Determination of airborne PAHs using passive sampling with 2,6 diphenyl-p-phenylenoxide as adsorbant

sampling rates were calculated by cross comparison with simultaneous active sampler data. Higher sampling rates were observed for low-molecular-weight PAHs, upon prolonged sampling periods at outdoor ambient air conditions. Although only moderate sampling rates were observed when coupled with a highly sensitive analysis technique based on thermal desorption gas chromatography and mass spectrometry, the passive sampler was shown to be competitive with the other passive samplers in literature. These results are promising and support the future use of this sampler for more quantitative analysis of gaseous and particle-bound PAHs.


INTRODUCTION
Air pollution caused by polycyclic aromatic hydrocarbons (PAHs) is an issue receiving continuous attention in many of PAHs have been reported in various occupational environments (Branisteanu & Aiking, 1998;Jongeneelen, 2001). Special attention has been allocated to analysing PAHs and associated compounds from incomplete combustion processes, which are believed to play a role in the formation of reactive oxygen species (ROS) and are also considered as mutagenic and carcinogenic compounds (Bowen, 2003;Molto et al., 2009). Some selected physico-chemical characteristics of PAHs are denoted in Table I of the Appendix. Considering the negative impact on human health, initiatives for further investigation and better screening methods to monitor air pollution are required. Most of the current PAH sampling methods are labour-intensive and involve pump-driven techniques. Active sampling is accurate and highly quantitative, but complex in design, expensive and requires maintenance. In contrast, passive samplers are cheap, unobtrusive and have a high spatial resolution (Bohlin, 2010). Consequently, passive sampling has emerged as an innovative technique to overcome today's sampling and analytical shortcomings. for gases was introduced by Palmes and Gunnison (1973) in a form of a mathematical model. Later on, many different types of passive air samplers (PASs) were introduced for sampling of different compounds and the theory has been well established for gaseous compounds (Tompkins & Goldsmith, 1977;Huckins et al., 1990).
Most of the existing PASs are designed for gas sampling of semi-volatile organic compounds, and based on high capacity sampling against a linear sampling rate for long durations such as weeks or months. Polyurethane (SPMDs), XAD-resin based samplers and membrane samplers are such examples (Petty et al., 1993;Wania et al., 2003;Harner et al., 2013). In addition, a few low capacity samplers such as polymer coated glass (POG) (Harner et al., 2003) and solid phase micro extraction (SPME) samplers have also been reported (Bohlin, 2010).

March 2017
Journal of the National Science Foundation of Sri Lanka 45(1) PAHs are known to partition between the gas and particulate phase, with heavier PAHs preferring the particle phase. Eighty four percent of the total PAHs et al., 2011). Many existing PASs are designed to sample volatile species by gas diffusion. Hence, the idea of simultaneous sampling of the gas and particle phase has investigated. Nash and Leith (2010) introduced a small axial PAS for particulate aerosols including a collection substrate underneath a metal mesh cap. They accordingly for gaseous and particle-bound PAHs based on both gravity and diffusion.
Since deposition of PAHs is involved in passive sampling, the deposition velocities are equally important. deposition velocity for both particles and gases as the of the pollutant. For PAHs ranging between napthalene and perylene, the mean particle deposition velocity for (Horstmann & Mclachlan, 1998). semi-volatile compounds from air, and can be utilised both as a column packing material and a trapping agent for organic compounds. Thermal desorption of volatile content is possible due to its high thermal stability and 2 /g, pore volume is 2.4 cm 3 /g and the 1978; Clark et al., 1982;Alfeeli et al., 2011).
In this study, we have performed initial tests involving the sensitive analysis of TD-GC-MS in an urban roadside METHODOLOGY known as Tenax ® TA was purchased from Novakemi Ltd, Holland. After cleaning with n-hexane in acetone with dichloromethane. Immediately after that, granular on top. The sorbent was only partly dissolved resulting in semi-spherical granules on the surface. A metallic mesh used so that the sampler could collect gaseous PAHs Previously determined retention times from external calibration curves using standard reference material (SRM-1491) were used to identify different PAHs from the sampler, and the peak areas were converted were prepared using the above mentioned method and analysed by the same TD-GC-MS based method. For a relative retention time ± 2 seconds, as compared to the standard solution, were used.

Field evaluation of the samplers
active samples were obtained during the week by pumping passive samplers were allowed to continuously sample for seven days at the site. All samplers were placed inside a weather protective hood in order to secure the samplers from rain and strong winds. Due to limitations of pump runtime, the active sampling durations were not identical to those of passive sampling, and average concentrations were used to obtain an integrated value for the week. The sampling height was approximately 10 ft from the ground level.

RESULTS AND DISCUSSION
and Harner (2002) and the SPMD sampler designed by Liua et al. 2 and 370 cm 2 , respectively. The latter had incorporated a limit of detection (LOD) of 1 ng. A POG sampler with a surface area of 300 cm 2 was developed by Harner et al. (2013), where the LOD was 0.1 ng. In comparison, the PAS of this study has a smaller surface area of 1.27 cm 2 and an associated LOD of 0.1 -0.05 ng. In our case, majority The mean concentrations of PAHs (ng/m 3 ) from active samplers are found in Table 1. The most volatile species 32 ng/m 3 at the site. Detection criteria were also met for 2 and 3-ring PAHs for all active samplers, whereas 4 and 5-ring PAHs were below the LOD. Although demonstrating low that would be expected in an urban location in Europe.  et al., 1997). This comparison demonstrates that the BaP 3 found in this study location was low, but was in the same order of the magnitude to Scandinavian countries. Passive R S 3 /day for more volatile PAHs [relative standard deviation (RSD) indicating successful uptake by diffusion. Also for heavier PAHs useful R S were obtained, although demonstrating a 3 /day and scatter (mean RSD PAH compounds in the sampler as gaseous and particlebound would be impossible in practice. Therefore, it was assumed that PAHs behave according to their physicochemical characteristics. PAHs can be released in gas phase as well as supported onto particulate matter (PM) due to their high volatility (Mastral & Calleän, 2000). Most volatile PAH compounds with two or three aromatic rings have been found to be released mainly in gaseous form and the ones with three or more aromatic rings, associated with the PM (Lee et al., 1993). For instance, naphthalene is reported to be present entirely in the gas are known to be adsorbed to PM (Mastral & Calleän, 2000).
Previously published R S are in the order of one magnitude higher and they are also shown in Table 1 for comparison. However, those samplers are based on analytical methods demonstrating a higher LOD by one order of magnitude, thus resulting in a similar overall performance (Shoeib & Harner, 2002; Bohlin, 2010; Gouin et al., 2010).

CONCLUSION
promising sampling rates to measure the full suite of ambient air PAHs in a short-term urban air sampling campaign. The lower sampling rates exhibited by the passive samplers is compensated by the superior mass transfer by the thermal desorption method used in the collection substrate indicated simultaneous sampling of both particulate and gaseous phases, which is considered a great advantage due to the particle association of heavier PAHs.

Acknowledgement
Support provided by L. Hagglund and C. Lejon of the