Polybrominated diphenyl ethers (PBDEs) in soils along a rural-urban-rural transect: Sources, concentration gradients, and profiles

lable at ScienceDirect

Environmental Pollution 159 (2011) 3666e3672

Contents lists avai

Environmental Pollution

journal homepage: www.elsevier .com/locate/envpol

Polybrominated diphenyl ethers (PBDEs) in soils along a rural-urban-ruraltransect: Sources, concentration gradients, and profiles

Bondi Gevao a,*, Abdul Nabi Ghadban a, Saif Uddin a, Foday M. Jaward b, Majed Bahloul a, Jamal Zafar a

aDepartment of Environmental Science, Kuwait Institute for Scientific Research, P.O. Box 24885, Safat 13109, Kuwaitb Environmental and Occupational Health, College of Public Health, University of South Florida, 13201 Bruce B. Downs Blvd. MDC56, Tampa, FL 33612-3805, USA

a r t i c l e i n f o

Article history:Received 30 May 2011Received in revised form12 July 2011Accepted 23 July 2011

Keywords:Polybrominated diphenyl ethersSurface soilPersistent organic pollutantsAtmospheric deposition, Urban pulse

* Corresponding author.E-mail address: [email protected] (B. Gevao).

0269-7491/$ e see front matter � 2011 Elsevier Ltd.doi:10.1016/j.envpol.2011.07.021

a b s t r a c t

This study reports concentrations of PBDEs in surface soil samples collected along a 140 km transectacross Kuwait to assess the role of urban centers as sources of persistent organic pollutants to thesurrounding environment. The SPBDE concentrations varied by a factor of w250 and ranged from 289 to80,078 pg g�1 d.w. The concentrations of PBDEs in Kuwait City were significantly higher (p< 0.01) thanthose collected from sites outside the city supporting the hypothesis that urban centers are sources ofPBDEs. The congener profiles were dominated by BDE-209, accounting for 93% of the PBDEs in the soilsamples. The concentrations of all congeners (except BDE-209) were highly correlated with percentorganic carbon (%OC) (p> 0.05) when the data from Kuwait City was omitted from the analysis. Thesefindings suggest that soil concentrations outside the urban centers were close to equilibrium with theatmosphere.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Polybrominated diphenyl ethers (PBDEs) have beenwidely usedas flame retardants in a broad range of consumer productsincluding electrical components, household appliances, furniture,textiles, etc. (Hazrati and Harrad, 2006; Harrad et al., 2008). Theywere manufactured at three levels of bromination as technicalmixtures: the penta-; Octa-; and Deca-formulations. They bearacute similarities in molecular structure, and physico-chemicalproperties, with other persistent organic pollutants (POPs) likepolychlorinated biphenyls (PCBs), polychlorinated dibenzo-p-dioxins and furans (PCDD/Fs). PBDEs have a high potential to leachout of the polymers to which they are added since they are notcovalently bonded into the fabric of the polymers (Alaee et al.,2003). The ventilation of homes and offices is therefore likely tobe a source to ambient (outdoor) air (Wilford et al., 2004). Evidenceof high indoor-outdoor gradient of PBDEs was recently reported indifferent cities (Butt et al., 2004; Wilford et al., 2004), furtherimplicating indoor sources as contributors to ambient (outdoor)concentrations.

PBDEs and other semi-volatile organic compounds primarilyenter soil via wet and dry atmospheric deposition. Anotherpathway through which these compounds penetrate soil is the

All rights reserved.

application of sewage sludge to agricultural land. In Kuwait, sewagesludge from wastewater treatment plants are dumped in thedesert and this may constitute an important pathway for thesecompounds to the terrestrial environment. Plants have beenreported to scavenge POPs from air. Therefore, plant litter isanother pathway for the entry of POPs into soil. Although vegeta-tion has been reported to constitute an important reservoir forsemi-volatile organic compounds (SOCs) (Choi et al., 2008), it is notlikely to be an important delivery mechanism to soil in Kuwaitbecause of scant vegetative cover for most of the year.

Soils and sediments are important reservoirs of POPs since theyact as repositories during periods of maximum input into theenvironment. When atmospheric concentrations decrease, espe-cially following bans/restrictions on the use of these POPs, such aswith PCBs and organochlorine pesticides, soils and sediments oncethought of as permanent sinks have been shown to act as importantsecondary sources supplying these chemicals to the atmosphere. Itis therefore important to obtain information on the levels and fateof these compounds in these reservoirs. PBDEs have been reportedin various environmental media in Kuwait including indoor air/dust(Gevao et al., 2006a,b), ambient air (Gevao et al., 2006c, 2010),sediments (Gevao et al., 2006d), biota (Gevao et al., 2011), and insewage sludge (Gevao et al., 2008). To date, however, there isvirtually no information on the levels of PBDEs in soils in theMiddle East in general or in Kuwait in particular. In 2009, thecommercial penta-bromodiphenyl ethers and Octa-bromodiphenylethers were included in Annex A (elimination) of the Stockholm

B. Gevao et al. / Environmental Pollution 159 (2011) 3666e3672 3667

Convention target list of chemicals (UNEP, 2009). Prior to this date,however, the EU had banned these chemicals in 2004 (Cox andEthymiou, 2003) and there were voluntary restrictions on theiruse in the US since 2006 (de Wit et al., 2006). This study presents,for the first time, data on PBDEs in soils collected along a north(upwind)esouth (downwind) transect across Kuwait. A secondaryobjective of the study was to investigate the “urban pulse” effectpostulated by Harrad and Hunter (2006) which suggests that urbanareas are sources of PBDEs to remote regions. In the current study,pollutants “outgassing” from Kuwait City are expected to betransported in the southerly direction which is the dominant winddirection. It is expected that soils downwind of Kuwait City willcontain higher concentrations of PBDEs compared with thosesampled upwind. It is hoped that this study will provide back-ground information on PBDEs in soil against which the effective-ness of the control measures on their continued use in consumerproducts by the Stockholm Convention can be assessed in thefuture.

