Chemosphere 83 (2011) 1360–1365

Contents lists available at ScienceDirect

Chemosphere

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

‘‘Novel’’ brominated flame retardants in Belgian and UK indoor dust:Implications for human exposure

Nadeem Ali a, Stuart Harrad b, Emma Goosey b, Hugo Neels a, Adrian Covaci a,c,⇑a Toxicological Center, University of Antwerp, Universiteitsplein 1, 2610 Wilrijk–Antwerp, Belgiumb Division of Environmental Health and Risk Management, School of Geography, Earth and Environmental Sciences, University of Birmingham, Birmingham B15 2TT, UKc Laboratory for Ecophysiology, Biochemistry and Toxicology, University of Antwerp, Groenenborgerlaan 171, 2020 Antwerp, Belgium

a r t i c l e i n f o

Article history:Received 24 November 2010Received in revised form 2 February 2011Accepted 27 February 2011Available online 31 March 2011

Keywords:‘‘Novel’’ brominated flame retardants(NBFRs)Human exposureToddlerIndoor dustPolybrominated diphenyl ethers (PBDEs)

0045-6535/$ - see front matter � 2011 Elsevier Ltd. Adoi:10.1016/j.chemosphere.2011.02.078

⇑ Corresponding author at: Toxicological Center, Unsiteitsplein 1, 2610 Wilrijk–Antwerp, Belgium.

E-mail address: adrian.covaci@ua.ac.be (A. Covaci)

a b s t r a c t

Concentrations of several ‘‘novel’’ brominated flame retardants (NBFRs) are reported in indoor dust sam-ples from Belgian houses (n = 39) and offices (n = 6) and from day-care centers and schools in the WestMidlands of the UK (n = 36). Using a GC-ECNI/MS method, the following NBFRs were quantified: decab-romodiphenyl ethane (DBDPE) (range <20–2470 ng g�1), 1,2-bis(2,4,6-tribromophenoxy)ethane (BTBPE)(range <0.5–1740 ng g�1), tetrabromobisphenol A-bis(2,3-dibromopropylether) (TBBPA-DBPE) (range<20–9960 ng g�1), 2-ethylhexyl-2,3,4,5-tetrabromobenzoate (TBB) (range <2–436 ng g�1) and bis(2-ethylhexyl)-3,4,5,6-tetrabromophthalate (TBPH) (range <2–6175 ng g�1). Hexachlorocyclopentadienyl-dibromocyclooctane (HCDBCO), another NBFR, was below the detection limit of 2 ng g�1 dust in all dustsamples. No correlation was detected between concentrations of NBFRs and PBDEs. The ratio of TBB:TBPHin the dust samples ranged from 0.01 to 4.77 (average 0.42), compared to the ratio present in the com-mercial flame retardant product FM 550 (TBB:TBPH = 4:1). Furthermore, no correlation was detectedbetween concentrations in dust of TBB and TBPH. This may suggest different sources of these NBFRs,or similar sources but compound-specific differences in their indoor fate and transport. Exposure via dustingestion was estimated for both adults and toddlers under low-end (5th percentile), typical (median),and high-end (95th percentile concentrations) scenarios. These were calculated assuming 100% absorp-tion of intake dust and using mean dust ingestion (adults = 20 mg d�1; for toddlers = 50 mg d�1) and highdust ingestion (adults = 50 mg d�1; for toddlers = 200 mg d�1). Typical exposure with high dust ingestionestimates for adults were 0.01, 0.2, 0.01, 0.02 and 0.08 ng kg�1 bw d�1 and for toddlers 0.05, 1.9, 0.08, 0.4and 1.12 ng kg�1 bw d�1 for BTBPE, DBDPE, TBB, TBPH and TBBPA-DBPE, respectively. Our results showedthat, similar to PBDEs, toddlers have higher exposure to NBFRs than adults. This study documents thepresence of NBFRs in indoor environments, and emphasizes the need to evaluate the health implicationsof exposure to such chemicals.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Brominated flame retardants (BFRs) comprise a diverse varietyof brominated organic compounds used to ensure that manufac-tured goods comply with fire safety regulations. On the basis oftheir molecular structure, BFRs may be divided into three catego-ries: (i) aromatic, e.g. tetrabromobisphenol A (TBBP-A) anddecabromodiphenyl ether (BDE 209), (ii) cycloaliphatic, e.g. hexa-bromocyclododecane (HBCD), and (iii) aliphatic, e.g. dibromone-opentyl glycol (DBNPG). Since their first commercial use in 1965,a wide variety of BFRs have been developed (BSEF, 2010). Commonapplications are in building materials, vehicles, textiles, and in

ll rights reserved.

