Environ. Sci. Technol. 2005, 39, 9406-9411
Atmospheric PCB Concentrations at Terra Nova Bay, Antarctica A N D R E A G A M B A R O , * ,†,‡ LAURA MANODORI,† ROBERTA ZANGRANDO,‡ ALESSANDRA CINCINELLI,§ G A B R I E L E C A P O D A G L I O , †,‡ A N D P A O L O C E S C O N †,‡ Environmental Sciences Department, Ca’ Foscari University of Venice, 30123 Venice, Italy, Institute for the Dynamics of Environmental Processes, C.N.R., 30123 Venice, Italy, and Department of Chemistry, University of Florence, 50019 Florence, Italy
Concentrations of gas-phase polychlorobiphenyls (PCBs) were studied over an austral summer at a site in Terra Nova Bay, Antarctica. Gas-phase concentrations of individual PCB congeners in the atmosphere of Terra Nova Bay ranged from below the detection limit to 0.25 pg m-3, with a mean concentration of ∑PCB of 1.06 pg m-3. The PCB profile was dominated by lower-chlorinated PCB congeners; in fact >78% of the total PCB content was due to congeners with 1-4 chlorine atoms and only about 10% with 5-7 chlorines, whereas higher-chlorinated PCB congeners were below detection limits. The mean ∑PCB concentration obtained in this study were lower than those reported in previous Antarctic studies. Temporal concentration profiles of ∑PCB do not correspond to seasonal temperature changes. In consideration of the low PCB concentrations observed, the studies with the wind roses, the regression between ln P(PCB) and T-1, and the distribution of congeners, we can hypothesize that PCB local source contributions are not very important, whereas long-distance transport is the prevalent factor bringing PCBs to Terra Nova Bay.
Introduction Polychlorobiphenyls (PCBs) are a class of nonpolar semivolatile organic compounds which includes 209 congeners divided into 10 homologue classes. They are chemically very stable, persistent in the environment, and toxic with endocrine-disrupting properties, and they bio-accumulate in the food chain. For all these reasons, they are generally considered priority pollutants, thus making their monitoring in the environment and studies of their toxic effects on living organisms of prime importance. Long-range atmospheric transport can move PCBs away from source regions to more remote and pristine locations such as Antarctica, which has made PCBs ubiquitous in the global environment. Montone et al. (1) reported that comparable levels of contamination in seawater from north and south of the Antarctic Convergence indicated that the atmosphere, not the water, was the dominant pathway for the transport of PCB compounds to * Corresponding author fax: +39-41-2348549; e-mail: gambaro@ unive.it. † Ca’ Foscari University of Venice. ‡ Institute for the Dynamics of Environmental Processes. § University of Florence. 9406
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 24, 2005
the Antarctic. Further evidence of atmospheric transport of PCBs in the Antarctic environment was provided by Fuoco et al. (2) whose studies of Antarctic lake sediment found high contributions of atmospheric particulate matter as the primary vehicle of transport and diffusion of PCBs in the environment. In addition to atmospheric transport, local sources of PCBs, such as Antarctic Research Stations, should be considered (3). Finally, information on the impact of longrange transport and local sources can be obtained from an analysis of the PCB concentration in surface snow. The deposition and accumulation of PCBs in the Antarctic snow/ pack ice are, in fact, considered to be important steps in their transfer from the atmosphere to terrestrial and marine systems (4). In Antarctica, the presence of contaminants can threaten living resources since several persistent organic pollutants (POPs), including PCBs, accumulate in the tissues of organisms (4, 5). PCBs have also been detected in Arctic biota. In general, because they originated mainly in the Northern hemisphere, their levels were lower in the Antarctic biota than in Arctic biota. Data concerning concentrations of PCBs in Antarctic air are very scarce. They were mainly collected during cruises such as the Atlantic cruise from the United Kingdom to Halley, Antarctica (6), with the aim of establishing PCB oceanic background air concentrations and assessing their latitudinal distribution, and from a few sampling campaigns on the continent itself. Kallenborn et al. (7) measured PCBs in ambient air at Signy Island over a period of 17 weeks and found that mean concentrations for single congeners were comparable to those in Arctic air. Montone et al. (1) report that atmospheric levels of PCBs in the vicinity of the Brazilian Antarctic Research Station were generally low and the higher levels were associated with the passage of the frontal system coming from South Africa. Moreover Ockenden et al. (8) monitored the air for PCBs at two sites in the southern hemisphere, one over land and the other over water. They found that the highest concentrations were observed when temperatures were greater and the air concentrations were higher over water than over land. In this study we increase the limited atmospheric database on PCBs in remote areas by reporting PCB concentrations during the austral summer at a site in Terra Nova Bay, Antarctica, and we hypothesize their possible sources.
Experimental Section Sampling and Analysis. Air samples were collected at a coastal site off the Northern Victoria Land (Figure 1), about 3 km south of the Italian base of Terra Nova Bay (74° 42′ 56.3′′ S, 164° 06′ 52′′ E), during the austral summer between 4 November 2003 and 30 January 2004 (XIX Italian Antarctic Expedition). Sampling was performed over a 5-day period, providing a total of nine observations. Samples were obtained using high volume samplers (Tisch Environmental Inc., Cleves, OH) equipped with a quartz fiber filter (QFF; size 102 mm, SKC) and a polyurethane foam plug (PUF; height 75 mm, diameter 65 mm, SKC, Eighty Four, PA) to differentiate the particulate and the gas phases, respectively. Filters were not analyzed because no significant retention of PCB congeners was observed during our preliminary tests, as reported by other authors (1, 8). This is in contrast with the fact that the condensation and deposition of gas-phase samples at such low temperatures in Antarctica should be significant. Stripping processes from the filters, caused by the high operative flow (average: 0.34 m3 min-1) and the high volume of air 10.1021/es0510921 CCC: $30.25
2005 American Chemical Society Published on Web 11/04/2005
FIGURE 1. Map of sampling site.