2. Materials and methods

2.1. Chemicals and reagents

All solvents used in this work were of analytical grade and purchased throughVWR Scientific (USA). Silica (Baker, 100e200 mesh), Alumina and sodium sulfate(Baker) were also purchased through VWR Scientific (USA). The PBDE analyticalstandard (EO-5278) was purchased from CIL. The following PBDE congeners were inthe standard mix (28, 47, 99, 100, 153, 154, 183, and 209). Individual standards forBDEs 35 (EO-4109) and 181 (EO-4927) were purchased separately from CIL.

2.2. Sample collection

Soil samples were collected in November 2010. The mean, minimum andmaximum temperatures over sampling period were 26 �C, 21 �C and 34 �C

Fig. 1. Soil sampling locations along a major road from Iraq to Saudi A

respectively. Samples were collected at approximately 20 km intervals alonga 140 km northesouth transect from the KuwaiteIraq border through Kuwait to theSaudi border (Fig. 1). The transect was coincident with the predominant northwest-southeast wind direction across Kuwait, providing an opportunity to investigate the“urban pulse” theory postulated by Harrad and Hunter (2006). At each sampling sitesurface (0e5 cm), soil samples were collected using a stainless steel hand held auger.To ensure that samples from each site were representative, three samples werecollected in a triangular fashion about 5 m apart (approximately 10 m from the road)and pooled to give a composite sample prior to sub-sampling. The samples wereimmediately transferred into clean, solvent-rinsed, amber glass jars and stored ina cool box for transport to the laboratory. All utensils used in the collection, poolingand sub-sampling were thoroughly washed and rinsed with acetone and hexanebetween sampling sites to minimize the likelihood of cross-contamination. Eachsampling location was given a unique location reference and coordinates wererecorded using a global positioning system (GPS). All samples were given a uniquelabeling code that identified sample type, time of collection, and other importantinformation. In the laboratory, the samples were sieved onto solvent-rinsedaluminium foil to remove debris and other large particles. The sievewas cleaned andaluminium foil replaced between samples. The samples were immediately trans-ferred to clean; solvent-rinsed amber glass bottles and kept at �20 �C until analysis.

2.3. Extraction and analyses

Soil samples (w20 g) were extracted in a Soxhlet apparatus using 1:1 v/vmixture of DCM:hexane. Although the soil samples were dry, 10 g of prebaked(450 �C for 12 h) sodium sulfatewas added to remove any residual water, and spikedwith PBDE congeners (BDE 35 and BDE 181) to monitor analytical recovery. Theextracts were reduced in volume on a Turbovap�, solvent exchanged to hexane andinterfering compounds removed by column chromatography using 10 g silica and5 g alumina (and 0.5 cm anhydrous Na2SO4 at the top of the column to prevent thecolumn from contact with air). The compounds of interest were eluted with 100 ml1:1 mixture of hexane:DCM. The eluent was then blown down on a Turbovap�concentrator, transferred to 2 ml vials and blown down under a gentle stream ofnitrogen. 50 ml of dodecane was added during this blow down stage to ensure thesamples did not dry out. The samples were then transferred to 100 ml glass inserts,and spiked with Mirex (10 ml of 10 ng/ml) internal standard, used for volumecorrection and to adjust for variations in instrument response. The sample extractswere analyzed on a Shimadzu GC 2010 (Shimadzu, Tokyo, Japan) gas chromatograph

rabia. Arrow shows the dominant wind direction across Kuwait.

B. Gevao et al. / Environmental Pollution 159 (2011) 3666e36723668

using splitless injection on a 15 m DB5-ms column (0.25 mm i.d., 0.25 mm filmthickness) and helium as carrier gas. The oven program was 150 �C for 1 min,ramped at 20 �Cmin�1 to 250 �C, 4 �Cmin�1 to 290 �C, and held for 25 min. This gaschromatograph was coupled to a Shimadzu 2010 Mass Selective Detector operatedin electron capture negative chemical ionization (ECNCI) using selected ion moni-toring (SIM), with methane as reagent gas. The ions m/z 79 and 81 were monitoredfor PBDEs and 402/404 for Mirex. Operating conditions were as follows: injectortemperature was set at 250 �C; ion source 230 �C; quadrupole 106 �C; transfer line300 �C. Identification and quantification was carried out against 5 calibrationstandards of known concentrations.

