 0.3 0.8 f 0.3 1.8 f 0.6 3.0 f 0.5 1.1f 0.1 2.5 f 0.1 0.3 f 0.1 ND
3.5 f 0.1 1.7 f 0.3
4.5 f 0.3 ND 4.3 f 0.1 0.3 f 0 . 2
0.3 f 0.1 2.6 f 0.3 2.0 f 0.3 1.2 f 0.1 1.0 f 0.1 0.3 f 0.2 0.4 f 0.2 0.6 f 0.1 1.6 f 0.4 1.9 f 0.4 0.8 f 0.1 1.1 f 0.1 0.4 f 0.2 ND
7.9 f 0.1 15.3 f 0.1 0.3 f 0.1 0.6 f 0.1 3.2 f 0.3 4.4 i 1.0 0.8 f 0.1 0.8 f 0.1 0.8 f 0.1 0.6 f 0.2 ND 6.7 f 0.1 18.7 f 1.1 0.4 f 0.1 1.2 f 0.1 4.2 f 0.9 5.8 f 0.3 0.9 f 0.1 1.6 f 0.1 ND
'Calculated from three replicate injections of a single determination. No MeEtPb2' was detected in any of the samples and Et2MePbt was detected in only one sample (0.3 f 0.1 ng/g, immature, Hamilton Harbour). "Corrected for recovery. dND, not detected. Dialkyllead recoveries may have been lowered by tissue sample metabolic activity. Egg is composed primarily of storage proteins whereas the other examined tissues have cellular metabolic activity. However, trialkyllead recoveries should also have been lowered by dealkylation yet were not significantly different ( p = 0.05, Newman-Keuls multiple range test) among the four tissues. Chau et al. (7) reported lowered Me2Pb2+recoveries from fish samples and suggested that sample-induced decomposition of Me2Pb2+to inorganic lead occurred. Other possible factors involved in the lowering of dialkyllead recoveries are competitive binding to the sample hydrolysate and/or physical occlusion of R2Pb2+within insoluble hydrolysate products. Environmental Samples. Ionic alkyllead compounds were detected in all the sampled herring gull breeding colonies (Table 111). The GC-AAS chromatogram of a gull kidney sample (Figure 2) illustrates typical GC-AAS system performance during the study. The herring gull does not appear to bioaccumulate ionic alkylleads as mature and immature bird analyte levels were not significantly different (p = 0.05, paired comparison t test), except EbPb2+kidney levels, which contained several zero values. This indicates that the mature and immature bird alkylead levels result from equilibration to environmental levels of alkyllead which were characteristic-ofthe area surrounding the sampling site. The mature and immature bird analyte levels were therefore pooled for analysis of variance, which showed significant differences ( p = 0.05) between the sampling sites (Figure 3). Trimethyllead levels in kidney samples from Hamilton Harbour were significantly greater ( p = 0.05, Duncan's multiple range test) than in the other collected tissues. Liver samples from Agawa Rock, Double Island, Middle Island, and Agawa Rock kidney samples had significantly less (p = 0.05, Duncan's multiple range test) Me,Pb+ than most other tissues. Similarly, Me2Pb2+ levels in kidney tissue from Hamilton Harbour gulls were significantly greater ( p = 0.05, Duncan's multiple range test) than Me2Pb2+levels from other sites. Liver samples from Double Island gulls had significantly less ( p = 0.05, Duncan's multiple range test) Me2Pb2+than Hamilton Harbour tissue samples. Triethyllead levels, however, were significantly greater (p = 0.05, Duncan's multiple range test) in gull kidney samples collected from Agawa Rock than in gull tissues collected from other sites. The EbPb2+
Figure 2. GC-AAS chromatogram of herring gull kidney sample containing the following: 1, Me,BuPb; 2, Me,EtBuPb; 3, Me,Bu,Pb; 4, EtaBuPb; 5, Et,BU,Pb; 6, BuhPb.
