Calibration of a Passive Sampler for Both Gaseous and Particulate

The PUF disk and the GFF were supported by backup plates (20 mm diameter) ... of 16 PAHs was diluted with n-hexane (PPH-10JM, Chem Service Inc., U.S.)...
0 downloads 0 Views 182KB Size
Download PDF
Environ. Sci. Technol. 2007, 41, 568-573

Calibration of a Passive Sampler for Both Gaseous and Particulate Phase Polycyclic Aromatic Hydrocarbons SHU TAO,* YANAN LIU, WEI XU, CHANG LANG, SHUZHEN LIU, HAN DOU, AND WENXIN LIU Laboratory for Earth Surface Processes, College of Environmental Sciences, Peking University, Beijing 100871, China

A novel passive air sampler was designed and tested that individually collects the gaseous and particulate phase polycyclic aromatic hydrocarbons (PAHs) in air. The sampler was calibrated against a conventional active sampler in an indoor environment. A PUF (polyurethane foam) disk and a piece of GFF (glass fiber filter) were installed in a sampling shelter for collecting gaseous and particulate phase PAHs, respectively. The passive samplers were deployed in seven indoor locations for 86 days. Six times during this period, 24-h conventional active sampling was conducted for calibration at an average interval of 17days. Principle component analysis showed that the measured congener profile compositions were totally different between the gaseous and particulate phase PAHs, but similar between the passive and the active samples. This suggested that gaseous and particulate phase PAHs were primarily trapped by the PUF disk and GFF, respectively. Linear relationships between the passively and the actively measured and log-transformed concentrations were derived for calibration of both gaseous and particulate phase PAHs. The uptake rates of the sampler were 0.10 ( 0.014 m3/d and 0.007 ( 0.001 m3/d for gaseous and particulate phase PAHs, respectively. The rates were significantly lower than those reported in the literature using similar PUF samplers, mainly because of the special design with limited air circulation.

Introduction It is well recognized that polycyclic aromatic hydrocarbons (PAHs) are one of the most concerning classes of persistent organic pollutants in China. It was estimated that the annual emission of PAHs in China exceeds 25,000 tons with 20% being composed of the higher molecular weight carcinogenic compounds (1). Consequently, high PAH levels are often detected in various environmental media, particularly in the air (2-4). For instance, dispersion modeling demonstrated that the mean concentrations of total PAHs in gaseous and particulate phases in the atmosphere of Tianjin, a typical northern Chinese city, are 16 and 42 ng/m3, respectively. Over 40% of the local population was exposed to levels of benzo[a]pyrene equivalent to a concentration that exceeds the national ambient air standard of 10 ng/m3 (5). The passive sampler is an attractive and powerful tool for regional-scale air quality investigations (6-8), and has also * Corresponding author phone and fax: 0086-10-62751938; email: [email protected]. 568

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 2, 2007

been widely used in a range of studies to monitor indoor and outdoor air qualities of various pollutants including semivolatile organic compounds (8, 9). Several types of sampling media including semipermeable membrane devices (SPMDs), solid-phase microextraction (SPME), polymer-coated glass (POG), styrene-divinylbenzene copolymer (XAD resin), and polyurethane foam disks (PUF) have been employed for monitoring persistent organic pollutants using a passive air sampler (6-8, 10, 11). It has been documented that, although passive samplers are desirable, there are a number of limitations associated with this technology that should be taken into consideration (12). One of the many constraints so far of passive samplers developed for persistent organic pollutants is that the gaseous and particulate phase pollutants cannot be distinguished. For example, Jaward and colleagues found that the measured gaseous and particulate phase PAHs and polychlorinated naphthalenes using PUF disks were in good agreement with the total air concentrations measured using a conventional active method (13). They also reported that a number of other persistent organic pollutants sampled with PUF disks were in good agreement with total air concentrations obtained by traditional active sampling (8). In many cases, toxic chemicals are strongly associated with particulate matter in the environment; therefore it is desirable to develop passive particle samplers for measuring the gaseous and particulate phase pollutants individually (13). In the case of PAHs, there are a large number of congeners with different physiochemical properties and environmental behaviors. For example, naphthalene and other lower molecular weight species exist primarily in the gaseous phase, while higher molecular weight compounds like benzo[a]pyrene are mostly associated with particulates in the environment (14, 15). A passive sampler, which can collect gaseous and particulate phase PAHs individually and simultaneously, would be a useful tool for regional studies. This is particularly true in the Chinese urban environment where extremely high aerosol levels are often observed (16). The objectives of this study were to design and test a passive air sampler that can individually sample gaseous and particulate phase PAHs. In order to present the measurements in volumetric air concentrations, it is necessary to know the sampling rate of the passive air sampler for particular chemicals. Therefore, the sampler was calibrated by comparing the results of conventional active sampling that was performed simultaneously. Sixteen USEPA priority PAHs, namely, naphthalene (NAP), acenaphthene (ACE), acenaphthylene (ACY), fluorene (FLO), phenanthrene (PHE), anthracene (ANT), fluoranthene (FLA), pyrene (PYR), benz[a]anthracene (BaA), chrysene (CHR), benzo[b]fluoranthene (BbF), benzo[k]fluoranthene (BkF), benzo[a]pyrene (BaP), indeno[1,2,3-cd]pyrene (IcdP), dibenz[a,h]anthracene (DahA), and benzo[g,h,i]perylene (BghiP) were included in this study.