2.4. QA/QC

An analytical blank was processed for every 5 samples. In the absence ofappropriate blanks, anhydrous Na2SO4 (previously baked at 450 �C) was spiked withsurrogates and taken through the entire analytical process as for actual samples toserve as matrix blanks. A peak was positively identified if it was within�0.05 min ofthe retention time in the calibration standard and quantified only if the S/N� 3 andthe ratio of the target ion to its qualifier ion was within �20% of the standard value.The PBDEs present in the appropriate blanks were subtracted from those in thesample extracts. The method detection limits (MDLs) were calculated as the meanblankþ 3� SD. The method detection limits ranged from 110 pg (BDE 28) to 1.1 ng(BDE-209). These were converted to concentrations by dividing by the averageweight of soil (20 g) analyzed in the study, resulting in detection limits of 5.5 pg g�1

(BDE-28) to 55 pg g�1 (BDE-209). Average recoveries (%) for surrogates spiked insamples were between 70 (�10 SD) for BDE 35 and 84 (�5 SD) for BDE 181. Externalrecoveries for the entire method were carried out by spiking six pre-extractedthimbles containing 10 g anhydrous Na2SO4 that had been previously baked at450 �C for 12 h, with a working standard containing all PBDE congeners. Theextraction and work-up procedures were identical to those for actual samples. Therecoveries were found to be 90�12% for all congeners.

The accuracy and precision of the analytical method was further assessed byreplicate analyses (n¼ 6) of a certified indoor dust reference material (SRM 2585).The results compared very well with the certified values of all the congeners (Fig. 2).

3. Results and discussion

The concentrations of PBDEs were calculated by dividing theamounts by the actual weight of soil extracted after adjusting formoisture. A total of 8 PBDE congeners was regularly detected insamples and quantified. These are: BDE 28 (2,4,40-TriBDE); BDE-47(2,20,4,40-Tetra-BDE); BDE 99 (2,20,4,40,5-Penta-BDE); BDE 100(2,20,4,40,6-Penta-BDE); BDE-153 (2,20,4,40,5,50-Hexa-BDE); BDE-154(2,20,4,40,5,60-Hexa-BDE); BDE-183 (2,20,3,4,40,50, 6-Hepta-BDE) andBDE-209 (2,20,3,304,40,5,50,6,60-Deca-BDE). In the discussion thatfollows, SPBDE refers to the sum of all the congeners measured in

Fig. 2. Comparison between average (n¼ 6) concentrations of PBDE congeners in SRM2585 in this study and their certified values. The error bars represent standarddeviations.

this study whereas S7PBDEs refers to the concentrations of thepenta-congeners (BDEs 28, 47, 99, 100, 153, 154, and 183).

The concentrations of PBDEs in the samples are summarized inFig. 3 and the congener specific concentrations are given in Tables 1and 2 on a dry weight, and organic carbon normalized concentra-tion basis respectively. The data in Fig. 3 is arranged along thenorthesouth direction from the KuwaiteIraq border to theKuwaiteSaudi Arabia border which is coincident with thepredominant wind direction in the country; the natural gradientfor pollutant input from long-range atmospheric transport. TheSPBDE concentrations varied by a factor of w250 and ranged from289 to 80,078 pg g�1 d.w., whereas the concentrations of theS7PBDEs (the total contribution of the congeners that arepredominantly present in the penta-BDE formulations) rangedfrom 26 to 6132 pg g�1. The concentrations of PBDEs in Kuwait City(shaded portion) were significantly higher (p< 0.01) than thosecollected from sites outside the city evidenced by a clear pulse inthe profile given in Fig. 3. This magnitude of the pulse in concen-tration within the city, estimated as the ratio of the averageconcentration in samples collected from the city (sites 5, 6 and 7), tothe average concentration at all the other locations outside the cityis ca 5 and 24 for SPBDE and S7PBDEs (penta-BDEs) respectively.When organic carbon normalized soil concentrations are used themagnitude of the pulse is 3.3 and 7.6 for SPBDEs and S7PBDEsrespectively. This urban pulse for SPBDE is similar to that reportedby Harrad and Hunter (2006) for a similar study in Birmingham UKalthough the magnitude of the pulse for S7PBDEs (penta-conge-ners) is much higher. It has been hypothesized that urban centersare net sources of pollutants to the surrounding areas becausepollutants emitted from urban areas are transported in the atmo-sphere and deposited in remote locations. Since the predominantwind direction in the study area is from the northwest (upwind) tothe southeast (downwind) across Kuwait City, it is plausible toexpect higher concentrations of PBDEs in soil downwind of theurban source compared to those sampled upwind. The averageconcentrations of PBDEs in soil samples collected downwind ofKuwait City (sites 8e11) were 1.5 times the average concentrationscollected upwind (sites 1e4) of the City. This finding lends supportto the hypothesis that urban centers are sources of pollutants tosurrounding areas.

Table 3 compares the concentrations measured in this studywith those reported in other parts of the world where urban and/orbackground soils have been analyzed for PBDEs. In order to

Fig. 3. S7PBDE and BDE-209 concentrations in soil samples collected along the tran-sect across Kuwait.

Table 1Concentration of PBDEs (pg g�1 d.w.) in soil samples along the IraqeKuwaiteSaudi Arabia transect.

