- Email: [email protected]
Optimisation steps of an innovative air sampling method for semi volatile organic compounds
Atmospheric Environment 79 (2013) 780e786
Contents lists available at ScienceDirect
Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv
Optimisation steps of an innovative air sampling method for semi volatile organic compounds Borislav Lazarov a, *, Rudi Swinnen a, Maarten Spruyt a, Eddy Goelen a, Marianne Stranger a, Gilbert Desmet b, Eric Wauters b a b
Environmental Risk and Health Unit, VITO, Boeretang 200, 2400 Mol, Belgium Flemish Environmental Agency (VMM), Raymonde de Larochelaan 1, B-9051 Sint-Denijs-Westrem, Belgium
h i g h l i g h t s An innovative method for the measurement several groups of SVOCs in air has been optimised. As the method collects gaseous as well as particulate matter SVOCs, it avoids underestimating the total air concentration. The applicability of the method to sample PAHs, PCBs and PEs from different urban environments was confirmed. The presented strategy reduces the risk of contamination during sample preparation steps compared to more traditional techniques.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 26 April 2013 Received in revised form 19 July 2013 Accepted 24 July 2013
This work describes optimisation steps of an innovative method for the measurement several groups of semi-volatile organic compounds (SVOCs) in air, collecting both gaseous and particulate air fractions. It is based on active air sampling on sorption tubes (consisting of polydimethylsiloxane (PDMS) and Tenax TA), followed by thermal desorption and gas chromatography mass spectrometry analysis (TDeGCeMS). The optimised method was validated in the laboratory for the measurement of selected target compounds from the following chemical classes: polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs) and phthalate esters (PEs). It was applied in different Belgian urban outdoor as well as indoor environments. The new method is characterised by limits of detection in the range of 0.003e0.3 ng m3 for PAHs, 0.004e0.2 ng m3 for PCBs, 0.113 e0.201 ng m3 for PBDEs and 0.002e0.2 ng m3 for PEs, a linearity of 0.996 and a repeatability of less than 10% for all studied compounds. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Air monitoring Semi-volatile organic compounds Thermal desorption GCeMS Mixed bed sampling
1. Introduction According to the definition of the World Health Organisation (WHO), semi-volatile organic compounds (SVOCs) are organic chemicals with boiling points ranging from 240e260 C to 380e 400 C (World Health Organization, 1989). This range covers a large number of compounds that are present in outdoor and indoor air (e.g. polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs), phthalate esters (PEs). These classes of compounds may occur in meaningful abundances both in the gas phase and on the surface of airborne particles, which recognises them as a human health concern. Although in recent years, the study of the occurrence, fate and
* Corresponding author. Tel.: þ32 14 33 69 27; fax: þ32 14 33 55 99. E-mail addresses: [email protected], [email protected] (B. Lazarov). 1352-2310/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atmosenv.2013.07.059
human exposure to these compounds has become an important research topic, SVOCs have not been so widely studied as certain other classes of air pollutants such as VOCs, airborne particles and inorganic gaseous pollutants (Weschler and Nazaroff, 2008). The high degree of analytical challenges in measuring SVOCs has impeded progress in studying them. Sampling of SVOCs is mostly done by the collection of both the gas and the particle phase, using high volume samplers. In most cases the particulate phase is trapped on a filter (quartz or glass fibre) and the breakthrough of the gas phase is subsequently trapped on a sorbent (PUF or XAD) (EN ISO 16000e13, 2008; Batterman et al., 2009; Elflein et al., 2003). A widely used method for sample treatment after collecting several m3 of air, is solvent desorption of the filters (Soxhlet extraction, accelerated solvent extraction, sonication) followed by analysing the compounds of interest by GCeMS (gas chromatography coupled to mass chromatography) or HPLC techniques. There are several limitations and
B. Lazarov et al. / Atmospheric Environment 79 (2013) 780e786
disadvantages related to this traditional method for sampling and detection of SVOCs. These include the lower sampling efficiency of compounds with higher vapour pressure (Bidleman et al., 1986) and the underestimation of the actual content caused by filter reactions with oxidative gases (Schauer et al., 2003). The method also implies the use of high-volume sampling pumps, which are not suitable for personal and indoor air sampling (due to dimensions and noise nuisance) and need electricity power supply, which is not always convenient when sampling outdoors. Another disadvantage is that the analysis method is time-consuming and requires the samples to be manipulated, which means that there is considerable risk of contamination and/or losing some of the compounds. Several new methods, based on diffusion denuders (Temime et al., 2002), passive samplers (Wania et al., 2003; Namiesnik et al., 2005; Ni et al., 2007; May et al., 2011), sorbentimpregnated filters (Galarneau et al., 2006) or molecular imprinted polymers (Krupadam et al., 2010) have been developed as alternatives for the high-volume sampling method. These methods have advantages compared to the high-volume sampling method, such as more efficient sampling of the compounds with higher vapour pressure (diffusion denuders), elimination of pumping unit (passive samplers) and reduction in size, simplified sample handling and decrease in solvent consumption (sorbent-impregnated filters, molecular imprinted polymers). However, those methods have other disadvantages such as long sampling time (denuders and passive samplers), and possible losses due to photo degradation reactions (passive samplers) that are not excluded as well. As an alternative for these methods, Wauters et al. (2008) have introduced another method for PAHs sampling, based on active sampling on sorption tubes consisting of polydimethylsiloxane (PDMS) foam, PDMS particles and a Tenax TA bed, followed by thermal desorption and analysis by GCeMS. The method is characterised by the advantages to only require low flow sampling (using quasi noiseless and portable sampling pumps), and to considerably reduce the risk of contamination during the sample preparation and analysis steps. The main objective of this reported study is to explore the possibilities of using the mixed bed sampling method to sample and analyse different classes of SVOCs, by performing several optimisation steps. The optimised method was validated for measurement of the listed target compounds (see Table 1) in different Belgian urban outdoor as well as indoor environments. 2. Material and methods High purity standards (98.0e99.9%) were used in this study. The standard calibration mix of PAHs in methylene chloride with a concentration 2000 mg mL1 was purchased from Restek, USA. Deuterated PAHs (d8-naphthalene, d10-fluorene, d10fluoranthene, d12-benzo(a)pyrene and d12-benzo(g,h,i)perylene) and PCBs were purchased from Promochem/C.N. Schmidt, The Netherlands. PBDEs standards were purchased from Wellington Laboratories, Canada. All standard solutions were prepared by dilution in methanol. The solvents were GC grade with purity >99.9% (Merck, Germany). The standard reference material of PAHs (ERM-AC213) was purchased from Institute for Reference Materials and Measurements (IRMM, Belgium). Nitrogen gas of 99.999% purity was used for spiking the thermal desorption sorbent tubes and 99.999% pure helium gas was used for chromatographic analysis.
781
Table 1 Target compounds and respective retention times and the quantification ion studied in air. Compound
RT, min
Q, m/z
PAHs Naphthalene (cas no 91-20-3) Acenaphthylene (cas no 208-96-8) Acenaphthene (cas no 83-32-9) Fluorene (cas no 86-73-7) Phenanthrene (cas no 85-01-8) Anthracene (cas no 120-12-7) Fluoranthene (cas no 206-44-0) Pyrene (cas no 129-00-0) Benz(a)anthracene (cas no 56-55-3) Chrysene (cas no 218-01-9) Benzo(b)fluoranthene (cas no 205-99-2) Benzo(k)fluoranthene (cas no 207-08-9) Benzo(a)pyrene (cas no 50-32-8) Dibenz(a,h)anthracene (cas no 53-70-3) Benzo(g,h,i)perylene (cas no 191-24-2) Indeno(1,2,3-cd)pyrene(cas no 193-39-5)
8.20 10.15 10.33 10.88 12.00 12.07 13.94 14.40 17.63 17.74 21.20 21.30 22.28 26.01 26.68 25.88
128 152 153 166 178 178 202 202 228 228 252 252 252 276 278 276
13.93
256
14.20
292
14.99
326
15.67
326
16.25
360
15.90
360
17.21
394
22.70
406
24.86
486
26.83
406
12.61 13.35 15.10 17.55 17.75
163 149 149 149 129
18.87
149
20.45
149
PCBs 2,4,40 -trichlorobiphenyl (PCB-28) (cas no 7012-37-5) 2,20 ,5,50 -tetrachlorobiphenyl (PCB-52) (cas no 35693-99-3) 2,20 ,4,5,50 -pentachlorobiphenyl (PCB-101) (cas no 37680-73-2) 2,30 ,4,40 ,5-pentachlorobiphenyl (PCB-118) (cas no 31508-00-6) 2,20 ,3,4,40 ,50 -hexachlorobiphenyl (PCB-138) (cas no 35065-28-2) 2,20 ,4,40 ,5,50 -hexachlorobiphenyl (PCB-153) (cas no 35065-27-1) 2,20 ,3,4,40 ,5,50 -heptachlorobiphenyl (PCB-180) (cas no 35065-23-3) PBDEs 2,4,40 -Tribromodiphenyl ether (BDE-28) (cas no 41318-75-6) 2,20 ,4,40 -tetra-bromodiphenyl ether (BDE-47) (cas no 5436-43-1) 2,20 ,4,40 ,5-penta-bromodiphenyl ether (BDE-99) (cas no 60348-60-9) PEs Dimethylphthalate (DMP) (cas no 131-11-3) Diethylphthalate (DEP) (cas no 84-66-2) Di-n-butylphthalate (DBP) (cas no 84-74-2) Benzyl butyl phthalate (BBP) (cas no 85-68-7) Bis(2-ethylhexyl)adipate (BEHA) (cas no 103-23-1) Bis(2-ethylhexyl)phthalate (DEHP) (cas no 117-81-7) Di-n-octyl phthalate (DNOP) (cas no 117-84-0)
GmbH). The cartridges for sampling and thermal desorption were stainless-steel tubes (Markes International Ltd.) with the following dimensions: 9 cm length, 6.53 o.d. and 5 mm i.d. and packed with PDMS/Tenax sorbent material. Prior to each use, the sampling tubes were conditioned by thermal cleaning under a nitrogen flow rate of 75 mL min1 at 300 C for 60 min, sealed with end caps and stored under nitrogen atmosphere to prevent any contamination of the sorbent. After loading, the samples were immediately sealed again with the end caps and stored under nitrogen atmosphere until analysis. The samples were analysed within seven days after the collection.