Materials and Methods Sampling Media and Device. As illustrated in Figure 1, the passive sampler consisted of two pieces of sampling media including a PUF disk (100 mm diameter × 10 mm thick, density 0.024 g/cm3) and a glass fiber filter (GFF, 80 mm diameter) in a sampling shelter that collected for the gaseous and particulate phase PAHs, respectively. The shelter was a cylinder made of stainless steel. To eliminate gravitational deposition of particles, the PUF disk was mounted to the ceiling of the shelter with only the downward face exposed. The GFF was suspended in the chamber to trap the particulate matter. The PUF disk and the GFF were supported by backup 10.1021/es0617486 CCC: $37.00

 2007 American Chemical Society Published on Web 12/09/2006

FIGURE 1. Design of the passive air sampler with a PUF disk and a glass fiber filter (GFF) for individually collecting gaseous and particulate phase PAHs. plates (20 mm diameter) mounted on a central screw stem using nuts. The samplers were assembled and disassembled in the laboratory immediately before and after the deployment and transportion to and from the sites. The PUF disks were extracted by Soxhlet in a 1:1 mixture of n-hexane and cyclohexane for 4 h. The GFFs were preconditioned by heating in a furnace at 500 °C for 4 h. The shelter was air-sealed except for 12 evenly distributed 10 mm diameter openings located at the bottom for air exchange. The purposes of such a design were to prevent direct air flow through the shelter, minimize the turbulence in the device, reduce the possibility of aerosols being trapped by the PUF disk, and allow part of the aerosols to settle gravitationally on the GFF. The uptake rate of the sampler was reduced by such a design, which was acceptable since the sampler was specifically designed for long-term monitoring. In each sampler, there was also an XAD cartridge installed. The measured uptake of XAD was significantly correlated to PUF uptake at p < 0.01. The XAD results are not included in this paper because no extra information was provided except that a linear uptake kinetics was confirmed. Sampling. The passive air samplers were deployed at seven indoor locations including residence apartments (S1 and S2) and business offices (S3-S7) for 86 days from January 8 to April 3, 2006. The sites were selected to cover a relatively large range of PAH contamination levels. The space heating in the area was centralized and there was no indoor source from the heating at the sampling sites. Two passive air samplers were set at each site as replicates where six active samples were collected on days 1, 13, 28, 48, 69, and 86, respectively. For active sampling, assembled cartridges with glass PUF holders with metalline screens (Supelco) containing low volume PUF plugs from Supelco (22 mm diameter × 7.6 cm) along with GFFs (22 mm in diameter) were employed for the collection of gaseous and particulate phase PAHs. The sampling rate was set at 2 L/min using a TMP 1500 pump (Jiangsu Eltong Electric Corp. Co., Ltd.). All of the sampling media were removed and stored in sealed, solventcleaned containers at -20 °C until extraction. Extraction and Analysis. PUF disks and plugs were extracted by Soxhlet in a 1:1 mixture of n-hexane and cyclohexane for 4 h. GFFs were extracted by Soxhlet using the same solvent mixture for 10 h. The extracts were concentrated by rotary evaporation to about 1 mL. All samples were analyzed on a gas chromatograph (Agilent 6890) connected to a mass selective detector (Agilent 5973). A 30 m × 0.25 mm i.d. × 0.25 µm film thickness HP5MS capillary column was used with helium as the carrier gas at a constant flow rate of 1 mL/min. The column was