levels were not significantly different (p = 0.05, Duncan's multiple range test) among collected tissues (Figure 3). Thus, two trends in the alkyllead levels were apparent. The methyllead concentration trend, Lake Ontario > Lake Huron = Lake Erie > Lake Superior, was opposite to the ethyllead concentration trend, Lake Superior > Lake Huron > Lake Erie N Lake Ontario. Gull egg homogenate samples contained primarily Me3Pb+ (9 out of 10 samples) (Table IV). The lowest levels of Me3Pb+were observed in egg homogenates from Lake Huron and Lake Superior whereas eggs collected from Lake Erie or the Lake Ontario region (including the Niagara and Detroit Rivers) contained the highest levels (Table IV). Dimethyllead and diethyllead were present in only two and three samples,. respectively. Correlation (Pearson correlation coefficients) of Me3Pb+levels between mature bird tissues and egg samples from the same breeding colony were significant (Table V, part A). Immature bird tissues and egg Me3Pb+levels were not significantly (p = 0.05) correlated (Table V, part B). As only mature birds produce egg clutches, the significant correlation suggests that alkyllead burdens in the egg occur during egg formation. The automotive source of methyllead into the environment would probably not account for the extensive presence of Me,Pb+ in the gull tissue samples as Et4Pb is the predominant organolead gasoline additive used in North Environ. Sci. Technol., Vol. 20, No. 10, 1986
1035
20
+J
15
JJ P)
z
m
\
-
m
C C
: 10 +J
m
L +J
c al 0
C 0
u c m
g
5
MeJPb
Me2Pb
EtJPb
Et2Pb
Alkyllead
Flgure 3. Effect on gull tissue ionic alkyllead levels (as alkylbutyllead) by sampllng site. Solid boxes, Hamilton Harbour liver; vertically lined boxes, Hamilton Harbour kidney; diagonally lined ( / / Iboxes, ) Agawa Rock liver; diagonally lined (\\\) boxes, Agawa Rock kidney; crosshatched boxes, Double Island liver; horizontally lined boxes, Double Island kidney; squared boxes, Middle Island liver; open boxes, Middle Island kidney. Bars with the same letter in each grouping are not significantly different (p = 0.05, Duncan's multiple range test).
Table IV. Ionic Alkyllead (as Alkylbutyllead) Levels in Herring Gull Whole Egg Homogenate from Great Lakes Colonies
Table V. Correlations between Trimethyllead Concentrations of Gull Egg Homogenate and Other Tissues liver
source Lake Superior Agawa Rock Lake Huron Channel Shelter Is. Double Island Lake Erie Middle Island Lake Ontario Scotch Bonnet Is. Snake Is. Muggs Is. Hamilton Harbour Niagara River Fighting Is., Detroit River
meanb concentrationCf SD, ng/g wet wt, for analytesn MesPbt MezPb2+ Et,Pb+ EtzPbZt 0.2 f 0.1 ND
ND
ND
ND
0.2 f 0.1 ND
ND ND
ND ND
0.3 f 0.1 ND
ND
ND
0.3 f 0.2 ND 0.3 f 0.1 ND ND ND
0.2 f 0.1
NDd
0.5 f 0.3 0.5 f 0.1 0.8 f 0.3 0.5 f 0.1 0.6 f 0.3 0.5 f 0.1
ND 0.6 f 0.2 0.7 f 0.2 ND ND ND
ND 0.3 f 0.1 ND 0.2 f 0.4 ND
No MezEtPbt, MeEtzPbt, or MeEtPbZt was detected in any of the samples. Calculated from three replicate injections of a single determination. Corrected for recovery. ND, not detected.
America. Canadian leaded gasoline contains only Et4Pb (3,35), whereas American leaded gasoline contains either Et4Pb or methylethyllead mixtures with Me4Pb constituting a minor fraction of total lead content (36, 37). Long-range movement of methyllead from American automotive sources appears unlikely as there is no significant difference (p = 0.05, Duncan's multiple range test) between the methyllead tissue levels of birds collected from Agawa Rock and Middle Island despite the large differences in proximity to major urban centers. Also, street dust and 1038 Environ. Sci. Technol., Vol. 20, No. 10, 1986
egg prob >Irl
kidney
(A) Mature Birds ( N = 4) 0.99773 0.96447 0.0023" 0.0355b
(B) Immature Birds ( N = 4) egg 0.93149 0.93024 prob >Irl 0.0685 0.0698 "0.01 significance.