2.1. Sampling
2.2. Analysis
Air samples were collected by active air sampling with a constant flow air sampling pump GSA SG350 (GSA Messgerätebau
All the analyses were performed on a TD-GC-MS system, which consisted of a TD100 Thermal desorber (TD) equipped with a multi-
782
B. Lazarov et al. / Atmospheric Environment 79 (2013) 780e786
tube autosampler (Markes International Ltd.) and coupled to a gas chromatograph Thermo Trace GC Ultra and a mass selective detector Thermo DSQII (Thermo Fisher Scientific Inc.). Thermal desorption of the sampling tubes was carried out at 300 C with a flow rate of 20 mL min1 for 12 min, followed by cold trap (10 C) focussing. Thermal desorber system control was performed using Thermal Desorber System Control Program version 4.4.1 (Markes International Ltd.). The GCeMS system control, data logging and data handling were performed using Xcalibur 2.0 software (Thermo Fisher Scientific Inc.). The chromatographic separation of PAHs, PCBs, PEs and PBDEs was conducted on Rxi-5Sil MS capillary column (Restek) 30 m 0.25 mm i.d. 0.25 mm film thickness. Helium was used as a carrier gas. The gas chromatographic conditions are listed in Table 2. All compounds were analysed by using time scheduled Selected Ion Monitoring (SIM) of the most intensive ion fragment of each compound (listed in Table 1). The MS ion source and the transfer line from the GC to the MS were kept at 250 C and 320 C respectively. 3. Results and discussion 3.1. Method optimisation for the assessment of PAHs in air 3.1.1. Dynamic retention efficiency Dynamic retention efficiency of five deuterated PAHs was evaluated at different flow rates and durations on the adsorbing PDMS/Tenax sampling tubes according to ASTM D4861-11 (ASTM International, 2011). Therefore, sampling tubes were spiked with a standard solution of d8-naphthalene, d10-fluorene, d10fluoranthene and d12-benzo(g,h,i)perylene in methanol. The absolute adsorbed quantity of each compound on every tube was 3000 pg. Clean air (RH 50%) was then passed through each tube at variable flow rates and sampling times, allowing an evaluation of the impact of different sampled air volumes on the dynamic retention efficiency of the deuterated compounds, as listed in Table 3. A comparison of the sampling efficiency at different sampling duration and air volume settings (Table 3) indicates that no breakthrough occurs for sampling volumes up to 732 L. In the current study, an air volume of 480 L (333 mL min1 for 24 h) was selected as suitable sampling volume for PAHs sample collection. Since this reported sampling method collects gas phase SVOCs as well as particulate matter, the air flow rate will influence the cutoff of the sampled particulate fraction in field studies. Therefore, taking into account the dimensions of the sampling tubes, the theoretical aerodynamic particle cut-off of the tubes respecting classical gravitational deposition velocity at different air flow rates
Table 3 Dynamic retention efficiency at different air volumes. Sample volume, L Sample rate, mL min1 Sampling time, h
180 124 24
176 98 30
282 98 48
445 247 30
732 254 48
d8-naphthalene d10-fluorene d10-fluoranthene d12-benzo(g,h,i)perylene
95% 93% 95% 99%
96% 94% 95% 102%
98% 99% 99% 108%
106% 102% 108% 128%
98% 100% 107% 122%
has been calculated (Brockmann, 2001) in calms (i.e. wind speed less than 0.5 m s1). According to this study, a flow rate of 50 mL min1 collects PM37, 100 mL min1 collects PM57, and 200 mL min1 collects PM80. These values imply that, when the method is applied at a flow rate of 333 mL min1, a PM fraction with an aerodynamic diameter exceeding PM80 is collected, which definitely includes all required SVOC particles. Furthermore, the measurement of SVOC compounds originating from combustion sources, such as PAHs which typically occur in a PM size fraction below 1 mm (Yang et al., 2007) won’t be influenced by the operational flow rate. However, collected SVOCs typically originating from product emissions, such as PEs and PBDEs which tend to occur both in smaller PM fractions (<2 mm) (Liu et al., 2012) and in larger (resuspended) dust fractions (Schripp et al., 2010), might be influenced by the flow rate. 3.1.2. Optimisation of TD parameters To obtain the best analytical conditions for PAH analyses in terms of sensitivity and reproducibility, the TD parameters, such as cold trap desorption temperature and desorption time, were examined and optimised. Sampling tubes, spiked with a mixture of the target PAHs, were evaluated at different cold trap desorption temperatures. Based on previous experiments (Wauters et al., 2008) and on the technical potential of the instrument, three cold trap desorption temperatures of 350, 375 and 400 C were studied. The amount of each individual compound in the tube was approximately 1000 pg. The intensity of the MS response of each compound is shown in Fig. 1. The results show that there is no significant influence of the trap desorption temperature on the MS response for compounds with lower boiling points. Furthermore, the intensity of the MS response for the compounds with higher boiling points, increases with an increasing the
350°C
Signal intensity nsity
375°C
400°C
1400000 1200000 1000000 800000
Table 2 Gas chromatographic conditions.
600000
Compound group
Oven conditions
Carrier gas conditions
PAHs
80 C (7 min) e 25 C min1 e 255 C e 5 C min1 e 325 C
PCBs
80 C (7 min) e 25 C min1 e 255 C e 5 C min1 e 300 C
PBDEs
100 300 100 255
65 kPa (7 min) e 8 kPa min1 e 121 kPa e 2 kPa min1 e 149 kPa 65 kPa (7 min) e 8 kPa min1 e 125 kPa e 2 kPa min1 e 140 kPa (1 min) 60 kPa (7 min) e 4 kPa min1 e 140 kPa (13 min) 70 kPa (5 min) e 10 kPa min1 e 120 kPa e 1 kPa min1 e 135 kPa
Benzo(g,h,i)perylene
Dibenz(a,h)anthracene
Benzo(a)pyrene
Indeno(1,2,3-cd)pyrene
Benzo(k)fluoranthene
Chrysene
Benzo(b)fluoranthene
Pyrene
Anthracene
Fluoranthene
Fluorene
Phenanthrene
Acenaphthene
Naphthalene
(7 min) e 10 C min1 e (13 min) (5 min) e 25 C min1 e e 5 C min1 e 300 C
0 Acenaphthylene
C C C C
200000 Benz(a)anthracene
PEs
400000
Fig. 1. Trap desorption temperature effect on MS response intensity for each PAH.
B. Lazarov et al. / Atmospheric Environment 79 (2013) 780e786
trap desorption temperature. This resulted in the selection of a cold trap desorption temperature of 400 C as an optimal temperature. The cold trap desorption time was optimised as well. Sampling tubes, spiked with 1000 pg of each individual PAH, were evaluated at a 400 C cold trap desorption temperature in four different desorption times: 10, 11, 13 and 15 min. As shown in Fig. 2, the MS response for the individual compounds was independent of the desorption duration. Therefore, a cold trap desorption time of 10 min was selected for this procedure. 3.2. Method performance The performance of the method was evaluated via an estimation of the linearity, the accuracy and the repeatability of the MS response. The Xcalibur 2.0 software (Thermo Fisher Scientific Inc.) was used for quantification, based on peak area. For the estimation of the linear dynamic range, five-point calibration curves were constructed in the range of 200e8000 pg, using a least-square regression analysis. As a result, correlation coefficients higher than 0.998 were obtained for all compounds (Table 4). Because of the higher expected quantity in air samples, the linear dynamic range of naphthalene was tested from 1 ng up to 250 ng. The obtained correlation coefficient in this range, also using a least-square regression analysis, was 0.9952. Limits of detection (LODs) were determined based on the analysis of tubes spiked with standard solution, and calculated as the quantity of the compound giving a signal to noise ratio of 3. The calculations were done for a sampling volume of 480 L. The LODs ranged between 0.003 ng m3 for Fluorene and 0.380 ng m3 for Benzo(k)fluoranthene. Table 4 lists the LOD for each studied PAH compound. The accuracy and repeatability of the technique were calculated from the replicate (n ¼ 6) analysis of sampling tubes, spiked within one day, with a standard reference material (SRM) ERM-AC213 of approximately 3000 pg of each compound. The recovery of the analysed compounds was within a range of 80e97%, except for benzo(b)fluoranthene which was characterised by a recovery of 146%. The higher recovery of benzo(b)fluoranthene might be caused by interference of benzo(j)fluoranthene, which was present in the SRM, but not studied in this method. The calculated relative standard deviations (RSDs) were less than 10% for most of the PAH compounds, except for benzo(k)fluoranthene e 12.6% (Table 5). The carry-over effect of each compound was investigated at three concentration levels: 250 pg, 1000 pg and 8000 pg. Six
Signal intensity
10 min
11 min
13 min
15 min
3000000 2500000 2000000 1500000
783
Table 4 Limit of detection and correlation coefficients obtained for the analysis of selected PAHs. Compound
LOD, ng m3
R2
Naphthalene Acenaphthylene Acenaphthene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benz(a)anthracene Chrysene Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(a)pyrene Indeno(1,2,3-cd)pyrene Dibenz(a,h)anthracene Benzo(g,h,i)perylene
0.033 0.250 0.005 0.003 0.005 0.194 0.097 0.106 0.008 0.008 0.026 0.380 0.062 0.540 0.095 0.450
0.9997 0.9998 0.9999 0.9999 0.9999 0.9985 0.9999 0.9998 0.9996 0.9999 0.9997 0.9997 0.9999 0.9992 0.9994 0.9998
sampling tubes, spiked with standard solution of PAHs, were analysed followed by the analysis of a set of two blank sampling tubes. The carry-over effect was then calculated as the percentage of the compound that could be quantified in the blank sampling tubes. The result indicated a carry-over effect on the investigated PAHs of less than 25% (see Table 6), independent of the spiked PAHs concentration levels. As expected, the carry-over effect increased with the increasing boiling point of the compounds, due to the tendency of these compounds to be trapped at so called “cold spots” of the system. In the analytical sequence, this small carry-over effect is taken into account by measuring a blank tube after each real sample. 3.3. Extension of the scope to other SVOCs The possibility of widening the application scope of this method for PAH sampling to the assessment of other semi-volatile organic air pollutants has been suggested by Wauters et al. (2008). Therefore, this study also explored the suitability of the PDMS/Tenax method to monitor Polychlorinated Biphenyls as well as Polybrominated diphenyl ethers and Phthalates. The examined compounds in each group are shown in Table 1. 3.3.1. PCBs Table 7 shows the limit of detection, the recovery and the repeatability of the method for PCBs. The recovery and repeatability of the technique for PCB detection was calculated from the replicate (n ¼ 4) analysis of sampling tubes, spiked with a standard solution of PCBs. The obtained results show a recovery in the range of 99e 108% and an RSD less than 5%. The linearity of the method for PCBs was evaluated by six point calibration curve at the range of 0.1e 10 ng. The calculated correlation coefficients were greater than 0.996 for each compound.