programmed from 60 to 280 °C at 6 °C/min, and then held isothermal for 20 min. The MSD was operated in electron impact mode at 70 eV, and the ion source temperature was 280 °C. The mass spectra were recorded using selected ion monitoring (SIM) mode. Quantification was performed by the internal standard method using 2-fluoro-1,1′-biphenyl and p-terphenyl-d14 (J&K Chemical, USA, 2.0 µg/mL). Quality Control. Routine quality assessment procedures were followed. All of the passive air sampling media were deployed and measured in duplicate to check for reproducibility. The average coefficients of variation of the duplicate samples were 22% for eight lower molecular weight PAHs from NAP to PYR sampled by the PUF disk, and 34% for eight median molecular weight PAHs from FLO to BkF sampled by the GFF. Two or three samplers, which were sealed and stored at -18 °C during the sampling period, served as procedure blanks for either passive or active sampling. The measured procedure blanks were generally more than 1 order of magnitude lower than the sample measurements for various congeners (Supporting Information). All of the results were blank corrected using the averages of all procedure blanks. The method detection limits were 0.85 (NAP) to 6.8 (BghiP) ng/mL for the extractants (Supporting Information). Method recoveries were determined by spiking the sampling media with a working standard. For the 16 spiked individual PAHs, the recoveries from NAP to BghiP were from 66% to 114% for PUF disk and from 60% to 115% for GFF (Supporting Information). Samples were spiked with a range of deuterated PAHs (NAP-d8, ANE-d10, ACE-d10, ANT-d10, CHR-d12, and Perelyne-d12) to monitor the extraction and cleanup procedures. Results showed good recoveries of the surrogates that ranged from 65% to 113% for the PUF disk and 89% to 112% for the GFF (Supporting Information). All the solvents used were analytical grade and were purified by distillation. The standard mixture of 16 PAHs was diluted with n-hexane (PPH-10JM, Chem Service Inc., U.S.). All glassware was cleaned using an ultrasonic cleaner (Kunshan KQ-500B) and heated to 400 °C for 6 h.

Results and Discussion Temporal and Spatial Variations. From a sampler calibration perspective, it was desirable that the PAHs at the seven sites varied over a wide concentration range. The total gaseous PAH concentrations of 16 congeners (PAH16) and the total gaseous concentrations without NAP (PAH15) measured using both the active and the passive samplers are presented in Figure 2 as mean values of all replicates. The lowest levels were recorded at site S1, (PAH15 were 58.7 ( 68.1 ng/m3 and 6.2 ( 3.0 ng/sampler‚day for active and passive samplers, respectively) and the highest PAH15 levels (gaseous phase) were detected at site S3 at 206 ( 122 ng/m3 and 19.7 ( 1.7 ng/sampler‚day for the active and passive samplers, respectively. It might be difficult to get results with larger range in an urban indoor environment of a single city. For particulate phase PAHs (not shown), the differences between the maximum and the minimum values were around 6 times for both the active (from 12.5 ( 67.9 to 76.7 ( 29.0 ng/m3) and passive (from 0.065 ( 0.017 to 0.38 ( 0.048 ng/sampler‚day) air sampling. According to the results of a one-way analysis of variance test, the differences among the sites were statistically significant for both gaseous and particulate phase PAHs (p < 0.001). Over the period studied, PAHs remained at relatively constant levels except for the second active sample when both the gaseous and particulate phase PAHs were much higher than those collected on other days (Figure 3). Although the study was conducted in winter when the space heating was on, the weather was basically under the control of cold high with strong winds from the northwest. In the week that the second sampling (d2) was conducted, however, the air VOL. 41, NO. 2, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

569

FIGURE 2. Measured PAH15 and PAH16 in the gaseous phase using both the active and passive air samplers with PUF at seven sampling sites.

FIGURE 3. Variations of actively collected gaseous and particulate phase PAH15 (left) and PAH16 (right) during the study period.