0.05 significance.
Table VI. Ionic Alkyllead (as Alkylbutyllead) Levels in Domestic Chickens meanb concentration" f SD, ng/g wet wt, for analytes" Me3Pbt Me2Pb2 Et,Pbt Et2PbZt
source egg homogenate kidney brain liver
ND ND 3.0 f 0.3 6.0 f 0.2 1.1 f 0.1 ND 1.8 f 0.1 2.7 f 0.5
ND ND 0.2 f 0.1 1.4 f 0.2 ND ND 0.2 f 0.4 0.4 f 0.4
"No Me2EtPbt, MeEtzPbt, or MeEtPbZt was detected in any of the samples. b N = 4 replicate determinations. Corrected for recoverv. d ND. not detected. ~~
~~
soil samples collected in the Montreal region contained only ethyllead compounds, with no methylleads detected (38). Chicken tissue, sampled from the same region, where Et,Pb is used exclusively as the gasoline additive, however, contained more methyllead than ethyllead (Table VI), indicating that the presence and ubiquitousness of Me,Pb+ was not limited to gulls in the Great Lakes region. Other
Table VII. Correlations between Alkyllead Concentrations of Gull Tissue from Combined Bird" and Lake Sediment Lead Levelsb
Me3 kidney prob > Irl Me2 liver prob > Irl Mez kidney prob > Irl Et3 liver prob > Irl Et3 kidney prob > Irl Etz liver prob > ]rl Etz kidney prob > Jr( Pb sediment prob > Irl
Me3 liver
Me3 kidney
Me2 liver
Me2 kidney
Et3 liver
Et3 kidney
Et2 liver
Et2 kidney
0.93012 0.0008d 0.73750 0.036SC 0.89512 0.0027d -0.38201 0.3504 -0.54102 0.1662 -0.33288 0.4204 -0.38403 0.3476 0.8984 0.0024d
0.76956 0.0255O 0.98128 0.0001d -0.39185 0.3370 -0.49304 0.2144 -0.42325 0.2961 -0.48811 0.2198 0.9277 0.0009d
0.72785 0.0407O -0.38704 0.3435 '-0.48147 0.2271 -0.42332 0.2960 -0.38222 0.3501 0.5081 0.1986
-0.50458 0.2022 -0.56719 0.1426 -0.55426 0.1540 -0.56752 0.1423 0.9348 0.0007d
0.89496 0.0027d 0.93971 0.0005d 0.77227 0.0247' -0.4226 0.2969
0.85025 0.0075d 0.88009 0.003gd -0.4862 0.2218
0.84901 0.0077d -0.4123 0.3101
-0.4641 0.2467
ON = 8. bValues taken from Hodson et al. (39). O0.05 significance. dO.O1 significance.