1000000 Table 5 Accuracy and repeatability of the method for PAH detection.
500000 Benzo(g,h,i)perylene
Fig. 2. Trap desorption time effect on MS response intensity of each PAH.
Dibenz(a,h)anthracene
Benzo(a)pyrene
Indeno(1,2,3-cd)pyrene
Benzo(k)fluoranthene
Chrysene
Benzo(b)fluoranthene
Benz(a)anthracene
Pyrene
Fluoranthene
Anthracene
Fluorene
Phenanthrene
Acenaphthene
Naphthalene
Acenaphthylene
0
Compound
True value, pg
Accuracy, %
RSD, %
Benz(a)anthracene Chrysene Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(a)pyrene Indeno(1,2,3-cd)pyrene Dibenz(a,h)anthracene Benzo(g,h,i)perylene
3044 3015 3005 3015 2857 2955 2719 3024
84 80 146 86 88 89 97 88
5.3 6.8 9.3 12.6 4.2 5.6 6.1 7.2
784
B. Lazarov et al. / Atmospheric Environment 79 (2013) 780e786
Table 6 Carry-over during thermal desorption of selected PAHs. Compound
250 pg
1000 pg
8000 pg
Naphthalene Acenaphthylene Acenaphthene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benz(a)anthracene Chrysene Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(a)pyrene Indeno(1,2,3-cd)pyrene Dibenz(a,h)anthracene Benzo(g,h,i)perylene
13.7% 3.0% 6.0% 5.0% 12.1% 9.9% 5.7% 6.7% 11.3% 14.2% 12.1% 10.8% 10.5% 18.0% 19.4% 19.1%
3.7% 1.0% 1.7% 1.8% 11.6% 9.1% 6.9% 5.5% 9.6% 11.6% 9.6% 10.3% 11.0% 17.7% 20.4% 18.6%
1.5% 1.9% 1.7% 2.9% 2.7% 2.6% 7.3% 7.8% 9.4% 12.0% 11.4% 12.0% 14.0% 24.6% 25.2% 24.7%
The LODs of the selected PCBs were determined by analysing sampling tubes spiked with a standard solution of the compounds in a quantity, giving a signal to noise ratio of 3. Based on the previous optimisation steps for PAHs, the calculations were done for a sampling volume of 480 L. The calculated LODs are in a range of 0.004 ng m3 for PCB-180 up to 0.225 ng m3 for PCB-118 (Table 7). Based on the above discussed method performance characteristics, the method shows good results compared to the other reported studies for the analysis of PCBs in air based on active sampling (Barro et al., 2005; Heinzow et al., 2007). 3.3.2. PBDEs Similar to the PCBs, selected method validation parameters to evaluate the applicability of the method for sampling PBDEs were assessed. The recovery and repeatability of the technique were calculated from a replicate (n ¼ 4) analysis of sampling tubes spiked with PBDEs. The linearity was estimated in the range of 3e32 ng of each compound. The obtained correlation coefficients were greater than 0.990. The LODs were determined by analysing sampling tubes spiked with amount of the compounds giving a signal to noise ratio of 3. Also for PBDEs, these first exploratory steps were calculated for a sampling volume of 480 L. The obtained LODs were in the range of 0.113e0.201 ng m3, recovery up to 98%, the repeatability was in the order of 2% (see Table 7). Table 7 Limit of detection, recovery and repeatability of the method for selected compounds. Compound
LOD, ng m3
Recovery, %
PCBs (n ¼ 4) PCB-28 PCB-52 PCB-101 PCB-118 PCB-138 PCB-153 PCB-180
0.036 0.003 0.012 0.225 0.042 0.005 0.004
105 108 99 103 101 102 105
3.0 2.1 4.0 5.1 2.5 2.9 1.1
PBDE (n ¼ 4) BDE-28 BDE-47 BDE-99
0.113 0.174 0.201
95 95 98
1.9 1.7 2.1
PEs (n ¼ 5) DMP DEP DBP BBP BEHA DEHP DNOP
0.229 0.002 0.005 0.025 0.015 0.002 0.009
109 109 99 110 125 113 95
7.8 20.0 13.7 10.6 4.7 7.3 6.4
RSD, %
Table 8 The median, lowest and highest values of measured concentrations of selected PAHs from city of Antwerp. Component
Median ng m3
Lowest ng m3
Highest ng m3
Naphthalene Acenaphthylene Acenaphthene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benz(a)anthracene Chrysene Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(a)pyrene Indeno(1,2,3-cd)pyrene Dibenz(a,h)anthracene Benzo(g,h,i)perylene
188.1 5.43 5.94 7.40 11.3 1.02 2.41 2.15 0.40 0.58 0.51 0.36 0.32 0.40 0.43 0.51
60.5 0.40 3.59 3.83 1.17 0.45 0.73 0.96 0.29 0.29 0.29 0.29 0.29 0.29 0.29 0.29
1126 55.3 20.6 25.6 47.9 1.67 9.92 13.7 0.51 3.69 1.37 0.73 0.51 0.51 0.61 0.81
3.3.3. PEs The evaluation of the applicability of the method to sample PEs was performed as well. Similar to the previous groups of chemicals, the determination of LODs was done by analysing sampling tubes, spiked with an amount of the compounds giving a signal to noise ratio of 3 and calculated for a sampling volume of 480 L. The obtained LODs were in the range of 0.001 ng m3 (DEP) e 0.069 ng m3 (DMP). Due to the higher sensitivity of the method for selected PEs, the sampling volume was reduced to 144 L in order to improve the sampling productivity. The calculated LODs for this reduced sampling volume are in range of 0.002 ng m3 (DEP)e0.229 ng m3 (DMP) (Table 7). The linearity of the method was estimated in the range of 1.5e 45 ng of each compound. The obtained correlation coefficients were greater than 0.996. The recovery and repeatability of the technique was calculated from a replicate (n ¼ 5) analysis of sampling tubes, spiked with standard mixture of PEs. The obtained recoveries for selected PEs are in range of 95% (DNOP) e 125% (BEHA) and the repeatabilities are below 13.7%, except for DEP e 20% (Table 7). 4. Field samples of SVOCs As a final step of the reported study, the optimised method was applied in the field to sample PAHs, PCBs, PEs and PBDEs in different urban environments in Flanders, Belgium. Twenty-six air samples were collected by active sampling on PDMS/Tenax and analysed for PAHs using the optimised method during three months (MarcheMay 2012) in the city of Antwerp, Belgium. Furthermore, the reproducibility of the method was tested through the analysis of parallel samples in two different labs. For each sample, the sampling time was 24 h with a total sampled air volume of 480 L. The compound concentrations were calculated using external calibration curves. Table 10 shows the median, minimum and maximum values of the PAHs concentrations detected during the three months sampling period. It should be noted that the results listed in Table 8 show consistently higher concentration levels when compared to data reported in previous PAH field studies in the same region, using the classical method for PAHs sampling on PUF foam (Du Four et al., 2005; Ravindra et al., 2006). The difference between the concentration levels tends to be higher for the more volatile (lower molecular mass) compounds than for the less volatile (higher molecular mass) ones.
B. Lazarov et al. / Atmospheric Environment 79 (2013) 780e786 Table 9 Results of analysing a parallel samples in two different labs using the optimised method. Compound
Lab1 ng m3
Lab2 ng m3
Difference from the average value
Naphthalene Acenaphthylene Acenaphthene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benz(a)anthracene Chrysene Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(a)pyrene Indeno(1,2,3-cd)pyrene Dibenz(a,h)anthracene Benzo(g,h,i)perylene
102 1.00 7.90 10.4 11.0 0.35 2.37 0.95 0.55 0.64 1.78 1.28 1.03 1.29 1.55 0.61
131 1.54 7.00 16.9 16.6 1.11 3.13 0.95 n.d.a 0.82 1.74 1.15 1.42 1.77 1.96 1.69
12% 21% 6% 24% 20% 52% 14% 0% e 12% 1% 5% 16% 16% 12% 47%
a
Concentration (ng m3) Sample 1
Sample 2
Sample 3
Average
RSD
11.3 3.11 0.89 0.34 0.38 0.30 n.d.a n.d.a n.d.a
12.1 3.40 1.19 0.37 0.39 0.40 n.d.a n.d.a n.d.a
12.1 3.54 1.31 0.38 0.37 0.45 n.d.a n.d.a n.d.a
11.8 3.4 1.13 0.36 0.38 0.38 e e e
3.9% 6.5% 19.2% 6.5% 1.9% 19.9% e e e
n.d. e not detected.