TABLE 1. Averages of Meteorological Data during the Sampling Periods tempwind wind erature relative direction velocity (°C) humidity (%) (North) (m s-1) I II III IV V VI VII VIII IX
4-9 November -10.2 9-15 November -10.2 4-9 December -1.3 9-14 December -1.9 14-19 December -4.7 3-10 January -0.4 10-17 January -2.9 17-23 January -3.1 23-30 January -3.5
26.1 11.9 50.1 58.7 65.4 65.5 65.2 44.6 61.5
77.5 69.8 208.1 184.0 226.5 193.1 201.0 220.9 221.0
0.8 0.1 6.6 6.1 2.4 5.2 6.9 6.3 5.8
collected (about 2500 m3), could be one possible cause of this. Every month a calibration was carried out to check flow rates. Table 1 reports the average meteorological data (temperature, relative humidity, wind direction, and velocity) over every sampling period. PUF plugs were pre-extracted two times with n-pentane: dichloromethane (2:1, v/v) in a Soxhlet extractor for 24 h in Italy and two times at the Italian station in Antarctica before sampling. They were then stored in solvent-rinsed metal tins, firmly sealed, and stored at -20 °C until extraction. The analytical method employed has been described elsewhere (9). Briefly 100 µL of six 13C-labeled PCB (28, 52, 101, 156, 138, 180; Cambridge Isotope Laboratories, Andover, MA) solutions at 200 pg µL-1 concentration in n-hexane were spiked onto the samples immediately prior to extraction. The extraction of PUFs was performed in Soxhlet apparatus for 24 h using 495 mL of an n-pentane-dichloromethane mixture (2:1 v/v). Sample extracts were dried with anhydrous Na2SO4 (about 2 g) and their volumes were reduced to 5 mL under a gentle nitrogen flow. Cleanup was performed by adding the sample to a column, slurry packed at the top with activated Florisil and at the bottom with anhydrous Na2SO4, and eluting with 30 mL of n-hexane. The eluate was reduced to 100 µl under a gentle stream of nitrogen and analyzed by gas chromatography-high-resolution mass spectrometry. GC-MS Analysis. One MAT 95XP high-resolution magnetic mass spectrometer (ThermoFinnigan), equipped with a Hewlett-Packard Model 5890 series II gas chromatograph, was used to analyze all samples. Gas chromatographic separation was performed on a fused silica capillary column (J & W Scientific DB-5MS, 60 m × 0.250 mm × 0.25 µm) and data were acquired in the electron impact (EI) mode (45 eV).
The operating conditions for PCB analysis were as follows: injector temperature 300 °C; transfer line temperature 300 °C; oven temperature program 120 °C (1 min), 20 °C min-1 to 150 °C, 8 min at 150 °C, 4 °C min-1 to 235 °C, 10 min at 235°C, 12 °C min-1 to 290 °C, 34 min at 290 °C (postrun); carrier gas (helium), 1.2 mL min-1; injection mode, splitless (split valve open after 1 min) with purge flow 50 mL min-1. High-resolution data were acquired with R ) 10 500 using perfluorotributylamine for continuous calibration of the instrument. The multiple ion detection technique with seven time windows was used for monitoring the exact mass of the most intense ion in the molecular ion cluster (188.0393, 222.0003, 255.9614, 291.9194, 325.8804, 359.8415, 393.8025, 429.7606, 463.7216, 497.6826). Quantification was performed by comparing the area of the chromatographic peak of the PCB with those of the same homologue 13C-labeled PCB; results were corrected by periodically evaluating individual response factors. Quality Control. Quality control of the analytical procedure for PCB determination in Antarctic samples was a very critical step. Due to both very low PCB concentrations and the time lag between pre-cleaning of the PUFs in Italy, sampling in Terra Nova Bay, and the extraction of samples back in Italy, the risk of sample contamination was high. Procedural blanks were, therefore, evaluated by analyzing a large number of two types of blank samples. The first of these are the laboratory blanks (LB) where the Soxhlet apparatus was filled using 495 mL of an n-pentanedichloromethane mixture (2:1 v/v) and extracted for 24 h as though it were a sample. The extract went through cleanup and PCBs were determined as for samples. The second are the field blanks (FB) where the PUFs were pre-cleaned in Italy by two 24 h extractions and in Terra Nova Bay by two more 24 h extractions in Soxhlet apparatus using 495 mL of an n-pentane-dichloromethane mixture (2:1 v/v) as for samples. The procedural blanks were assembled in sampling modules and then transferred back to Italy in a metal can where they were analyzed as samples. The repeatability and the accuracy of the method were estimated from four consecutive measurements of spiked PUF plugs with known amounts (200 pg) of 57 PCBs and subsequent extraction and analysis using the procedure reported above. For the repeatability test, the relative standard deviation ranged from 2 to 20% and the results of the accuracy test showed that, for most of the PCBs, amounts were within (20%. These values are in agreement with those reported by Montone et al. (1), who found that the recovery of the internal standards, spiked prior to sample extraction, ranged from 50 to 120%. They are also in agreement with those found in our previous work (9) where the relative standard deviation ranged from 1 to15% for spiked PCB congeners (2000 pg) and the recoveries for most of the PCBs were g75% and none were