Site Description Congeners

28 47 100 99 154 153 183 209 SPBDEs 47:99ratio

1 100 km Upwind of Kuwait City Center 1.1 12.2 1.0 7.4 1.1 1.9 1.3 424.6 450.6 1.652 80 km Upwind of Kuwait City Center 1.1 11.4 1.5 9.4 2.0 1.2 1.7 2118.6 2146.8 1.213 60 km Upwind of Kuwait City Center 2.1 62.3 22.5 130.3 15.9 23.1 3.8 7340.1 7600.0 0.484 40 km Upwind of Kuwait City Center 3.7 35.5 2.7 19.3 3.5 1.7 0.8 362.7 430.0 1.845 20 km Upwind of Kuwait City Center 3.3 68.7 13.9 107.9 8.2 11.8 12.7 2894.8 3121.3 0.646 Kuwait City Center 63.4 2099.6 392.0 2956.5 224.8 294.5 101.7 60,464.8 66,597.2 0.717 20 km Downwind of Kuwait City Center 13.8 321.6 78.1 584.2 71.0 79.6 31.7 78,898.5 80,078.5 0.558 40 km Downwind of Kuwait City Center 4.0 81.4 25.0 149.2 14.6 22.0 18.0 15,280.1 15,594.2 0.559 60 km Downwind of Kuwait City Center 1.9 21.8 2.4 13.8 1.5 2.7 3.7 241.7 289.5 1.5810 80 km Downwind of Kuwait City Center 0.5 8.0 1.5 14.0 2.0 1.7 2.7 3449.2 3479.6 0.5811 100 km Downwind of Kuwait City Center 3.4 46.0 7.8 40.9 7.4 5.0 2.4 2864.0 2977.0 1.12

B. Gevao et al. / Environmental Pollution 159 (2011) 3666e3672 3669

meaningfully compare the data between studies, the concentra-tions of the congeners frequently reported in the literature (definedin this study as S7PBDEs) are given with the total concentrationsreported in the respective studies. The concentrations of theS7PBDEs in background soils in this study (24.7e296 pg g�1 d.w.)are higher than those reported in Harbin City, China(2.45e55.9 pg g�1 d.w.) (Xu et al., 2009); comparable to rural soilsin Birmingham UK (73.5e285) (Harrad and Hunter, 2006), flood-plain soil of the Saginaw River Watershed, Michigan, USA(20e830 pg g�1 d.w.), mean background soils in the UK(400 pg g�1 d.w.) (Hassanin et al., 2004) and soils from a botanicalgarden in Manaus, Brazil (434 pg g�1 d.w.) (Thorenz et al., 2010).The S7PBDEs in soil samples from Kuwait City(226e6031 pg g�1 d.w.) is also comparable to that reported forurban soil samples from Birmingham, UK (401e3890 pg g�1 d.w.)(Harrad and Hunter, 2006), in urban soils from Mainz, Germany(1043 pg g�1 d.w.) (Thorenz et al., 2010); Pujalt, Spain(6100 pg g�1 d.w.) (Eljarrat et al., 2008), and in Shanghai, China(23.6e3799 pg g�1 d.w.). The concentrations reported in this studyand those cited above, represent the lower range of PBDE values insoils around the world as much higher values have been recordedin soils from electronic recycling activities (Wang et al., 2005; Caiand Jiang, 2006; Leung et al., 2007; Yang et al., 2008; Ma et al.,2009) and agricultural soils that have been amended with sewagesludge (Matscheko et al., 2002; Eljarrat et al., 2008).

3.1. Technical mixtures in soil samples

The congener mixture in the soil samples suggests that twomain technical formulations can be identified: deca- and penta-mixtures (Fig. 4). The congener profiles were dominated by BDE-209, the dominant congener in the technical deca-formulation,

Table 2Concentration of PBDEs (pg g�1 OC) in soil samples collected along a transect across Kuw

Site Description Congener

28 47 100

1 100 km Upwind of Kuwait City Center 147.5 1596.2 131.2 80 km Upwind of Kuwait City Center 201.9 2054.8 264.3 60 km Upwind of Kuwait City Center 161.8 4767.5 1721.4 40 km Upwind of Kuwait City Center 400.6 3804.0 292.5 20 km Upwind of Kuwait City Center 229.1 4793.9 969.6 Kuwait City Center 1896.2 62,818.0 11,729.7 20 km Downwind of Kuwait City Center 192.8 4489.9 1090.8 40 km Downwind of Kuwait City Center 222.8 4553.6 1396.9 60 km Downwind of Kuwait City Center 167.6 1899.5 209.10 80 km Downwind of Kuwait City Center 46.6 738.2 135.11 100 km Downwind of Kuwait City Center 472.7 6448.7 1096.

accounting for 93% (range, 83e99%) of the PBDEs in the soilsamples. This was followed in decreasing order of magnitude byBDE-47 (range, 0.2e8.3%; mean, 2.5%), BDE 99 (range, 0.4e4.8%;mean 2.2%) with the other congeners individually contributing lessthan 1% to the total concentrations in each sample. The concen-trations of BDE-209 ranged from 242 to 78,899 pg g�1 with a meanof 15,849 pg g�1 d.w. The BDE-209 profile is markedly similar to theS7PBDE profile in that the levels in background soils were generallylow with a pulse in Kuwait City (see Fig. 3). If BDE-209 is discardedfrom the analyses and the congener mixture is normalized toS7PBDEs, it becomes apparent that the penta-formulation is also animportant source of PBDEs in Kuwait (Fig. 5). The congener distri-butions expressed as a percent of S7PBDEs (excluding BDE-209), aregiven in Fig. 5 for samples collected upwind of Kuwait City, withinKuwait City, and downwind of Kuwait City. Also plotted in Fig. 5 arethe percent contributions of these congeners in two penta-formu-lations (Bromkal 70-5DE and DE-71) for comparison. Althoughsubtle differences exist, the congener profiles in soil samples areremarkably similar to those in both penta-formulations assumingthat the congener composition reported by La Guardia et al. (2006)is representative of penta-BDE technical mixtures. Hassanin et al.(2004) argued that this similarity in the soil profiles to that in thetechnical penta-formulations indicate that the transfer of conge-ners from source-sink operate with similar efficiencies across thepenta-PBDE congeners and that minimal weathering would haveoccurred during atmospheric transport or within the soil. Thisindeed may be the case in this study as the distance between themain urban source to the furthermost soil sampling location wasless than 100 kmwhich may not be sufficient transport distance forsignificant changes in composition to occur.