Concentration, ng m3 S1
S2
S3
63.7 8.8 3.4 437 213 65.2 583 524 116 3.5 20.6 2.1 4.4 6.0 3.8 Interferences n.d.a n.d.a n.d.a
S4
S5
S6
S7
S8
S9
S10
2.3 3.6 17.6 0.4 1.6
56.4 1380 153 9.8 12.2
14.6 69.2 95.7 7.1 4.7
n.d.a
n.d.a
n.d.a
57.8 210 283 10.0 4.1 681 n.d.a
10.9 53.8 334 12.2 3.5 100 n.d.a
454 161 1190 15.3 18.1 87.6 n.d.a
16.7 85.4 71.5 13.9 5.6 58.0 n.d.a
n.d. e not detected.
5. Conclusion
Table 10 Measured concentrations of the selected PCBs and PBDEs from city of Genk, Belgium.
a
Compound
a
This reflection is also confirmed by field test outcomes of an experiment reported by Wauters et al. (2008), which was focussed on a comparison between PDMS/Tenax sampling followed by thermo desorption and the determination of PAHs by sampling on PUF foam. The results from the analysis of the parallel samples in different labs are listed in Table 9. The obtained concentrations levels by the two labs are very similar, with a mean difference of 14% relative to the average concentration. For the PCBs and PBDEs field experiment, six samples were collected and analysed according to the optimised method. All samples were collected during 24 h at a sampling flow rate of 333 mL min1, which gives sampling volume of 480 L per sample. The samples were taken from outdoor locations in the city of Genk, Belgium. The compound concentrations were calculated using external calibration curves. Table 10 shows the measured concentrations of the compounds. Additionally, phthalates have been measured in indoor environments. The results are reported in Table 11. Based on this preliminary test, the air concentrations of the selected PBDEs in outdoor air were below the detection limit. In order to measure PBDEs in air, a different environmental setting (indoor air concentrations are expected to be higher than outdoor levels (Batterman et al., 2009)), a higher sampling volume, and/or a more sensitive detector than the single quadrupole which was applied in this study, might be needed. A more focussed optimisation strategy is currently being developed by the authors in order to fully explore the possibilities of this method for flame retardant detection.
PCB 28 PCB 52 PCB 101 PCB 118 PCB 138 PCB 153 BDE-28 BDE-47 BDE-99
Table 11 Measured concentrations of the selected phthalates from different sampling locations across Flanders, Belgium.
DMP DEP DBP BBP BEHA DEHP DNOP
n.d. e not detected.
Compound
785
The described air sampling method, based on active air sampling on PDMS/Tenax tubes, followed by thermal desorption and GCeMS analysis, showed a good efficiency for the determination of selected classes of semi-volatile compounds, such as PAHs. The method is fast, sensitive and relatively easy to apply in the field. No additional sample preparation steps are required, which significantly reduces the risk of sample contamination and sample losses as well as the required sample analysis time. Using low flow air sampling equipment makes this technique advantageous for applications in indoor environments or for personal sampling as well. The present optimisation was performed in a group-by-group approach, which resulted in small differences between the analytical procedures for each SVOC group. Try-outs have however indicated that methods for analysing PAHs, PCBs and PEs could be combined in one method for the simultaneous analysis of selected target compounds. Due to peak overlap, PBDEs congeners should be analysed using a separate analytical procedure. The applicability of the technique in real samples was tested in different urban settings. The method showed an increased sensitivity to detect PAHs compared to usually applied methods for air sampling; furthermore it also allows a reduction of the traditionally required sampling times. The applicability of the method to sample PCBs and PEs is also confirmed. However, due to the low concentrations of PBDEs in outdoor air, an improvement of the limit of detection for this group of SVOCs may be beneficial for outdoor as well as indoor air sampling. The reported optimisation steps are currently being expanded to the more thorough study of the method for flame retardant assessment in indoor environments. An increase of the air sampling volume, or the use of a more sensitive detection technique are initiatives that are being considered to further enlarge the applicability of this promising air sampling method. Acknowledgement Support was obtained from the project “Officair” funded by the 7the Framework Programme (contract 265267) under the Theme ENV 2010.1.2.2.1. Further support was obtained from the project “Inflame” funded by the 7the Framework Programme (contract 164600) under the Theme People-2010-ITN. References ASTM International, 2011. ASTM D4861-11 Standard Practice for Sampling and Selection of Analytical Techniques for Pesticides and Polychlorinated Biphenyls in Air. D4861-11 ed. D4861e11. Barro, R., Ares, S., Garcia-Jares, C., Llompart, M., Cela, R., 2005. A simple and fast micromethod for the analysis of polychlorinated biphenyls in air by sorbent enrichment and ultrasound-assisted solvent extraction. Anal. Bioanal. Chem. 381, 255e260.
786
B. Lazarov et al. / Atmospheric Environment 79 (2013) 780e786
Batterman, S., Chen, T.C., Chernyak, S., Godwin, C., 2009. Design and performance evaluation of a medium flow sampler for airborne brominated flame retardants (BFRs). J. Environ. Monit. 11, 858e866. Bidleman, T.F., Billings, W.N., Foreman, W.T., 1986. Vapor-particle partitioning of semivolatile organic compounds: estimates from field collections. Environ. Sci. Technol. 20, 1038e1043. Brockmann, J.E., 2001. Sampling and transport of aerosols. In: Baron, P.A., Willeke, K. (Eds.), Aerosol Measurement e Principles, Techniques, and Applications, second ed. John Wiley & Sons, pp. 143e170. Du Four, V.A., Janssen, C.R., Brits, E., Larebeke, N.V., 2005. Genotoxic and mutagenic activity of environmental air samples from different rural, urban and industrial sites in Flanders, Belgium. Mutat. Res. Genet. Toxicol. Environ. Mutagen. 588, 106e117. Elflein, L., Berger-Preiss, E., Levsen, K., Wünsch, G., 2003. Development of a gas chromatographyemass spectrometry method for the determination of household insecticides in indoor air. J. Chromatogr. A 985, 147e157. EN ISO 16000e13, 2008. Indoor Air e Part 13: Determination of Total (Gas and Particle-phase) Polychlorinated Dioxin-like Biphenyls (PCBs) and Polychlorinated Dibenzo-p-dioxins/dibenzofurans (PCDDs/PCDFs) e Collection on Sorbent-backed Filters. Galarneau, E., Harner, T., Shoeib, M., Kozma, M., Lane, D., 2006. A preliminary investigation of sorbent-impregnated filters (SIFs) as an alternative to polyurethane foam (PUF) for sampling gas-phase semivolatile organic compounds in air. Atmos. Environ. 40, 5734e5740. Heinzow, B., Mohr, S., Ostendorp, G., Kerst, M., Körner, W., 2007. PCB and dioxin-like PCB in indoor air of public buildings contaminated with different PCB sourcesderiving toxicity equivalent concentrations from standard PCB congeners. Chemosphere 67, 1746e1753. Krupadam, R., Bhagat, B., Khan, M., 2010. Highly sensitive determination of polycyclic aromatic hydrocarbons in ambient air dust by gas chromatography-mass spectrometry after molecularly imprinted polymer extraction. Anal. Bioanal. Chem. 397, 3097e3106. Liu, C., Morrison, G.C., Zhang, Y., 2012. Role of aerosols in enhancing SVOC flux between air and indoor surfaces and its influence on exposure. Atmos. Environ. 55, 347e356.
May, A.A., Ashman, P., Huang, J., Dhaniyala, S., Holsen, T.M., 2011. Evaluation of the polyurethane foam (PUF) disk passive air sampler: computational modeling and experimental measurements. Atmos. Environ. 45, 4354e4359. Namiesnik, J., Zabiegala, B., Kot-Wasik, A., Partyka, M., Wasik, A., 2005. Passive sampling and/or extraction techniques in environmental analysis: a review. Anal. Bioanal. Chem. 381, 279e301. Ni, Y., Kumagai, K., Yanagisawa, Y., 2007. Measuring emissions of organophosphate flame retardants using a passive flux sampler. Atmos. Environ. 41, 3235e3240. Ravindra, K., Bencs, L., Wauters, E., de Hoog, J., Deutsch, F., Roekens, E., Bleux, N., Berghmans, P., van Grieken, R., 2006. Seasonal and site-specific variation in vapour and aerosol phase PAHs over Flanders (Belgium) and their relation with anthropogenic activities. Atmos. Environ. 40, 771e785. Schauer, C., Niessner, R., Pöschl, U., 2003. Polycyclic aromatic hydrocarbons in urban air particulate matter: decadal and seasonal trends, chemical degradation, and sampling artifacts. Environ. Sci. Technol. 37, 2861e2868. Schripp, T., Fauck, C., Salthammer, T., 2010. Chamber studies on mass-transfer of di(2-ethylhexyl)phthalate (DEHP) and di-n-butylphthalate (DnBP) from emission sources into house dust. Atmos. Environ. 44, 2840e2845. Temime, B., Francois, S., Monod, A., Wortham, H., 2002. An experimental set up of a PAH vapour generator and its use to test an annular denuder. Environ. Pollut. 120, 609e616. Wania, F., Shen, L., Lei, Y.D., Teixeira, C., Muir, D.C.G., 2003. Development and calibration of a resin-based passive sampling system for monitoring persistent organic pollutants in the atmosphere. Environ. Sci. Technol. 37, 1352e1359. Wauters, E., Van Caeter, P., Desmet, G., David, F., Devos, C., Sandra, P., 2008. Improved accuracy in the determination of polycyclic aromatic hydrocarbons in air using 24 h sampling on a mixed bed followed by thermal desorption capillary gas chromatographyemass spectrometry. J. Chromatogr. A 1190, 286e293. Weschler, C.J., Nazaroff, W.W., 2008. Semivolatile organic compounds in indoor environments. Atmos. Environ. 42, 9018e9040. World Health Organization, 1989. Indoor air quality: organic pollutants. EURO Rep. Stud. 111. Berlin (West). Yang, C.R., Lin, T.C., Chang, F.H., 2007. Particle size distribution and PAH concentrations of incense smoke in a combustion chamber. Environ. Pollut. 145, 606e 615.