methyllead sources into the environment, therefore, likely exist. Although environmental methylation of inorganic lead remains controversial, it is a possible source of methyllead. Correlation (Pearson correlation coefficients) between alkyllead tissue concentrations of combined (mature and immature) birds and lake sediment inorganic lead levels (from ref 39) were generally significant (p = 0.01, Table VII) for the methylleads, but not the ethylleads. Gull egg homogenate Me3Pb+ levels (Table IV) were also significantly correlated (r = 0.86788, p = 0.01) with lake sediment inorganic lead levels (from ref 39), indicating possible methylation but not ethylation of inorganic lead. Host biomethylation of ingested inorganic lead, although unresearched, is another possible methyllead source. Correlation of the alkyllead levels in gull tissue suggest that sequential metabolic dealkylation of the alkylleads, as reported in mammals (40), occurred. Trimethyllead and Me2Pb2+levels in liver and kidney samples of combined (mature and immature) birds (Table VII) were significantly correlated (p = 0.05, Pearson correlation coefficients). The correlation was independent of the tissue type or the methyllead species compared. Triethyllead and Et2Pb2+ concentrations were similarly correlated (Table ,VII). However, there were no significant (p = 0.05) correlations between any of the combinations of ethyllead with methyllead levels in any of the tissues. Sequential dealkylation of the alkylleads would tend to produce correlated Me3Pb+-Me2Pb2+and Et+,Pb+-EbPb2+body burdens. The absence of significant correlations between ethyllead and methyllead tissue levels indicate independent ethyl- and methyllead sources to the gull populations. Although correlations between ethyllead and methyllead tissue levels were not significant ( p = 0.05) a negative correlation occurred in every case (Table VII), reflecting the regional trends in alkyllead tissue concentrations described earlier. Ethyllead levels in Hamilton Harbour gulls were lower than those found in gulls collected from Agawa Rocks (Table 111),even though the former region has a larger automotive source of Et4Pb. The prevalence of ethyllead in only the less polluted regions cannot be presently explained and requires further study.
Conclusions Trimethyllead is ubiquitous, having been found in two avian species and all examined tissues. Hamilton Harbour birds contained much higher levels of Me3Pb+than Et3Pb+ with no known automotive source of Me3Pb+. Triethyllead
tissue levels were greatest in Agawa Rock gull tissues, a relatively undeveloped region with a smaller potential automotive source of ethyllead salts than the region surrounding Hamilton Harbour. Significant correlations between alkyllead levels in sampled herring gull tissue and lake sediment lead levels suggest methylation (possibly microbial) but not ethylation of inorganic lead. The ethyland methyllead tissue levels in the gull colonies were not significantly correlated, indicating independent ethyl- and methyllead sources to the gull populations. The ethyl- and methyllead levels were, however, inversely related, with opposite regional trends occurring.
Acknowledgments Our sincere appreciation to Keith Marshall, Canadian Wildlife Service, for the gull tissue samples. Registry No. Me3Pb+, 14570-16-2; Me2EtPbf, 103730-90-1; Me2Pb2+,21774-13-0; Et3Pbf, 14570-15-1; EtzPb2+,24952-65-6;
Pb, 7439-92-1.
Literature Cited (1) Reamer, D. C.; Zoller, W. H.; O'Haver, T. C. Anal Chem. 1978,50, 1449-1453. (2) Nielsen, T.;Egsgaard, H.; Larsen, E. Anal. Chim.Acta 1981, 124, 1-13.
( 3 ) Radzuik, B.; Thomassen, Y.; Van Loon, J. C.; Chau, Y. K. Anal. Chim. Acta 1979, 105, 255-262. (4) Harrison, R. M. J. Environ. Sci. Health, Part A 1976,11, 417-423.
(5) Chakraborti, D.; De Jonghe, W. R. A,; Van Mol, W. E.; Van Cleuvenbergen, R. J. A.; Adams, F. C. Anal. Chem. 1984, 56, 2692-2697. (6) Sirota, G. R.; Uthe, J. F. Anal. Chem. 1977,49,823-825. (7) Chau, Y. K.; Wong, P. T. S.; Bengert, G. A.; Dunn, J. L. Anal. Chem. 1984,56, 271-274. (8) Cruz, R. B.; Lorouso, C.; George, S.; Thomassen, Y.; Kin-
rade, J. D.; Butler, L. R. P.; Lye, J.; Van Loon, J. C.