The subtle deviations between the observed congener profilesin soil samples and those in the technical penta-formulations may

ait.

99 154 153 183 209 SPBDEs

6 966.6 145.9 247.1 164.1 55,511.3 58,910.27 1691.4 360.5 211.4 302.5 381,465.9 386,553.14 9966.4 1213.6 1766.6 288.7 561,499.6 581,385.70 2071.3 374.8 182.6 83.9 38,856.9 460,65.98 7533.5 572.0 822.4 888.6 202,082.8 217,892.11 88,456.5 6725.4 8811.5 3041.9 1,809,084.7 1,992,563.38 8155.4 990.5 1110.6 442.1 1,101,356.2 1,117,828.23 8347.7 819.2 1230.3 1007.2 855,087.7 872,664.73 1199.6 130.1 239.2 318.9 210,73.4 25,237.66 1281.9 186.9 157.6 246.7 316,671.7 319,465.25 5737.9 1041.6 699.8 336.5 401,331.2 417,164.9

Table 3Comparison of PBDE soil concentrations (pg g�1 d.w.) between this study and other studies.

Location Congener

BDE-28 BDE-47 BDE-99 BDE-100 BDE-153 BDE-154 BDE-183 BDE-209 SPBDEs Reference

Birmingham, UK (urban) 13e116 101e909 117e1710 38e392 41.8e410 29.7e357 n/a n/a 401e3890b Harrad and Hunter,2006

Birmingham, UK (rural) 4.3e10.8 34.6e69.0 88.1e131 17.1e25.2 13.6e25.6 14.1e19.8 73.2e285b Harrad and Hunter,2006

UK grassland 12e21 7e520 78e3200 8e470 19e600 8e240 10e900 n/a 15e5000a Hassanin et al., 2004UK woodland 8e200 50e1400 190e3200 11e360 38e1200 14e420 10e7000 n/a 75e5600a Hassanin et al., 2004Norwegian woodland 8e49 12e860 63e1400 18e230 11e270 13e310 9e130 90e2600a Hassanin et al., 2004Harbin City, China <d.l.e6.3 <d.l.e29.9 <d.l.e29.5 <d.l.e7.24 <d.l.e5.01 <d.l.e2.35 <d.l.e8.60 <d.l.e1750 <d.l.e55.9c Xu et al., 2009Shiawassee floodplain,

USA<10e200 90e6290 100e5780 200e1050 <10e490 <10e520 600e41,170 210e14,670 Yum et al., 2008

Saginaw floodplain, USA <10e20 10e680 10e520 <10e160 <10e70 <10e60 <40e19,220 30e1400 Yum et al., 2008Saginaw floodplain, USA <10e20 10e420 <10e270 <10e90 <10e20 <10e20 <40e2160 20e830 Yum et al., 2008Pujalt, Spain 690 630 1080 940 930 1870 14,600 6100d Eljarrat et al., 2008Mainz (urban), Germany 18 490 91 92 290 43 19 760 1043c Thorenz et al., 2010Manaus, Brazil <d.l. <d.l. 71 <d.l. 330 33 <d.l. 500 434c Thorenz et al., 2010Bratislava (urban),

Slovakia20 65 34 <d.l. 16 26 <d.l. <d.l. 161c Thorenz et al., 2010

Shanghai (urban), China <d.l.e20.1 3.2e230 0.17e137 0.91e101 <d.l.e28.4 <d.l.e10.4 <d.l.e92.9 1.29e2910 23.6e3799e Jiang et al., 2010Kuwait-urban (This

study)3.3e63.4 68.7e2099.6 107e2956.5 13.9e392 11.8e294.5 8.2e22.4 12.7e101.7 3121.3e66,597 213e6031b

Kuwait-rural (This study) 0.5e4.0 8.0e81.4 7.4e149.2 1.0e25.0 1.2e23.1 1.1e15.9 0.8e18.0 289e15,594 24.7e296b

a Sum of congeners 47, 99, 100, 153, 154.b Sum of congeners 28, 47, 99, 100, 153, 154.c Sum of 17, 28, 47, 66, 100, 99, 154, 153, and 183.d Sum of 47,100, 99, 154, 153, 183.e Sum of 28 congeners.

B. Gevao et al. / Environmental Pollution 159 (2011) 3666e36723670

be due to several factors including the range of commercialmixtures used in treating products. Two penta-formulations arecommercially available: DE-71, produced by the Great LakesChemicals and the European formulation Bromkal 70-5DE. Bothformulations have similar congenermixtures except that DE-71 has10% higher levels of BDE 99 than BDE-47 (La Guardia et al., 2006).Also, the composition of congeners in technical mixtures from thesame manufacturer may not be constant between batches. Forinstance, the percent composition of the two major congeners,BDE-47 and BDE-99 in Bromkal 70-5DE reported by Sjödin et al.(1998) are 37% and 35% respectively while Rayne and Ikonomou(2002) report 51% and 34%, and La Guardia et al. (2006) report38% and 48% for an identical mixture.