Contents lists available at ScienceDirect
Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv
Optimisation steps of an innovative air sampling method for semi volatile organic compounds Borislav Lazarov a, *, Rudi Swinnen a, Maarten Spruyt a, Eddy Goelen a, Marianne Stranger a, Gilbert Desmet b, Eric Wauters b a b
Environmental Risk and Health Unit, VITO, Boeretang 200, 2400 Mol, Belgium Flemish Environmental Agency (VMM), Raymonde de Larochelaan 1, B-9051 Sint-Denijs-Westrem, Belgium
h i g h l i g h t s An innovative method for the measurement several groups of SVOCs in air has been optimised. As the method collects gaseous as well as particulate matter SVOCs, it avoids underestimating the total air concentration. The applicability of the method to sample PAHs, PCBs and PEs from different urban environments was confirmed. The presented strategy reduces the risk of contamination during sample preparation steps compared to more traditional techniques.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 26 April 2013 Received in revised form 19 July 2013 Accepted 24 July 2013
This work describes optimisation steps of an innovative method for the measurement several groups of semi-volatile organic compounds (SVOCs) in air, collecting both gaseous and particulate air fractions. It is based on active air sampling on sorption tubes (consisting of polydimethylsiloxane (PDMS) and Tenax TA), followed by thermal desorption and gas chromatography mass spectrometry analysis (TDeGCeMS). The optimised method was validated in the laboratory for the measurement of selected target compounds from the following chemical classes: polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs) and phthalate esters (PEs). It was applied in different Belgian urban outdoor as well as indoor environments. The new method is characterised by limits of detection in the range of 0.003e0.3 ng m3 for PAHs, 0.004e0.2 ng m3 for PCBs, 0.113 e0.201 ng m3 for PBDEs and 0.002e0.2 ng m3 for PEs, a linearity of 0.996 and a repeatability of less than 10% for all studied compounds. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Air monitoring Semi-volatile organic compounds Thermal desorption GCeMS Mixed bed sampling
1. Introduction According to the definition of the World Health Organisation (WHO), semi-volatile organic compounds (SVOCs) are organic chemicals with boiling points ranging from 240e260 C to 380e 400 C (World Health Organization, 1989). This range covers a large number of compounds that are present in outdoor and indoor air (e.g. polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs), phthalate esters (PEs). These classes of compounds may occur in meaningful abundances both in the gas phase and on the surface of airborne particles, which recognises them as a human health concern. Although in recent years, the study of the occurrence, fate and
* Corresponding author. Tel.: þ32 14 33 69 27; fax: þ32 14 33 55 99. E-mail addresses: [email protected], [email protected] (B. Lazarov). 1352-2310/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atmosenv.2013.07.059
human exposure to these compounds has become an important research topic, SVOCs have not been so widely studied as certain other classes of air pollutants such as VOCs, airborne particles and inorganic gaseous pollutants (Weschler and Nazaroff, 2008). The high degree of analytical challenges in measuring SVOCs has impeded progress in studying them. Sampling of SVOCs is mostly done by the collection of both the gas and the particle phase, using high volume samplers. In most cases the particulate phase is trapped on a filter (quartz or glass fibre) and the breakthrough of the gas phase is subsequently trapped on a sorbent (PUF or XAD) (EN ISO 16000e13, 2008; Batterman et al., 2009; Elflein et al., 2003). A widely used method for sample treatment after collecting several m3 of air, is solvent desorption of the filters (Soxhlet extraction, accelerated solvent extraction, sonication) followed by analysing the compounds of interest by GCeMS (gas chromatography coupled to mass chromatography) or HPLC techniques. There are several limitations and
B. Lazarov et al. / Atmospheric Environment 79 (2013) 780e786
disadvantages related to this traditional method for sampling and detection of SVOCs. These include the lower sampling efficiency of compounds with higher vapour pressure (Bidleman et al., 1986) and the underestimation of the actual content caused by filter reactions with oxidative gases (Schauer et al., 2003). The method also implies the use of high-volume sampling pumps, which are not suitable for personal and indoor air sampling (due to dimensions and noise nuisance) and need electricity power supply, which is not always convenient when sampling outdoors. Another disadvantage is that the analysis method is time-consuming and requires the samples to be manipulated, which means that there is considerable risk of contamination and/or losing some of the compounds. Several new methods, based on diffusion denuders (Temime et al., 2002), passive samplers (Wania et al., 2003; Namiesnik et al., 2005; Ni et al., 2007; May et al., 2011), sorbentimpregnated filters (Galarneau et al., 2006) or molecular imprinted polymers (Krupadam et al., 2010) have been developed as alternatives for the high-volume sampling method. These methods have advantages compared to the high-volume sampling method, such as more efficient sampling of the compounds with higher vapour pressure (diffusion denuders), elimination of pumping unit (passive samplers) and reduction in size, simplified sample handling and decrease in solvent consumption (sorbent-impregnated filters, molecular imprinted polymers). However, those methods have other disadvantages such as long sampling time (denuders and passive samplers), and possible losses due to photo degradation reactions (passive samplers) that are not excluded as well. As an alternative for these methods, Wauters et al. (2008) have introduced another method for PAHs sampling, based on active sampling on sorption tubes consisting of polydimethylsiloxane (PDMS) foam, PDMS particles and a Tenax TA bed, followed by thermal desorption and analysis by GCeMS. The method is characterised by the advantages to only require low flow sampling (using quasi noiseless and portable sampling pumps), and to considerably reduce the risk of contamination during the sample preparation and analysis steps. The main objective of this reported study is to explore the possibilities of using the mixed bed sampling method to sample and analyse different classes of SVOCs, by performing several optimisation steps. The optimised method was validated for measurement of the listed target compounds (see Table 1) in different Belgian urban outdoor as well as indoor environments. 2. Material and methods High purity standards (98.0e99.9%) were used in this study. The standard calibration mix of PAHs in methylene chloride with a concentration 2000 mg mL1 was purchased from Restek, USA. Deuterated PAHs (d8-naphthalene, d10-fluorene, d10fluoranthene, d12-benzo(a)pyrene and d12-benzo(g,h,i)perylene) and PCBs were purchased from Promochem/C.N. Schmidt, The Netherlands. PBDEs standards were purchased from Wellington Laboratories, Canada. All standard solutions were prepared by dilution in methanol. The solvents were GC grade with purity >99.9% (Merck, Germany). The standard reference material of PAHs (ERM-AC213) was purchased from Institute for Reference Materials and Measurements (IRMM, Belgium). Nitrogen gas of 99.999% purity was used for spiking the thermal desorption sorbent tubes and 99.999% pure helium gas was used for chromatographic analysis.
781
Table 1 Target compounds and respective retention times and the quantification ion studied in air. Compound
RT, min
Q, m/z
PAHs Naphthalene (cas no 91-20-3) Acenaphthylene (cas no 208-96-8) Acenaphthene (cas no 83-32-9) Fluorene (cas no 86-73-7) Phenanthrene (cas no 85-01-8) Anthracene (cas no 120-12-7) Fluoranthene (cas no 206-44-0) Pyrene (cas no 129-00-0) Benz(a)anthracene (cas no 56-55-3) Chrysene (cas no 218-01-9) Benzo(b)fluoranthene (cas no 205-99-2) Benzo(k)fluoranthene (cas no 207-08-9) Benzo(a)pyrene (cas no 50-32-8) Dibenz(a,h)anthracene (cas no 53-70-3) Benzo(g,h,i)perylene (cas no 191-24-2) Indeno(1,2,3-cd)pyrene(cas no 193-39-5)
8.20 10.15 10.33 10.88 12.00 12.07 13.94 14.40 17.63 17.74 21.20 21.30 22.28 26.01 26.68 25.88
128 152 153 166 178 178 202 202 228 228 252 252 252 276 278 276
13.93
256
14.20
292
14.99
326
15.67
326
16.25
360
15.90
360
17.21
394
22.70
406
24.86
486
26.83
406
12.61 13.35 15.10 17.55 17.75
163 149 149 149 129
18.87
149
20.45
149
PCBs 2,4,40 -trichlorobiphenyl (PCB-28) (cas no 7012-37-5) 2,20 ,5,50 -tetrachlorobiphenyl (PCB-52) (cas no 35693-99-3) 2,20 ,4,5,50 -pentachlorobiphenyl (PCB-101) (cas no 37680-73-2) 2,30 ,4,40 ,5-pentachlorobiphenyl (PCB-118) (cas no 31508-00-6) 2,20 ,3,4,40 ,50 -hexachlorobiphenyl (PCB-138) (cas no 35065-28-2) 2,20 ,4,40 ,5,50 -hexachlorobiphenyl (PCB-153) (cas no 35065-27-1) 2,20 ,3,4,40 ,5,50 -heptachlorobiphenyl (PCB-180) (cas no 35065-23-3) PBDEs 2,4,40 -Tribromodiphenyl ether (BDE-28) (cas no 41318-75-6) 2,20 ,4,40 -tetra-bromodiphenyl ether (BDE-47) (cas no 5436-43-1) 2,20 ,4,40 ,5-penta-bromodiphenyl ether (BDE-99) (cas no 60348-60-9) PEs Dimethylphthalate (DMP) (cas no 131-11-3) Diethylphthalate (DEP) (cas no 84-66-2) Di-n-butylphthalate (DBP) (cas no 84-74-2) Benzyl butyl phthalate (BBP) (cas no 85-68-7) Bis(2-ethylhexyl)adipate (BEHA) (cas no 103-23-1) Bis(2-ethylhexyl)phthalate (DEHP) (cas no 117-81-7) Di-n-octyl phthalate (DNOP) (cas no 117-84-0)
GmbH). The cartridges for sampling and thermal desorption were stainless-steel tubes (Markes International Ltd.) with the following dimensions: 9 cm length, 6.53 o.d. and 5 mm i.d. and packed with PDMS/Tenax sorbent material. Prior to each use, the sampling tubes were conditioned by thermal cleaning under a nitrogen flow rate of 75 mL min1 at 300 C for 60 min, sealed with end caps and stored under nitrogen atmosphere to prevent any contamination of the sorbent. After loading, the samples were immediately sealed again with the end caps and stored under nitrogen atmosphere until analysis. The samples were analysed within seven days after the collection.