iversity of Antwerp, Univer-

.

electrical and electronic goods. In a global context, the most-widely used BFRs are polybrominated diphenyl ethers (PBDEs),HBCD, and TBBP-A. These BFRs have been extensively studiedand their occurrence reported in a variety of environmental com-partments (de Wit, et al., 2002; Covaci et al., 2006; Frederiksenaet al., 2009). Studies have shown certain BFRs possess theproperties of persistent organic pollutants (POPs), i.e. Persistent,Bioaccumulative, Toxic and Long-range transportable (BSEF,2010). Due to various health and environmental threats posed byBFRs, certain commercial PBDE formulations i.e., Penta- andOcta-BDEs are listed under Stockholm Convention and have beenbanned in the European Union, Japan, China and several states ofthe USA (Darnerud, 2003; EU, 2003; Birnbaum and Staskal, 2004;Renner, 2004; Harrad et al., 2004, 2006; Thomas et al., 2006; Zhou,2006; Betts, 2008; Harrad et al., 2008a; BSEF, 2010), while the useof another major PBDE commercial formulation, i.e. Deca-BDE, in

N. Ali et al. / Chemosphere 83 (2011) 1360–1365 1361

certain consumer products has been banned in the EU since July2008 (European Court of Justice, 2008).

As a result of the restrictions on PBDEs and HBCDs, it is thoughtlikely that there has been an increased demand for alternative (ornovel) FRs to meet flammability standards. Indeed, recent researchhas established the presence of these ‘‘novel’’ BFRs (NBFRs) in bothindoor and outdoor environments Karlsson et al., 2007; Harradet al., 2008b; Stapleton et al., 2008; Ismail et al., 2009; Shi et al.,2009; Stapleton et al., 2009; de Wit et al., 2010; Covaci et al.,2011). Moreover, there is evidence that some NBFRs may be persis-tent and can accumulate in the environment (Shi et al., 2009). Theterm NBFRs relates to BFRs which are either new to the market orhave been observed only recently in the environment. In this study,the NBFRs examined are: decabromodiphenyl ethane (DBDPE), 1,2-bis(2,4,6-tribromophenoxy)ethane (BTBPE), tetrabromobisphenolA-bis(2,3-dibromopropylether) (TBBPA-DBPE), 2-ethylhexyl-2,3,4,5-tetrabromobenzoate (TBB), bis(2-ethylhexyl)-3,4,5,6-tetra-bromophthalate (TBPH), and hexachlorocyclopentadienyl-dib-romocyclooctane (HCDBCO).

The present study reports on the concentrations of NBFRs in in-door dust taken from Belgian houses (n = 39), Belgian offices (n = 6)and UK child day care center and primary school classrooms(n = 36). Our main objectives were: (i) to evaluate the presenceof NBFRs in floor dust from selected European indoor environ-ments; (ii) to estimate exposure to NBFRs of children up to age 6and adults via dust ingestion; (iii) in conjunction with existing dataon PBDE concentrations in the same samples, to evaluate the evi-dence that NBFRs are replacing PBDEs in indoor environments;and (iv) to elucidate the extent to which our target NBFRs havecommon sources.

2. Experimental section

2.1. Chemicals and materials

BTBPE, DBDPE, HCDBCO, TBB, and TBPH were purchased fromWellington Laboratories Inc, Canada, while TBBPA-DBPE was pur-chased from Chiron AS, Norway. The following internal standards(ISs) were used: BDE 77 (AccuStandard Inc, USA) for HCDBCOand TBB, BDE 128 (AccuStandard) for BTBPE and TBPH and13C-BDE 209 (isotopic purity of 99%, Wellington Laboratories, Can-ada) for DBDPE and TBBPE-DBPE. All standard stock solutions wereprepared in a mixture of iso-octane: toluene (8:2, v/v). Dichloro-methane (for analysis grade), and acetone, iso-octane and toluene(GC grade) were purchased from Merck (Darmstadt, Germany),while n-hexane, pesticide grade, was purchased from Acros Organ-ics (Geel, Belgium). Anhydrous Na2SO4, concentrated H2SO4 (98%)and silica gel were also purchased from Merck. Empty polypropyl-ene filtration tubes (3 mL) SPE cartridges and 500 mg/3 mL Supel-clean™ ENVI™ Florisil� cartridges were used and purchased fromSupelco (Bellefonte, PA, USA). Silica gel was twice washed withn-hexane in a glass flask and activated overnight at 160 �C. Priorto each experiment, silica was reactivated by heating for 2 h at160 �C. Prior to us, all glassware was washed with detergent, rinsedwith water, dried at 100 �C, and rinsed with hexane.