The differences in physico-chemical properties between thecongeners may be relevant in explaining the subtle differences

Fig. 4. Contribution of the commercial mixtures to the SPBDE concentration measuredin soil samples collected along the transect across Kuwait.

observed in the soil profiles and the technical penta-mixtures. Forexample, the BDE-47:-99 ratio in the soil samples range from 0.63in Kuwait City to 0.92 in background soils, slightly lower than theratios in the penta-mixtures (1.05, Bromkal 70-5DE (30); 0.95 LaGuardia et al. (29)) but similar to the ratio in DE-71 the penta-formulation manufactured in the United States (0.78, La Guardiaet al., 2006). Similar 47:99 ratios have been previously reportedin soil samples (e.g. Hassanin et al., 2004; Harrad and Hunter, 2006;Duan et al., 2010). It has been argued that a higher KOA for BDE 99compared to BDE-47 results in a preferentially higher atmosphericdeposition and better soil retention of BDE 99 resulting in a low47:99 in soils relative to their ratio in the technical penta-mixtures(Hassanin et al., 2004; Harrad and Hunter, 2006; Duan et al., 2010).

To further illustrate how differences in the physico-chemicalproperties of PBDE congeners would have resulted in the observedchanges in congener composition in soil samples, we plotted theratio (“R”) of the sum of BDE-47þ BDE-99þ BDE-100 to the sum ofBDE-153þ BDE-154 along the transect (Fig. 6). This ratio wasinitially proposed by Hites (2004) to examine changes in congener

Fig. 5. Congener mixture expressed as a percent of penta-BDE congeners.

B. Gevao et al. / Environmental Pollution 159 (2011) 3666e3672 3671

composition in guillemot eggs collected between 1980 and 2000from the Great Lakes catchment. It was speculated that the positivecorrelation between the ratio and year of sample collection mayhave been due to possible changes in the composition of thetechnical mixtures used in the USA. This ratio has subsequentlybeen used to assess the relative contributions of the penta- andocta-PBDE technical mixtures in samples (Yum et al., 2008; Jianget al., 2010). In this study, the ratio (“R”) in soil along the transectmay be interpreted as representing the degree of weathering of thetechnical formulation following its release from their sourceregions. This ratio (“R”) in the technical penta-mixture lies between10 (DE-71) and 12 (Bromkal 70-5DE). The ratio is expected todecrease with distance from source regions. Following their releaseto the atmosphere, the fate of different congeners will be dictatedby their properties. In the atmosphere, semi-volatile organiccompounds undergo temperature dependent gas-particle parti-tioning with the heavier congeners preferentially partitioning ontoparticles. They are consequently more likely to be preferentiallyscavenged and deposited from a given air mass than lowermolecular weight congeners which may preferentially be in the gasphase. As a result of their association with particles, highermolecular weight congeners are more likely deposited closer tosource region (urban centers) and also preferentially retained insoils relative to the lower molecular weight congeners. Theseprocesses of preferential deposition and/or retention in soil medi-ated by the physico-chemical properties will ultimately result inthe fractionation of the congeners which will result in the alter-ation of this ratio from that in the penta-formulation. It will beexpected that the magnitude of “R” in soil will decrease withdistance from potential source regions as heavier congeners arepreferentially retained in soil relative to the more volatile conge-ners. In this study, the ratio (“R”) decreased in the following order:Kuwait city (range, 9.5e11; mean, 10.4)> downwind of the urbansource (range, 6.2e7.6; mean, 7.3)> ratio upwind of Kuwait City(range, 3.4e5.5, mean, 5.6). The lowest value of “R” in this study

Fig. 6. Ratio of the sum of the concentrations of BDE-47þ -99þ -100 divided by thesum of BDE-153 plus BDE-154 in soil samples collected along the transect acrossKuwait.

(3.4) was from the northernmost site upwind of Kuwait City whichmostly receives its pollutant input from long-range transport fromother regions. This is followed by a gradual increase in “R” toa maximum in Kuwait City (mean 10.4) and a gentle decrease insamples downwind of the urban source. The inflection point in theprofile occurs in the city center with the maximum PBDE soilconcentrations. The decrease in the magnitude of “R” in the soilsamples collected downwind of urban source (Kuwait City) isgentler because the soil samples downwind will receive freshinputs of congeners outgassing from the city.

3.2. Soil organic matter

It is often suggested that soil organic matter plays a significantrole in the distribution of POPs in soils/sediments because theyare hydrophobic contaminants. Consequently, the concentrationsderived from soil/sediment analyses are often normalized to theorganic carbon (OC) content. In this study, the soil organic carboncontent of soils ranged from 0.4 to 2.5% except in Kuwait City wherethe values were between 3 and 7%. The organic carbon content ofthe soils (especially soil samples collected outside Kuwait City) aregenerally lower than those reported in other parts of the world(Hassanin et al., 2004; Harrad and Hunter, 2006; Yum et al., 2008;Thorenz et al., 2010). The higher OC content of soils within the Citymay be due to mulching of soils for greenery operations. Congenerspecific OC normalized soil concentrations in this study are given inTable 2 and are comparable to those reported in urban and remotesoils from Birmingham, UK (Harrad and Hunter, 2006) andNorwegian woodland soils (Hassanin et al., 2004) but lower thanthose reported for UK grassland and woodland soils (Hassaninet al., 2004). The %OC concentrations in soils tracked the S7PBDE(excluding 209) concentrations remarkably well (Fig. 7). WhenBDE-209 was included, however, differences in profiles wereevident especially in samples collected within Kuwait City wherethe OC content was significantly higher than in the other samples.Congener specific soil concentrations were highly correlated with %OC (Pearson correlation coefficients, R2, ranging between 0.7 and0.82, p> 0.05) when the data from Kuwait City was omitted fromthe analysis. However, when the entire dataset was included in theassessment, none of the congeners was significantly correlatedwith OC. These findings suggest that the soil concentrations outsidethe urban centers were close to equilibriumwith the atmosphere. It