Spectrochim. Acta, Part B 1980,35, 775-783. (9) Johnson, M. S.; Pluck, H.; Hutton, M.; Moore, G. Arch. Enuiron. Contam. Toxicol. 1982, 11, 761-767. (10) Nielsen, T.; Jensen, K. A.; Grandjean, P. Nature (London) 1978 274, 602-603. (11) Harrison, R. M.; Laxen, D. P. H. Enuiron. Sci. Technol. 1978,12, 1384-1391. (12) Grove, J. R. In Lead in the Marine Environment; Branica, M.; Konrad, Z., Eds.; Pergamon: Oxford, 1980; pp 45-52. (13) Noden, F. G. In Lead in the Marine Enuironment;Branica, M.; Konrad, Z., Eds.; Pergamon: Oxford, 1980; p p 83-91. (14) Jarvie, A. W. P.; Markall, R. N.; Potter, H. R. Environ. Res. 1981,25, 241-249. Environ. Sci. Technol., Vol. 20,No. 10, 1986
1037
Environ. Sci. Technol. 1986,2 0 , 1038-1043
(15) Hayakawa, K. Jpn. J. Hyg. 1972,26, 526-535. (16) Cremer, J. E.; Callaway, S. Br. J. Ind. Med. 1961, 18, 277-282. (17) Bolanowska, W. Br. J. Ind. Med. 1968, 25, 203-208. (18) Jaworski,J. F. “Effectsof Lead in the Environment-1978”; NRCC Publication No. 16736; 1979; pp 615-616. (19) Grandjean, P.; Nielsen, T. Residue Rev. 1979, 72, 97-154. (20) Grandjean, P. In Lead us Health; Rutter, M.; Jones, R. R., Eds.; Wiley: New York, 1983; p 183. (21) Wong, P. T. S.; Chau, Y. K.; Luxon, P. L. Nature (London) 1975, 253, 263-264. (22) Chau, Y. K.; Wong, P. T. S. ACS Symp. Ser. 1978, No. 82, 39-53. (23) Jarvie, A. W. P.; Marshall, R. N.; Potter, H. R. Nature (London) 1975,255, 217-218. (24) Reisinger, K.; Stoeppler, M.; Nurnberg, H. W. Nature (London)1981, 291, 228-230. (25) Schmidt,U.; Huber, F. Nature (London)1976,259,157-158. (26) Jarvie, A. W. P.; Whitmore, A. P.; Markall, R. N.; Potter, H.R. Enuiron. Pollut., Ser. B 1983, 6, 81-94. (27) Ahmad, I.; Chau, Y. K.; Wong, P. T. S.; Carty, A. J.; Taylor, L. Nature (London) 1980, 287, 716-717. (28) Jarvie, A. W. P.; Whitmore,A. P. Enuiron. Technol. Lett. 1981,2, 197-204. (29) Lovelock, J. E.; Maggs, R. J. Nature (London)1973,241, 194-196. (30) Mineau, P.; Fox, G. A,; Norstrom, R. J.; Weseloh, D. V.;
Hallett, D. J.; Ellenton, J. A. In Toxic Contarninants in the Great Lakes;Nriagu, J. 0.;Simmons,M. S., Eds.; Wiley: New York, 1984; pp 425-452.
(31) Gilberson, M. Nature Can. 1975, 4, 22-25. (32) Forsyth, D. S.; Marshall, W. D. Anal. Chem. 1983, 55, 2 132-2 137. (33) Forsyth, D. S.; Marshall, W. D. Anal. Chem. 1985, 57, 1299-1305. (34) Kovats, E. In Advances in Chromatography;Giddings, J. C.; Keller, R. A., Eds.; Marcel Dekker: New York, 1965; V O ~1, . 229-247. (35) Chau, Y. K.; Wong, P. T. S.; Saitoh, H. J. Chromatogr.Sci. 1976, 14, 162-164. (36) DuPuis, M. D.; Hill, H. H., Jr. Anal. Chem. 1979, 51, 292-295. (37) Estes, S. A.; Uden, P. C.; Barnes, R. M. J. Chromatogr. 1982, 239, 181-189. (38) Blais, J. S.; Marshall, W. D. J. Environ. Qual. 1986, 15, 255-260. (39) Hodson, P. V.; Whittle, D. M.; Wong, P. T. S.; Borgmann, U.; Thomas, R. L.; Chau, Y. K.; Nriagu, J. 0.;Hallett, D. J. In Toxic Contaminants in the Great Lakes; Nriagu, J. 0.;Simmons, M. S., Eds.; Wiley: New York, 1984; pp 335-369. (40) Casida,J. E.; Kimmel, E. C.; Holm, B.; Widmark, G. Acta. Chem. Scand. 1971,25, 1497-1499.