2.1. Sampling
2.2. Analysis
Air samples were collected by active air sampling with a constant flow air sampling pump GSA SG350 (GSA Messgerätebau
All the analyses were performed on a TD-GC-MS system, which consisted of a TD100 Thermal desorber (TD) equipped with a multi-
782
B. Lazarov et al. / Atmospheric Environment 79 (2013) 780e786
tube autosampler (Markes International Ltd.) and coupled to a gas chromatograph Thermo Trace GC Ultra and a mass selective detector Thermo DSQII (Thermo Fisher Scientific Inc.). Thermal desorption of the sampling tubes was carried out at 300 C with a flow rate of 20 mL min1 for 12 min, followed by cold trap (10 C) focussing. Thermal desorber system control was performed using Thermal Desorber System Control Program version 4.4.1 (Markes International Ltd.). The GCeMS system control, data logging and data handling were performed using Xcalibur 2.0 software (Thermo Fisher Scientific Inc.). The chromatographic separation of PAHs, PCBs, PEs and PBDEs was conducted on Rxi-5Sil MS capillary column (Restek) 30 m 0.25 mm i.d. 0.25 mm film thickness. Helium was used as a carrier gas. The gas chromatographic conditions are listed in Table 2. All compounds were analysed by using time scheduled Selected Ion Monitoring (SIM) of the most intensive ion fragment of each compound (listed in Table 1). The MS ion source and the transfer line from the GC to the MS were kept at 250 C and 320 C respectively. 3. Results and discussion 3.1. Method optimisation for the assessment of PAHs in air 3.1.1. Dynamic retention efficiency Dynamic retention efficiency of five deuterated PAHs was evaluated at different flow rates and durations on the adsorbing PDMS/Tenax sampling tubes according to ASTM D4861-11 (ASTM International, 2011). Therefore, sampling tubes were spiked with a standard solution of d8-naphthalene, d10-fluorene, d10fluoranthene and d12-benzo(g,h,i)perylene in methanol. The absolute adsorbed quantity of each compound on every tube was 3000 pg. Clean air (RH 50%) was then passed through each tube at variable flow rates and sampling times, allowing an evaluation of the impact of different sampled air volumes on the dynamic retention efficiency of the deuterated compounds, as listed in Table 3. A comparison of the sampling efficiency at different sampling duration and air volume settings (Table 3) indicates that no breakthrough occurs for sampling volumes up to 732 L. In the current study, an air volume of 480 L (333 mL min1 for 24 h) was selected as suitable sampling volume for PAHs sample collection. Since this reported sampling method collects gas phase SVOCs as well as particulate matter, the air flow rate will influence the cutoff of the sampled particulate fraction in field studies. Therefore, taking into account the dimensions of the sampling tubes, the theoretical aerodynamic particle cut-off of the tubes respecting classical gravitational deposition velocity at different air flow rates
Table 3 Dynamic retention efficiency at different air volumes. Sample volume, L Sample rate, mL min1 Sampling time, h
180 124 24
176 98 30
282 98 48
445 247 30
732 254 48
d8-naphthalene d10-fluorene d10-fluoranthene d12-benzo(g,h,i)perylene
95% 93% 95% 99%
96% 94% 95% 102%
98% 99% 99% 108%
106% 102% 108% 128%
98% 100% 107% 122%
has been calculated (Brockmann, 2001) in calms (i.e. wind speed less than 0.5 m s1). According to this study, a flow rate of 50 mL min1 collects PM37, 100 mL min1 collects PM57, and 200 mL min1 collects PM80. These values imply that, when the method is applied at a flow rate of 333 mL min1, a PM fraction with an aerodynamic diameter exceeding PM80 is collected, which definitely includes all required SVOC particles. Furthermore, the measurement of SVOC compounds originating from combustion sources, such as PAHs which typically occur in a PM size fraction below 1 mm (Yang et al., 2007) won’t be influenced by the operational flow rate. However, collected SVOCs typically originating from product emissions, such as PEs and PBDEs which tend to occur both in smaller PM fractions (<2 mm) (Liu et al., 2012) and in larger (resuspended) dust fractions (Schripp et al., 2010), might be influenced by the flow rate. 3.1.2. Optimisation of TD parameters To obtain the best analytical conditions for PAH analyses in terms of sensitivity and reproducibility, the TD parameters, such as cold trap desorption temperature and desorption time, were examined and optimised. Sampling tubes, spiked with a mixture of the target PAHs, were evaluated at different cold trap desorption temperatures. Based on previous experiments (Wauters et al., 2008) and on the technical potential of the instrument, three cold trap desorption temperatures of 350, 375 and 400 C were studied. The amount of each individual compound in the tube was approximately 1000 pg. The intensity of the MS response of each compound is shown in Fig. 1. The results show that there is no significant influence of the trap desorption temperature on the MS response for compounds with lower boiling points. Furthermore, the intensity of the MS response for the compounds with higher boiling points, increases with an increasing the
350°C
Signal intensity nsity
375°C
400°C
1400000 1200000 1000000 800000
Table 2 Gas chromatographic conditions.
600000
Compound group
Oven conditions
Carrier gas conditions
PAHs
80 C (7 min) e 25 C min1 e 255 C e 5 C min1 e 325 C
PCBs
80 C (7 min) e 25 C min1 e 255 C e 5 C min1 e 300 C
PBDEs
100 300 100 255
65 kPa (7 min) e 8 kPa min1 e 121 kPa e 2 kPa min1 e 149 kPa 65 kPa (7 min) e 8 kPa min1 e 125 kPa e 2 kPa min1 e 140 kPa (1 min) 60 kPa (7 min) e 4 kPa min1 e 140 kPa (13 min) 70 kPa (5 min) e 10 kPa min1 e 120 kPa e 1 kPa min1 e 135 kPa
Benzo(g,h,i)perylene
Dibenz(a,h)anthracene
Benzo(a)pyrene
Indeno(1,2,3-cd)pyrene
Benzo(k)fluoranthene
Chrysene
Benzo(b)fluoranthene
Pyrene
Anthracene
Fluoranthene
Fluorene
Phenanthrene
Acenaphthene
Naphthalene
(7 min) e 10 C min1 e (13 min) (5 min) e 25 C min1 e e 5 C min1 e 300 C
0 Acenaphthylene
C C C C
200000 Benz(a)anthracene
PEs
400000
Fig. 1. Trap desorption temperature effect on MS response intensity for each PAH.
B. Lazarov et al. / Atmospheric Environment 79 (2013) 780e786
trap desorption temperature. This resulted in the selection of a cold trap desorption temperature of 400 C as an optimal temperature. The cold trap desorption time was optimised as well. Sampling tubes, spiked with 1000 pg of each individual PAH, were evaluated at a 400 C cold trap desorption temperature in four different desorption times: 10, 11, 13 and 15 min. As shown in Fig. 2, the MS response for the individual compounds was independent of the desorption duration. Therefore, a cold trap desorption time of 10 min was selected for this procedure. 3.2. Method performance The performance of the method was evaluated via an estimation of the linearity, the accuracy and the repeatability of the MS response. The Xcalibur 2.0 software (Thermo Fisher Scientific Inc.) was used for quantification, based on peak area. For the estimation of the linear dynamic range, five-point calibration curves were constructed in the range of 200e8000 pg, using a least-square regression analysis. As a result, correlation coefficients higher than 0.998 were obtained for all compounds (Table 4). Because of the higher expected quantity in air samples, the linear dynamic range of naphthalene was tested from 1 ng up to 250 ng. The obtained correlation coefficient in this range, also using a least-square regression analysis, was 0.9952. Limits of detection (LODs) were determined based on the analysis of tubes spiked with standard solution, and calculated as the quantity of the compound giving a signal to noise ratio of 3. The calculations were done for a sampling volume of 480 L. The LODs ranged between 0.003 ng m3 for Fluorene and 0.380 ng m3 for Benzo(k)fluoranthene. Table 4 lists the LOD for each studied PAH compound. The accuracy and repeatability of the technique were calculated from the replicate (n ¼ 6) analysis of sampling tubes, spiked within one day, with a standard reference material (SRM) ERM-AC213 of approximately 3000 pg of each compound. The recovery of the analysed compounds was within a range of 80e97%, except for benzo(b)fluoranthene which was characterised by a recovery of 146%. The higher recovery of benzo(b)fluoranthene might be caused by interference of benzo(j)fluoranthene, which was present in the SRM, but not studied in this method. The calculated relative standard deviations (RSDs) were less than 10% for most of the PAH compounds, except for benzo(k)fluoranthene e 12.6% (Table 5). The carry-over effect of each compound was investigated at three concentration levels: 250 pg, 1000 pg and 8000 pg. Six
Signal intensity
10 min
11 min
13 min
15 min
3000000 2500000 2000000 1500000
783
Table 4 Limit of detection and correlation coefficients obtained for the analysis of selected PAHs. Compound
LOD, ng m3
R2
Naphthalene Acenaphthylene Acenaphthene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benz(a)anthracene Chrysene Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(a)pyrene Indeno(1,2,3-cd)pyrene Dibenz(a,h)anthracene Benzo(g,h,i)perylene
0.033 0.250 0.005 0.003 0.005 0.194 0.097 0.106 0.008 0.008 0.026 0.380 0.062 0.540 0.095 0.450
0.9997 0.9998 0.9999 0.9999 0.9999 0.9985 0.9999 0.9998 0.9996 0.9999 0.9997 0.9997 0.9999 0.9992 0.9994 0.9998
sampling tubes, spiked with standard solution of PAHs, were analysed followed by the analysis of a set of two blank sampling tubes. The carry-over effect was then calculated as the percentage of the compound that could be quantified in the blank sampling tubes. The result indicated a carry-over effect on the investigated PAHs of less than 25% (see Table 6), independent of the spiked PAHs concentration levels. As expected, the carry-over effect increased with the increasing boiling point of the compounds, due to the tendency of these compounds to be trapped at so called “cold spots” of the system. In the analytical sequence, this small carry-over effect is taken into account by measuring a blank tube after each real sample. 3.3. Extension of the scope to other SVOCs The possibility of widening the application scope of this method for PAH sampling to the assessment of other semi-volatile organic air pollutants has been suggested by Wauters et al. (2008). Therefore, this study also explored the suitability of the PDMS/Tenax method to monitor Polychlorinated Biphenyls as well as Polybrominated diphenyl ethers and Phthalates. The examined compounds in each group are shown in Table 1. 3.3.1. PCBs Table 7 shows the limit of detection, the recovery and the repeatability of the method for PCBs. The recovery and repeatability of the technique for PCB detection was calculated from the replicate (n ¼ 4) analysis of sampling tubes, spiked with a standard solution of PCBs. The obtained results show a recovery in the range of 99e 108% and an RSD less than 5%. The linearity of the method for PCBs was evaluated by six point calibration curve at the range of 0.1e 10 ng. The calculated correlation coefficients were greater than 0.996 for each compound.