Fig. 7. Concentrations of S7PBDEs and percent organic carbon content in soils alonga transect across Kuwait.

B. Gevao et al. / Environmental Pollution 159 (2011) 3666e36723672

can be further deduced from the data that the concentrations ofPBDEs in soils within Kuwait City were far from a steady state withair and are a function of primary emissions.

Acknowledgments

We are grateful to the management of Kuwait Institute forScientific Research for funding this research.

References

Alaee, M., Arias, P., Sjodin, A., Bergman, A., 2003. An overview of commercially usedbrominated flame retardants, their applications, their use patterns in differentcountries/regions and possible modes of release. Environ. Int. 29, 683e689.

Butt, C.M., Diamond, M.L., Truong, J., Ikonomou, M.G., ter Schure, A.H., 2004. Spatialdistribution of polybrominated diphenyl ethers in Southern Ontario asmeasured in indoor and outdoor window organic films. Environ. Sci. Technol.38, 724e731.

Cai, Z., Jiang, G., 2006. Determination of polybrominated diphenyl ethers in soilfrom e-waste recycling site. Talanta 70, 88e90.

Choi, S.D., Li, H., Su, Y., Gevao, B., Harner, T., Staebler, R.M., Wania, F., 2008. Depletionof gaseous polycyclic aromatic hydrocarbons by a forest canopy. Atmos. Chem.Phys. 8, 4105e4113.

Cox, P.M., Ethymiou, P., 2003. Directive 2003/11/EC of the European parliament andof the council of February 6, 2003 amending for the 24th time Council Directive76/669/EEC relating to restrictions on the marketting and use of certaindangerous substances and preparations (pentabromodiphenyl ether, octabro-modiphenyl ether). Off. J. Eur. Union 42, 45e46.

de Wit, C., Alaee, M., Muir, D.C., 2006. Levels and trends of brominated flameretardants in the Arctic. Chemosphere 64, 209e233.

Duan, Y.-P., Meng, X.-Z., Yang, C., Pan, Z.-Y., Chen, L., Yu, R., Li F., T., 2010. Poly-brominated diphenyl ethers in background surface soils from the Yangtze RiverDelta (YRD), China: occurrence, sources, and inventory. Environ. Sci. Pollut. Res.17, 948e956.

Eljarrat, E., Marsh, G., Labandeira, A., Barcelo, D., 2008. Effect of sewage sludescontaminated with polybrominated diphenylethers on agricultural soils. Che-mosphere 71, 1079e1086.

Gevao, B., Al-Bahloul, M., Al-Ghadban, A.N., Al-Omair, A., Ali, L., Zafar, J., Helaleh, M.,2006a. House dust as a source of human exposure to polybrominated diphenylethers in Kuwait. Chemosphere 64, 603e608.

Gevao, B., Al-Bahloul, M., Al-Ghadban, A.N., Ali, L., Al-Omair, A., Helaleh, M., Al-Matrouk, K., Zafar, J., 2006b. Polybrominated diphenyl ethers in indoor airKuwait: implications for human exposure. Atmos. Environ. 40, 1419e1426.

Gevao, B., Al-Omair, A., Sweetman, A., Al-Bahloul, M., Al-Ali, L., Helaleh, M., Zafar, J.,2006c. Passive-sampler derived air concentrations for polybrominated diphenylethers and polycyclic aromatic hydrocarbons in Kuwait. Environ. Toxicol. Chem.25, 1496e1502.

Gevao, B., Beg, M.U., Al-Ghadban, A.N., Al-Omair, A., Helaleh, M., Zafar, J., 2006d.Spatial distribution of polybrominated diphenyl ethers in coastal marine sedi-ments receiving industrial and municipal effluents in Kuwait. Chemosphere 62,1078e1086.

Gevao, B., Jaward, F.M., Al-Bahloul, M., Ud din, S., Beg, M.U., Zafar, J., 2011. Poly-brominated diphenyl ethers in three commercially important fish from theNorthwestern Arabian Gulf: occurrence, concentration, and profiles. Arch.Environ. Contam. Toxicol. 60, 636e642.

Gevao, B., Jaward, F.M., MacLeod, M., Jones, K.C., 2010. Diurnal fluctuations inpolybrominated diphenyl ether concentrations during and after a severedust storm episode in Kuwait City, Kuwait. Environ. Sci. Technol. 44,8114e8120.

Gevao, B., Muzaini, S., Helaleh, M., 2008. Occurrence and concentrations of poly-brominated diphenyl ethers in sewage sludge from three wastewater treatmentplants in Kuwait. Chemosphere 71, 242e247.