Received for review November 11,1985. Accepted May 16,1986. This study was financially supported by the Natural Sciences and Engineering Research Council of Canada (W.D.M.),Fonds F.C.A.C. pour l’aide et le soutien a la recherche, and J.W. McConnell Foundation (D.S.F.).
Vapor-Particle Partitioning of Semivolatile Organic Compounds: Estimates from Field Collections Terry F. Bidleman,” W. Net1 Billings, and Wllllam T. Foreman Department of Chemistry, Marine Science Program and Belle W. Baruch Institute for Marine Biology and Coastal Research, University of South Carolina, Columbia, South Carolina 29208
Pesticides, polychlorinated biphenyls (PCB), and other semivolatile organic compounds (SOC) exist in air as vapors and are associated with particulate matter. Factors influencing the vapor-to-particle distribution can be obtained from high-volume sampling experiments using a glass-fiber filter to collect particles and an adsorbent trap to collect vapors. Measurements of airborne organochlorine pesticides and PCB in four cities over a wide temperature range were used to estimate apparent partition coefficients A(TSP)/F, where A and F a r e the adsorbent- and filter-retained SOC concentrations (ng/m3) and TSP is the total suspended particle concentration (pg/m3). A(TSP)/F were related to the average sampling temperature ( T ,kelvin) through log [A(TSP)/F] = m / T 4- b. Fitted log [A(TSP)/F] a t 20 O C were closely correlated with log poL,the subcooled liquid vapor pressure. Heats of adsorption (AHA),calculated from slopes m, were a t most 2-4 kcal/mol greater than subcooled liquid heats of vaporization (AHvA)and in most cases indistinguishable from AHvaL a t the 95% confidence level. Introduction Semivolatile organic compounds (SOC) such as polychlorinated biphenyls (PCB), pesticides, heavy alkanes, and polycyclic aromatic hydrocarbons (PAH) exist in the atmosphere as vapors and associated with suspended particulate matter. The vapor-to-particle ratio (V/P) has an important influence on atmospheric removal by rain and dry deposition. Vapor scavenging is governed by Henry’s law constant, the ratio of vapor pressure to water 1038
Environ. Sci. Technol., Voi. 20, No. 10, 1986
solubility. Some lower molecular weight SOC are rained out by vapor dissolution in raindrops; examples include phenols ( I ) , two to four ring PAH (2-4), and hexachlorocyclohexane (HCH) (2-6). Other SOC that have substantial fractions in the particle phase and/or Henry’s law constants unfavorable for vapor scavenging are washed out of the atmosphere mainly on particles. Some compounds in the latter category are PCB and DDT (7,8),PAH of five or more rings ( 4 ) ,and n-alkanes (4). In the case of PCB and DDT, field-measured scavenging ratios greatly exceed those calculated from washout of the vapor phase (7). Within the suite of PCB congeners, those that are less volatile and have greater fractions on particles are preferentially deposited by rain and dry deposition (7, 8). Direct determination of V / P in ambient air is an experimentally challenging problem that has not yet been fully solved. SOC levels in ambient air are low, on the order of picograms to nanograms per cubic meter, and several hundreds or even thousands of cubic meters of air are usually sampled to provide enough material for analysis. A common high-volume (hi-vol) sampling technique is to pull air through a glass- or quartz-fiber filter (F)which retains particles followed by a solid adsorbent trap (A) to collect vapors. Frequently used adsorbents are polyurethane foam (PUF), Tenax-GC, XAD-2, and Florisil (9). The operational V/P is the ratio of the adsorbentretained/filter-retained concentrations (AIF), both expressed in ng/m3. How closely A/F represents VIP is uncertain. A/F may overestimate V I P because of the “blow-off effect”-
0013-936X/86/0920-1038$01.50/0
@ 1986 American Chemical Society