1000000 Table 5 Accuracy and repeatability of the method for PAH detection.
500000 Benzo(g,h,i)perylene
Fig. 2. Trap desorption time effect on MS response intensity of each PAH.
Dibenz(a,h)anthracene
Benzo(a)pyrene
Indeno(1,2,3-cd)pyrene
Benzo(k)fluoranthene
Chrysene
Benzo(b)fluoranthene
Benz(a)anthracene
Pyrene
Fluoranthene
Anthracene
Fluorene
Phenanthrene
Acenaphthene
Naphthalene
Acenaphthylene
0
Compound
True value, pg
Accuracy, %
RSD, %
Benz(a)anthracene Chrysene Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(a)pyrene Indeno(1,2,3-cd)pyrene Dibenz(a,h)anthracene Benzo(g,h,i)perylene
3044 3015 3005 3015 2857 2955 2719 3024
84 80 146 86 88 89 97 88
5.3 6.8 9.3 12.6 4.2 5.6 6.1 7.2
784
B. Lazarov et al. / Atmospheric Environment 79 (2013) 780e786
Table 6 Carry-over during thermal desorption of selected PAHs. Compound
250 pg
1000 pg
8000 pg
Naphthalene Acenaphthylene Acenaphthene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benz(a)anthracene Chrysene Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(a)pyrene Indeno(1,2,3-cd)pyrene Dibenz(a,h)anthracene Benzo(g,h,i)perylene
13.7% 3.0% 6.0% 5.0% 12.1% 9.9% 5.7% 6.7% 11.3% 14.2% 12.1% 10.8% 10.5% 18.0% 19.4% 19.1%
3.7% 1.0% 1.7% 1.8% 11.6% 9.1% 6.9% 5.5% 9.6% 11.6% 9.6% 10.3% 11.0% 17.7% 20.4% 18.6%
1.5% 1.9% 1.7% 2.9% 2.7% 2.6% 7.3% 7.8% 9.4% 12.0% 11.4% 12.0% 14.0% 24.6% 25.2% 24.7%
The LODs of the selected PCBs were determined by analysing sampling tubes spiked with a standard solution of the compounds in a quantity, giving a signal to noise ratio of 3. Based on the previous optimisation steps for PAHs, the calculations were done for a sampling volume of 480 L. The calculated LODs are in a range of 0.004 ng m3 for PCB-180 up to 0.225 ng m3 for PCB-118 (Table 7). Based on the above discussed method performance characteristics, the method shows good results compared to the other reported studies for the analysis of PCBs in air based on active sampling (Barro et al., 2005; Heinzow et al., 2007). 3.3.2. PBDEs Similar to the PCBs, selected method validation parameters to evaluate the applicability of the method for sampling PBDEs were assessed. The recovery and repeatability of the technique were calculated from a replicate (n ¼ 4) analysis of sampling tubes spiked with PBDEs. The linearity was estimated in the range of 3e32 ng of each compound. The obtained correlation coefficients were greater than 0.990. The LODs were determined by analysing sampling tubes spiked with amount of the compounds giving a signal to noise ratio of 3. Also for PBDEs, these first exploratory steps were calculated for a sampling volume of 480 L. The obtained LODs were in the range of 0.113e0.201 ng m3, recovery up to 98%, the repeatability was in the order of 2% (see Table 7). Table 7 Limit of detection, recovery and repeatability of the method for selected compounds. Compound
LOD, ng m3
Recovery, %
PCBs (n ¼ 4) PCB-28 PCB-52 PCB-101 PCB-118 PCB-138 PCB-153 PCB-180
0.036 0.003 0.012 0.225 0.042 0.005 0.004
105 108 99 103 101 102 105
3.0 2.1 4.0 5.1 2.5 2.9 1.1
PBDE (n ¼ 4) BDE-28 BDE-47 BDE-99
0.113 0.174 0.201
95 95 98
1.9 1.7 2.1
PEs (n ¼ 5) DMP DEP DBP BBP BEHA DEHP DNOP
0.229 0.002 0.005 0.025 0.015 0.002 0.009
109 109 99 110 125 113 95
7.8 20.0 13.7 10.6 4.7 7.3 6.4
RSD, %
Table 8 The median, lowest and highest values of measured concentrations of selected PAHs from city of Antwerp. Component
Median ng m3
Lowest ng m3
Highest ng m3
Naphthalene Acenaphthylene Acenaphthene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benz(a)anthracene Chrysene Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(a)pyrene Indeno(1,2,3-cd)pyrene Dibenz(a,h)anthracene Benzo(g,h,i)perylene
188.1 5.43 5.94 7.40 11.3 1.02 2.41 2.15 0.40 0.58 0.51 0.36 0.32 0.40 0.43 0.51
60.5 0.40 3.59 3.83 1.17 0.45 0.73 0.96 0.29 0.29 0.29 0.29 0.29 0.29 0.29 0.29
1126 55.3 20.6 25.6 47.9 1.67 9.92 13.7 0.51 3.69 1.37 0.73 0.51 0.51 0.61 0.81
3.3.3. PEs The evaluation of the applicability of the method to sample PEs was performed as well. Similar to the previous groups of chemicals, the determination of LODs was done by analysing sampling tubes, spiked with an amount of the compounds giving a signal to noise ratio of 3 and calculated for a sampling volume of 480 L. The obtained LODs were in the range of 0.001 ng m3 (DEP) e 0.069 ng m3 (DMP). Due to the higher sensitivity of the method for selected PEs, the sampling volume was reduced to 144 L in order to improve the sampling productivity. The calculated LODs for this reduced sampling volume are in range of 0.002 ng m3 (DEP)e0.229 ng m3 (DMP) (Table 7). The linearity of the method was estimated in the range of 1.5e 45 ng of each compound. The obtained correlation coefficients were greater than 0.996. The recovery and repeatability of the technique was calculated from a replicate (n ¼ 5) analysis of sampling tubes, spiked with standard mixture of PEs. The obtained recoveries for selected PEs are in range of 95% (DNOP) e 125% (BEHA) and the repeatabilities are below 13.7%, except for DEP e 20% (Table 7). 4. Field samples of SVOCs As a final step of the reported study, the optimised method was applied in the field to sample PAHs, PCBs, PEs and PBDEs in different urban environments in Flanders, Belgium. Twenty-six air samples were collected by active sampling on PDMS/Tenax and analysed for PAHs using the optimised method during three months (MarcheMay 2012) in the city of Antwerp, Belgium. Furthermore, the reproducibility of the method was tested through the analysis of parallel samples in two different labs. For each sample, the sampling time was 24 h with a total sampled air volume of 480 L. The compound concentrations were calculated using external calibration curves. Table 10 shows the median, minimum and maximum values of the PAHs concentrations detected during the three months sampling period. It should be noted that the results listed in Table 8 show consistently higher concentration levels when compared to data reported in previous PAH field studies in the same region, using the classical method for PAHs sampling on PUF foam (Du Four et al., 2005; Ravindra et al., 2006). The difference between the concentration levels tends to be higher for the more volatile (lower molecular mass) compounds than for the less volatile (higher molecular mass) ones.
B. Lazarov et al. / Atmospheric Environment 79 (2013) 780e786 Table 9 Results of analysing a parallel samples in two different labs using the optimised method. Compound
Lab1 ng m3
Lab2 ng m3
Difference from the average value
Naphthalene Acenaphthylene Acenaphthene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benz(a)anthracene Chrysene Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(a)pyrene Indeno(1,2,3-cd)pyrene Dibenz(a,h)anthracene Benzo(g,h,i)perylene
102 1.00 7.90 10.4 11.0 0.35 2.37 0.95 0.55 0.64 1.78 1.28 1.03 1.29 1.55 0.61
131 1.54 7.00 16.9 16.6 1.11 3.13 0.95 n.d.a 0.82 1.74 1.15 1.42 1.77 1.96 1.69
12% 21% 6% 24% 20% 52% 14% 0% e 12% 1% 5% 16% 16% 12% 47%
a
Concentration (ng m3) Sample 1
Sample 2
Sample 3
Average
RSD
11.3 3.11 0.89 0.34 0.38 0.30 n.d.a n.d.a n.d.a
12.1 3.40 1.19 0.37 0.39 0.40 n.d.a n.d.a n.d.a
12.1 3.54 1.31 0.38 0.37 0.45 n.d.a n.d.a n.d.a
11.8 3.4 1.13 0.36 0.38 0.38 e e e
3.9% 6.5% 19.2% 6.5% 1.9% 19.9% e e e
n.d. e not detected.