FIGURE 4. Profiles of actively and passively sampled gaseous (left) and particulate (right) phase PAHs. was particularly stagnant leading to the accumulation of large amounts of air pollutants. It was reported that PAH concentrations in the ambient air in the same area were under strong influence of temperature, humidity, and atmospheric stability (17) and there was a significant correlation in PAH level in air between indoor and outdoor environments (18). PAH Profiles of the Actively and Passively Collected Samples. The sixteen PAH congeners behaved differently in the environment due to differences in their physiochemical properties. The lower molecular weight species primarily occurred in the gaseous phase, while higher molecular weight compounds were predominately sorbed on airborne particles (14, 15). Consequently, the PAH congener profiles of gas and particulate phases in the atmosphere were considerably dissimilar. Such differences were addressed in this study to assess whether the gaseous phase, the particulate phase, or both were trapped by the PUF disk or the GFF in the passive air sampler. The gaseous and particulate phase PAH profiles, either passively or actively sampled as mean values of seven sites are depicted in Figure 4 for comparison. The measured PAH patterns of both the actively and passively collected samples generally gave good agreement with those reported in the literature. The similarities between the passive and active air samples as well as the dissimilarities between the gaseous and particulate phases are well demonstrated. The only difference between the passively and actively sampled particulate PAH profiles was the higher molecular weight species from BkF to BghiP, which were much higher in the active air samples than in the passive air samples. This can be explained by the low sampling efficiencies of passive air sampling for particulate phase PAHs, and the measured values of these congeners in many samples were below or close to the detection limits. This caused an underestimation 570

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 2, 2007

of the mean values when zeros were assigned to the undetectable PAHs. Similarity in PAH Patterns Among the Samples. The aforementioned profiles were further analyzed and compared for seven individual sites instead of the site means using principal component analysis (PCA). PCA was also employed by Wilford et al. to compare the relative congener composition of polybrominated diphenyl ether flame retardants in indoor and outdoor air samples. It was found that the technical formulations are relatively enriched in the heavier congeners, while indoor air was enriched in the more volatile constituents (9). In this study, a total of 28 samples including seven active and seven passive samples of both gaseous and particulate phases were employed in the principal component analysis. Six active measurements of each site were averaged. NAP was not included due to the relatively large uncertainty in the measurements of this volatile compound. The relative concentrations were used in the analysis. The total number of nondetectable samples was 58 accounting for 14% of the total sample size of 420. Of the 58, 19 were DahA. Therefore, the measurements of DahA were not included in the analysis and the undetectable data in the data set used for the analysis were less than 10%. The factor loadings of the first three principal components (F1, F2, and F3), which accounted for 86% of the total variance, are listed in Table 1. Those with loadings greater than 0.60 are marked bold. The first principal component accounted for over 34% of the total variance, and was mainly an association of the positive contributions of median molecular weight PAHs from PHE to CHR. F2 represents principally a linear combination of higher molecular weight PAHs including BbF, BkF, BaP, and BghiP. F3 was primarily negative contributions of lower molecular weight compounds such as ACE and FLO. ANT

TABLE 1. Factor Loadings of the First (F1), Second (F2), and Third (F3) Principal Components F1 F2 F3

ACE

ACY

FLO

PHE

ANT

FLA

PYR

BaA

CHR

BbF

BkF

BaP

IcdP

BghiP

var.,%

-0.41 -0.33 -0.69

-0.40 -0.68 -0.22

-0.01 -0.59 -0.60

0.83 -0.36 -0.10

0.27 -0.03 0.80

0.90 0.16 0.27

0.87 0.34 0.24

0.62 0.56 0.26

0.85 0.28 0.26

0.59 0.60 0.20

0.18 0.92 0.07

0.06 0.94 0.13

0.65 0.50 -0.09

0.17 0.94 0.11

34 33 13

FIGURE 5. Scatter plot of the factor scores of the first two principal components (F1 and F2) of PAH profiles measured using both the active (Gactive and Pactive for gaseous and particulate phases, respectively) and the passive (Gpassive and Ppassive for gaseous and particulate phases, respectively) samplers. also played an important role in F3. To illustrate the similarity or dissimilarity of the gaseous and the particulate phase PAHs collected either actively or passively, the factor scores of the first two principal components (F1 and F2) are plotted in Figure 5. In general, PAH compositions of the seven samples in the same class (passively or actively sampled gaseous or particulate phase PAHs) were similar to one another. This was suggested by the clustering of data points of the same color, even though the levels of the seven samples collected were significantly different (Figure 2). This is particularly true for both the actively and passively sampled gaseous phase PAHs and passively sampled particulate phase PAHs. The scatter plot also shows that the factor scores of gaseous phase PAHs (pink and orange dots) were distinguished from those of particulate phase PAHs (blue and green dots) in the F1-F2 plan. F1 values of all gaseous PAHs, both passively and actively sampled, varied in a narrow range from -1.2 to 0.10. This indicated a strong influence of lower molecular weight compounds with negative factor loadings and weak effects of median to higher molecular weight congeners. F2 values of gaseous PAHs were close to one another, ranging from -0.9 to -0.1, suggesting similar strong effects of ACE, ACY, and FLO. Compared with the active sample, the passively sampled gaseous phase PAHs shifted slightly toward the positive F1 and negative F2. When the PUF disks were suspended in the shelter with both sides exposed to a stronger air circulation, it was found that the PUF disks were effective at trapping gaseous chemicals as well as fine particles. The PAHs trapped by these PUF disks were representative of the derived total air concentrations of both gaseous and particulate phase PAHs (8, 9). In this study, however, when PUF disks were mounted to the ceilings of the chambers with only bottom openings of the shelter to diminish air circulation, it appeared that the influence of the particulate phase PAHs in the air was largely eliminated. However, the trapping of very fine particles on the PUF disks could not be completely avoided, and the influence of the very fine particles on the PAH profiles may explain the slight difference between the passively and actively sampled gaseous phase PAHs.