Harrad, S., Hunter, S., 2006. Concentrations of polybrominated diphenyl ethers inair and soil on a rural-urban transect across a major UK conurbation. Environ.Sci. Technol. 40, 4548e4553.

Harrad, S., Ibarra, C., Diamond, M., Melymuk, L., Robson, M., Douwes, J., Roosens, L.,Dirtu, A.C., Covaci, A., 2008. Polybrominated diphenyl ethers in domestic indoordust from Canada, New Zealand, United Kingdom and United States. Environ.Int. 34, 232e238.

Hassanin, A., Breivik, K., Meijer, S.N., Steinnes, E., Thomas, G.O., Jones, K.C., 2004.PBDEs in European background soils: levels and factors controlling theirdistribution. Environ. Sci. Technol. 38, 738e745.

Hazrati, S., Harrad, S., 2006. Causes of variability in concentrations of poly-chlorinated biphenyls and polybrominated diphenyl ethers in indoor air.Environ. Sci. Technol. 40, 7584e7589.

Hites, R.A., 2004. Polybrominated diphenyl ethers in the environment and inpeople: a meta-analysis of concentrations. Environ. Sci. Technol. 38, 945e956.

Jiang, Y., Wang, X., Zhu, K., Wu, M.S., Sheng, G., Fu, J., 2010. Occurrence, composi-tional profiles and possible sources of polybrominated diphenyl ethers in urbansoils of Shanghai, China. Chemosphere 80, 131e136.

La Guardia, L., Hale, R.C., Harvey, E., 2006. Detailed polybrominated diphenyl ether(PBDE) congener composition of the widely used penta-, octa-, and deca-PBDEtechnical flame-retardant mixtures. Environ. Sci. Technol. 40, 6247e6254.

Leung, A.O., Luksemburg, W.J., Wong, A.S., Wong, M.H., 2007. Spatial distribution ofpolybrominated diphenyl ethers and polychlorinated dibenzo-p-dioxins anddibenzofurans in soil and combusted residue at Guiyu, an electronic wasterecycling site in southeast China. Environ. Sci. Technol. 41, 2730e2737.

Ma, J., Addink, R., Yun, S., Cheng, J., Wang, W., Kannan, K., 2009. Polybrominateddibenzo-p-dioxins/dibenzofurans and polybrominated diphenyl ethers in soil,vegetation, workshop-floor dust, and electronic shredder residue from anelectronic waste recycling facility and in soils from a chemical industrialcomplex in eastern China. Environ. Sci. Technol. 43, 7350e7356.

Matscheko, N., Tysklind, M., de Wit, C., Bergek, S., Andersson, R., Sellstrom, U., 2002.Application of sewage sludge to arable land-soil concentrations of poly-brominated diphenyl ethers and polychorinated dibenzo-p-dioxins, dibenzo-furans, and biphenyls, and their accumulation in earthworms. Environ. Toxicol.Chem. 21, 2515e2525.

Rayne, S., Ikonomou, M.G., 2002. Reconstructing source PBDE congener profilesfrom SPMDs in the Fraser River, British Columbia, Canada: comparison tocommercial mixtures. Environ. Toxicol. Chem. 21, 2292e2300.

Sjödin, A., Jakobsson, E., Kierkegaard, A., Marsh, G., Sellstrom, U., 1998. Gas chro-matographic identification and quantification of polybrominated diphenylethers in a commercial product, Bromkal 70-5DE. J. Chromatogr. A 822, 83e89.

Thorenz, U.R., Bandowe, B.A.M., Sobocka, J., Wilcke, W., 2010. Method optimizationto measure polybrominated diphenyl etteh (PBDE) concentrations in soils ofBratislava, Slovakia. Environ. Pollut. 158, 2208e2217.

UNEP, 2009. The nine new POPs under the Stockholm convention. Retrieved March12, 2010, from. http://chm.pops.int/Programmes/NewPOPs/The9newPOPs/tabid/672/language/en-US/Default.aspx.

Wang, D., Cai, Z., Jiang, G., Leung, A., Wong, M.H., Wong, W.K., 2005. Determinationof polybrominated diphenyl ethers in soil and sediment from an electronicwaste recycling facility. Chemosphere 60, 810e816.

Wilford, B.H., Harner, T., Zhu, J., Shoeib, M., Jones, K.C., 2004. A passive samplingsurvey of polybrominated diphenyl ether flame retardants in indoor andoutdoor air in Ottawa, Canada: implications for sources and exposure. Environ.Sci. Technol. 38, 5312e5318.

Xu, W., Nanqi, R., Hong, Q., Wanli, M., Yifan, L., 2009. Levels and distribution ofbrominated flame retardants in the soil of Harbin in China. J. Environ. Sci. 21,1541e1546.

Yang, Z.Z., Zhao, X.R., Zhao, Q., Qin, Z.F., Qin, X.F., Xu, X.B., Jin, Z.X., Xu, C.X.,2008. Polybrominated diphenyl ethers in leaves and soil from typical elec-tronic waste polluted area in South China. Bull. Environ. Contam. Toxicol. 80,340e344.

Yum, S.H., Addink, R., McCabe, J.M., Ostaszewski, A., Mackenzie-Taylor, D.,Taylor, A.B., Kannan, K., 2008. Polybrominated diphenyl ethers and poly-brominated biphenyls in sediment and floodplain soils of the Saginaw riverwatershed, Michigan, USA. Arch. Environ. Contam. Toxicol. 55, 1e10.