Concentration, ng m3 S1
S2
S3
63.7 8.8 3.4 437 213 65.2 583 524 116 3.5 20.6 2.1 4.4 6.0 3.8 Interferences n.d.a n.d.a n.d.a
S4
S5
S6
S7
S8
S9
S10
2.3 3.6 17.6 0.4 1.6
56.4 1380 153 9.8 12.2
14.6 69.2 95.7 7.1 4.7
n.d.a
n.d.a
n.d.a
57.8 210 283 10.0 4.1 681 n.d.a
10.9 53.8 334 12.2 3.5 100 n.d.a
454 161 1190 15.3 18.1 87.6 n.d.a
16.7 85.4 71.5 13.9 5.6 58.0 n.d.a
n.d. e not detected.
5. Conclusion
Table 10 Measured concentrations of the selected PCBs and PBDEs from city of Genk, Belgium.
a
Compound
a
This reflection is also confirmed by field test outcomes of an experiment reported by Wauters et al. (2008), which was focussed on a comparison between PDMS/Tenax sampling followed by thermo desorption and the determination of PAHs by sampling on PUF foam. The results from the analysis of the parallel samples in different labs are listed in Table 9. The obtained concentrations levels by the two labs are very similar, with a mean difference of 14% relative to the average concentration. For the PCBs and PBDEs field experiment, six samples were collected and analysed according to the optimised method. All samples were collected during 24 h at a sampling flow rate of 333 mL min1, which gives sampling volume of 480 L per sample. The samples were taken from outdoor locations in the city of Genk, Belgium. The compound concentrations were calculated using external calibration curves. Table 10 shows the measured concentrations of the compounds. Additionally, phthalates have been measured in indoor environments. The results are reported in Table 11. Based on this preliminary test, the air concentrations of the selected PBDEs in outdoor air were below the detection limit. In order to measure PBDEs in air, a different environmental setting (indoor air concentrations are expected to be higher than outdoor levels (Batterman et al., 2009)), a higher sampling volume, and/or a more sensitive detector than the single quadrupole which was applied in this study, might be needed. A more focussed optimisation strategy is currently being developed by the authors in order to fully explore the possibilities of this method for flame retardant detection.
PCB 28 PCB 52 PCB 101 PCB 118 PCB 138 PCB 153 BDE-28 BDE-47 BDE-99
Table 11 Measured concentrations of the selected phthalates from different sampling locations across Flanders, Belgium.
DMP DEP DBP BBP BEHA DEHP DNOP
n.d. e not detected.
Compound
785
The described air sampling method, based on active air sampling on PDMS/Tenax tubes, followed by thermal desorption and GCeMS analysis, showed a good efficiency for the determination of selected classes of semi-volatile compounds, such as PAHs. The method is fast, sensitive and relatively easy to apply in the field. No additional sample preparation steps are required, which significantly reduces the risk of sample contamination and sample losses as well as the required sample analysis time. Using low flow air sampling equipment makes this technique advantageous for applications in indoor environments or for personal sampling as well. The present optimisation was performed in a group-by-group approach, which resulted in small differences between the analytical procedures for each SVOC group. Try-outs have however indicated that methods for analysing PAHs, PCBs and PEs could be combined in one method for the simultaneous analysis of selected target compounds. Due to peak overlap, PBDEs congeners should be analysed using a separate analytical procedure. The applicability of the technique in real samples was tested in different urban settings. The method showed an increased sensitivity to detect PAHs compared to usually applied methods for air sampling; furthermore it also allows a reduction of the traditionally required sampling times. The applicability of the method to sample PCBs and PEs is also confirmed. However, due to the low concentrations of PBDEs in outdoor air, an improvement of the limit of detection for this group of SVOCs may be beneficial for outdoor as well as indoor air sampling. The reported optimisation steps are currently being expanded to the more thorough study of the method for flame retardant assessment in indoor environments. An increase of the air sampling volume, or the use of a more sensitive detection technique are initiatives that are being considered to further enlarge the applicability of this promising air sampling method. Acknowledgement Support was obtained from the project “Officair” funded by the 7the Framework Programme (contract 265267) under the Theme ENV 2010.1.2.2.1. Further support was obtained from the project “Inflame” funded by the 7the Framework Programme (contract 164600) under the Theme People-2010-ITN. References ASTM International, 2011. ASTM D4861-11 Standard Practice for Sampling and Selection of Analytical Techniques for Pesticides and Polychlorinated Biphenyls in Air. D4861-11 ed. D4861e11. Barro, R., Ares, S., Garcia-Jares, C., Llompart, M., Cela, R., 2005. A simple and fast micromethod for the analysis of polychlorinated biphenyls in air by sorbent enrichment and ultrasound-assisted solvent extraction. Anal. Bioanal. Chem. 381, 255e260.
786
B. Lazarov et al. / Atmospheric Environment 79 (2013) 780e786
Batterman, S., Chen, T.C., Chernyak, S., Godwin, C., 2009. Design and performance evaluation of a medium flow sampler for airborne brominated flame retardants (BFRs). J. Environ. Monit. 11, 858e866. Bidleman, T.F., Billings, W.N., Foreman, W.T., 1986. Vapor-particle partitioning of semivolatile organic compounds: estimates from field collections. Environ. Sci. Technol. 20, 1038e1043. Brockmann, J.E., 2001. Sampling and transport of aerosols. In: Baron, P.A., Willeke, K. (Eds.), Aerosol Measurement e Principles, Techniques, and Applications, second ed. John Wiley & Sons, pp. 143e170. Du Four, V.A., Janssen, C.R., Brits, E., Larebeke, N.V., 2005. Genotoxic and mutagenic activity of environmental air samples from different rural, urban and industrial sites in Flanders, Belgium. Mutat. Res. Genet. Toxicol. Environ. Mutagen. 588, 106e117. Elflein, L., Berger-Preiss, E., Levsen, K., Wünsch, G., 2003. Development of a gas chromatographyemass spectrometry method for the determination of household insecticides in indoor air. J. Chromatogr. A 985, 147e157. EN ISO 16000e13, 2008. Indoor Air e Part 13: Determination of Total (Gas and Particle-phase) Polychlorinated Dioxin-like Biphenyls (PCBs) and Polychlorinated Dibenzo-p-dioxins/dibenzofurans (PCDDs/PCDFs) e Collection on Sorbent-backed Filters. Galarneau, E., Harner, T., Shoeib, M., Kozma, M., Lane, D., 2006. A preliminary investigation of sorbent-impregnated filters (SIFs) as an alternative to polyurethane foam (PUF) for sampling gas-phase semivolatile organic compounds in air. Atmos. Environ. 40, 5734e5740. Heinzow, B., Mohr, S., Ostendorp, G., Kerst, M., Körner, W., 2007. PCB and dioxin-like PCB in indoor air of public buildings contaminated with different PCB sourcesderiving toxicity equivalent concentrations from standard PCB congeners. Chemosphere 67, 1746e1753. Krupadam, R., Bhagat, B., Khan, M., 2010. Highly sensitive determination of polycyclic aromatic hydrocarbons in ambient air dust by gas chromatography-mass spectrometry after molecularly imprinted polymer extraction. Anal. Bioanal. Chem. 397, 3097e3106. Liu, C., Morrison, G.C., Zhang, Y., 2012. Role of aerosols in enhancing SVOC flux between air and indoor surfaces and its influence on exposure. Atmos. Environ. 55, 347e356.
May, A.A., Ashman, P., Huang, J., Dhaniyala, S., Holsen, T.M., 2011. Evaluation of the polyurethane foam (PUF) disk passive air sampler: computational modeling and experimental measurements. Atmos. Environ. 45, 4354e4359. Namiesnik, J., Zabiegala, B., Kot-Wasik, A., Partyka, M., Wasik, A., 2005. Passive sampling and/or extraction techniques in environmental analysis: a review. Anal. Bioanal. Chem. 381, 279e301. Ni, Y., Kumagai, K., Yanagisawa, Y., 2007. Measuring emissions of organophosphate flame retardants using a passive flux sampler. Atmos. Environ. 41, 3235e3240. Ravindra, K., Bencs, L., Wauters, E., de Hoog, J., Deutsch, F., Roekens, E., Bleux, N., Berghmans, P., van Grieken, R., 2006. Seasonal and site-specific variation in vapour and aerosol phase PAHs over Flanders (Belgium) and their relation with anthropogenic activities. Atmos. Environ. 40, 771e785. Schauer, C., Niessner, R., Pöschl, U., 2003. Polycyclic aromatic hydrocarbons in urban air particulate matter: decadal and seasonal trends, chemical degradation, and sampling artifacts. Environ. Sci. Technol. 37, 2861e2868. Schripp, T., Fauck, C., Salthammer, T., 2010. Chamber studies on mass-transfer of di(2-ethylhexyl)phthalate (DEHP) and di-n-butylphthalate (DnBP) from emission sources into house dust. Atmos. Environ. 44, 2840e2845. Temime, B., Francois, S., Monod, A., Wortham, H., 2002. An experimental set up of a PAH vapour generator and its use to test an annular denuder. Environ. Pollut. 120, 609e616. Wania, F., Shen, L., Lei, Y.D., Teixeira, C., Muir, D.C.G., 2003. Development and calibration of a resin-based passive sampling system for monitoring persistent organic pollutants in the atmosphere. Environ. Sci. Technol. 37, 1352e1359. Wauters, E., Van Caeter, P., Desmet, G., David, F., Devos, C., Sandra, P., 2008. Improved accuracy in the determination of polycyclic aromatic hydrocarbons in air using 24 h sampling on a mixed bed followed by thermal desorption capillary gas chromatographyemass spectrometry. J. Chromatogr. A 1190, 286e293. Weschler, C.J., Nazaroff, W.W., 2008. Semivolatile organic compounds in indoor environments. Atmos. Environ. 42, 9018e9040. World Health Organization, 1989. Indoor air quality: organic pollutants. EURO Rep. Stud. 111. Berlin (West). Yang, C.R., Lin, T.C., Chang, F.H., 2007. Particle size distribution and PAH concentrations of incense smoke in a combustion chamber. Environ. Pollut. 145, 606e 615.