FIGURE 6. Relationship of the log-transformed PAH8L between the passive and the active PUF disks. The results of all seven sampling sites are applied. For PAH profiles of the particulate samples, the differences between the passively and actively collected samples were much larger with F1 and F2 varying from -0.8 to 2.1 and from -1.4 to 2.8, respectively. Dots representing the passively sampled particulate phase PAHs were clustered around a narrow range at the bottom right-hand corner of the plot and differed significantly with the gaseous samples, primarily on F1. This implied a positive contribution of median molecular weight PAHs from PHE to CHR (also refers to Figure 4, right). The difference between the actively and passively collected particulate phase PAHs was associated with both F1 and F2, which are primarily linear combinations of median and higher molecular weight congeners, respectively. In fact, due to very low sampling efficiency of the passive sampler, higher molecular weight compounds in many of the seven passive samples were either very close to or even below the detection limits. This resulted in a strong deviation from the active samples along the F2 axis. In summary, the results demonstrated that the passive and active samplings gave comparable congener profiles suggesting that the gaseous and particulate phase PAHs were sampled separately by the passive samplers. Calibration of the Passive Air Sampler for Gaseous Phase PAHs. Since the gaseous and particulate phase PAHs were passively sampled separately in this study, the sampling media were also calibrated individually for each of the two phases. Using log-transformed concentrations, the relationship between the actively and the passively measured gaseous phase PAHs was examined. For the higher molecular weight PAHs, the measurements were either very close to or even below the detection limits. Therefore, they were not included in the analysis. For most of the lower molecular weight species, significant positive correlations were revealed at a level of 0.05. Such correlations are illustrated in Figure 6 using PAH8L (total concentration of eight lower molecular weight PAHs from ACE to BaA) as a typical example. The linear regression equation based on the logtransformed PAH8L accounted for 88% of the variation. Taking into consideration the fact that the active samples were only collected for six individual days during which the VOL. 41, NO. 2, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

571

FLO to CHR individually with correlation coefficients varying from 0.65 to 0.94, all significant at a level of 0.05. The three lower molecular weight PAHs (NAP, ACE, and ACY) and five higher molecular weight PAHs (BkF, BaP, IcdP, DahA and BghiP) were not included in the plot because the measured concentrations in the passive samples were low, very close to, or even below the detection limits. Like gaseous phase PAHs, the passive air sampler was also calibrated against the active one for a total of eight particulate phase PAHs (PAH8Mp). The calibration equation for particulate PAH8M is as follows:

PAH8Mp (ng/m3) ) 100.992 lg P(ng/sampler·day) + 2.175

FIGURE 7. Relationship of the log-transformed PAH8M between the passive and the active GFFs. The results of all seven sampling sites are applied. PAH concentrations varied significantly (Figure 3), while the passive samplers were continuously exposed for 86 days during the study period, such a result was to be expected. Since there was no direct airflow through the shelter, it was assumed that the sampling rate was restricted by the limited air exchange in the sampler housing and mostly under the control of molecular diffusion. If this was the case, the sampling rates of individual PAH congeners should be negatively proportional to the square roots of their molecular sizes (19). However, such a relationship was not observed in this study and there was not a significant difference in the ratio of the passively to actively sampled quantities among the individual PAH congeners. This could possibly be explained by the following reason: the molecular sizes of the PAH congeners from NAP to PYR are not remarkably different (only from 128 to 202 with the difference of 1.26 times in square roots). Without a significant difference among the congeners, the passive sampler was calibrated against the active sampler using the PAH8L data, rather than that of individual PAH. Since the linear correlation was significant on a log-scale, the least-squares fit calibration equation for calculating the concentration of gaseous PAH8L (PAH8Lg) in terms of ng/m3 based on the passively sampled results was as follows:

PAH8Lg (ng/m3) ) 101.02 lg P(ng/sampler·day) + 0.969 where P was the passively measured PAH8L in ng/sampler‚ day. For each site, a ratio between the passively and actively measured PAH8L values was calculated (ng/sampler‚day)/ (ng/m3) and the mean value of the ratios of the seven sites was reported as the efficiency of the passive air sampler for the eight gaseous PAHs in terms of volume of air sampled per day. The calculated result was 0.10 ( 0.014 m3/d and was much lower than those values reported in the literature (3-4 m3/d) using a PUF disk sampler of a different design (9). Such a relatively low uptake rate was more or less expected because air circulation in the sampling shelter and the trapping of aerosols by the PUF disks in this study were limited. Calibration of Passively Sampled Particulate Phase PAHs. Figure 7 illustrates the linear relationship between the passively and actively measured particulate PAH8M (logtransformed), which is a total concentration of median molecular weight PAHs with 3-4 rings from FLO to BbF. The coefficient of determination was as high as 0.894. In fact, significant correlations were identified for seven species from 572

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 2, 2007

where P is the passively measured PAH8M concentration in ng/sampler‚day. The uptake efficiency of PAH8M was 0.008 ( 0.002 m3/d, which was over an order of magnitude below that of gaseous PAH8L. Such a low uptake rate was likely caused by the narrow and zigzag path from bottom opening to the surface of GFF. Since PAHs in the particulate phase were trapped gravitationally in the sampler, the sampling rate for particulate phase PAHs may be under strong influence of level and size distribution of airborne particles in the air and further studies on the factors affecting the uptake efficiency and modification of the device are desirable. Even when gaseous and particulate phase PAHs are pooled together, the uptake rate (0.11 m3/d) was still significantly lower than that of other passive PUF sampling devices reported in the literature, primarily due to the restriction of air exchange in the sampler housing (9, 20). At such a low uptake efficiency, the sampler operated most likely in kinetic mode, which is favorable for relatively long-term sampling. However, our samplers were calibrated in an indoor environment with limited samples and a relatively narrow range of exposure concentrations at relatively calm conditions. It is necessary to further test the device in the field where exposure concentrations vary in orders of magnitude and wind and temperature may influence its performance. Additional studies are also needed to investigate the size distribution of the particulate matter collected by the GFF against that in the ambient environment. One of the disadvantages of the passive sampler tested in this study is that the sampler has to be transported with great care to prevent the loss of particulate matter collected on the GFF. Hopefully, the sampler can be further improved to solve this problem.

Acknowledgments Funding for this study was provided by National Scientific Foundation of China (40332015/ 40021101) and National Basic Research Program (2007CB407301). We thank Ms. Susie Genualdi for polishing the English of the manuscript.

Supporting Information Available Tabulation of procedure blanks, recoveries, and detection limits, and results of both active and passive sampling. This material is available free of charge via the Internet at http:// pubs.acs.org.

Literature Cited (1) Xu, S. S.; Liu, W. X.; Tao, S. Emission of polycyclic aromatic hydrocarbons in China. Environ. Sci. Technol. 2006, 40, 702708, and personal communication. (2) Liu, Y. J.; Zhu, L. Z.; Shen, X. Y. Polycyclic aromatic hydrocarbons (PAHs) in indoor and outdoor air of Hangzhou, China. Environ. Sci. Technol. 2001, 35, 840-844. (3) Mai, B. X.; Qi, S. H.; Zeng, E. Y.; Yang, Q. S.; Zhang, G.; Fu, J. M.; Sheng, G. Y.; Peng, P. G.; Wang, Z. S. Distribution of polycyclic aromatic hydrocarbons in the coastal region off Macao, China:

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

assessment of input sources and transport pathways using compositional analysis. Environ. Sci. Technol. 2003, 37, 48554863. Chen, S. J.; Luo, X. J.; Mai, B. X.; Sheng, G. Y.; Fu, J. M.; Zeng, E. Y. Distribution and mass inventories of polycyclic aromatic hydrocarbons and organochlorine pesticides in sediments of the Pearl River Estuary and the northern South China Sea. Environ. Sci. Technol. 2006, 40, 709-714. Tao, S.; Li, X. R.; Yang, Y.; Lu, X. X.; Chen, H. T.; Shen, W. R. Dispersion modeling of polycyclic aromatic hydrocarbons from various sources in Tianjin, China. Environ. Sci. Technol. 2006, 40, 4586-4591. Wania, F.; Shen, L.; Lei, Y. D.; Teixeira, C.; Muir, D. C. G. Development and calibration of a resin-based passive sampling system for monitoring persistent organic pollutants in the atmosphere. Environ. Sci. Technol. 2003, 37, 1352-1359. Farrar, N. J.; Harner, T.; Shoeib, M.; Sweetman, A.; Jones, K. C. Field deployment of thin film passive air samplers for persistent organic pollutants: a study in the urban atmospheric boundary layer. Environ. Sci. Technol. 2005, 39, 42-28. Jaward, F. M.; Zhang, G., Nam, J. J.; Sweetman, A. J.; Obbard, J. P.; Kobara, Y.; Jones, K. C. Passive air sampling of polychlorinated biphenyls, organochlorine compounds, and polybrominated diphenyl ethers across Asia. Environ. Sci. Technol. 2005, 39, 8638-8645. Wilford, B. H.; Harner, T.; Zhu, J.; Shoeib, M.; Jones, K. C. Passive sampling survey of polybrominated diphenyl ether flame retardants in indoor and outdoor air in Ottawa, Canada: implications for sources and exposure. Environ. Sci. Technol. 2004, 38, 5312-5318. Harner, T.; Farrar N. J.; Shoeib, M.; Jones, K. C.; Gobas, F. A. P. C. Characterization of polymer-coated glass as a passive air sampler for persistent organic pollutants. Environ. Sci. Technol. 2003, 37, 2486-2493. Bartkow, M. E.; Huckins, J. N.; Mu ¨ ller, J. F. Field-based evaluation of semipermeable membrane devices (SPMDs) as passive air

(12) (13) (14) (15) (16)

(17) (18)

(19) (20)

samplers of polyaromatic hydrocarbons (PAHs), Atmos. Environ. 2004, 38, 5983-5990. Krupa, S. V.; Legge, A. H. Passive sampling of ambient, gaseous air pollutants: an assessment from an ecological perspective. Environ. Pollut. 2000, 107, 31-45. Jaward, F. M.; Farrar, N. J.; Harner, T.; Sweetman, A.; Jones, K. C. Passive air sampling of PAHs and PCNs across Europe. Environ. Toxicol. Chem. 2004, 23, 1355-1364. Harner, T.; Bidleman, T. F. Measurement of octanol-air partition coefficient for polycyclic aromatic hydrocarbons and polychlorinated naphthalenes. J. Chem. Eng. Data 1998, 43, 40-46. Kaupp, H.; McLachlan, M. S. Gas/particle partitioning of PCDD/ Fs, PCBs, PCNs and PAHs. Chemosphere 1999, 38, 3411-3421. Wang, W.; Tang, D. G.; Liu, H. J.; Wang, M. Y.; Du, J. Aircraft measurement of atmospheric pollutants in winter in north China, II: Research on pollution properties of atmospheric aerosol. Res. Environ. Sci. 2000, 13, 10-13. Liu, Y. N.; Tao, S.; Dou, H.; Zhang, T. W.; Zhang, X. L.; Dawson, R. Exposure of traffic police to Polycyclic aromatic hydrocarbons in Beijing, China. Chemosphere, in press. Lang, C.; Liu, Y. N.; Wu, S. P.; Zuo, Q.; Lan, T.; Wang, X. J.; Tao, S. Gas phase PAHs in indoor air from Peking University campus in non-heating season. Acta Sci. Circumstantiae 2004, 24, 655660. Thibodeaux, L. J. Environmental Chemodynamics: Movement of Chemicals in Air, Water, and Soil, 2nd ed.; Wiley: New York, 1996. Shoeib, M.; Harner, T. Characterization and comparison of three passive air samplers for persistent organic pollutants. Environ. Sci. Technol. 2002, 36, 4142-4151.

Received for review July 21, 2006. Revised manuscript received October 15, 2006. Accepted October 19, 2006. ES0617486

VOL. 41, NO. 2, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

573