2.2. Dust sampling

Belgian homes (n = 39) and offices (n = 6) dust samples werecollected as reported in detail in Roosens et al. (2010). The samepaper reports concentrations of PBDEs, HBCDs and PFCs in thesesamples. Briefly, house and office dust samples were collected be-tween January and June 2008 in Antwerp, Belgium, according to astandardized protocol described by Harrad et al. (2008a). To collectdust one square meter of carpet was vacuumed for 2 min or where

carpet was absent, 4 m2 of bare floor was vacuumed for 4 min. Acombined dust sample per house was collected from commonly-frequented rooms, including living areas, kitchens, studies andbedrooms. Samples were collected using nylon sampling socks fit-ted within the nozzle of the vacuum cleaner. After sampling, sockswere closed and sealed in zipped plastic bags. Before and aftersampling different houses, the collecting instruments were cleanedthoroughly using water and soap and a hexane-impregnated dis-posable wipe. Each dust sample was sieved through a 500 lmmesh, homogenized and stored at 4 �C prior to analysis.

Dust samples from classrooms in child day-care centers and pri-mary schools (n = 36) in the West Midlands, UK were collectedduring winter 2007-spring 2008. A detailed description is givenelsewhere (Harrad et al., 2010a), together with concentrations ofPBDEs and other contaminants. In short, samples were collectedusing a portable vacuum cleaner. To retain dust, a nylon sock witha 25 lm mesh size (Allied Filter Fabrics Ltd., Australia) was in-serted into the nozzle of the vacuum cleaner. In essence, the sam-pling procedures and materials corresponded with those deployedin Belgium. However, in contrast to the Belgian samples, the entirefloor surface of each classroom was vacuumed thoroughly (typi-cally for 4 min, depending on classroom size). This approach wastaken to account for the tendency of toddlers and young childrento explore all parts of a room. The socks containing the dust sam-ples were placed in resealable polyethylene bags for transporta-tion. The samples were sieved through a 500 lm mesh andstored in the dark at 4 �C until extraction. Nevertheless, as reportedpreviously, the extent to which such differences in sampling meth-ods influence analytes concentrations detected in dust samples isnot know (Harrad et al., 2010b).

2.3. Sample preparation

The method used for the sample preparation and instrumentanalysis is described elsewhere (Ali et al., accepted for publication).Briefly, an accurately weighed aliquot of each sample, typically75 mg, was spiked with 10 ng BDE 77, 10 ng BDE 128 and 30 ng13C-BDE 209) used as internal standards, and extracted with 2 mLof n-hexane/acetone mixture (3:1, v/v). Each sample was vortexedfor 2 min before sonication for 5 min. The process was repeatedthree times and then all samples were centrifuged for 10 min. Thesupernatant of each sample was collected and the above procedurewas repeated once more. Prior to clean up, the pooled supernatantwas evaporated to incipient dryness by a gentle nitrogen streamthen resolubilized in 1 mL of n-hexane and vortexed for 1 min.

The cleanup method involved two steps. In the 1st step, extractsin 1 mL of n-hexane were loaded on cartridges with 500 mgactivated silica and eluted with 10 mL hexane and 10 mL DCM,collected separately. Both fractions were concentrated to approxi-mately 1 mL under a gentle nitrogen stream. In the 2nd step, thehexane fraction was loaded on a 1 g acid silica (44%) cartridge,while the DCM fraction was loaded onto a 500 mg Florisil� car-tridge. Since all three ISs eluted in the hexane fraction during thefractionation on the activated silica, the same amount of IS asadded prior to the 1st step, was added to the DCM fraction beforeloading onto Florisil. This was necessary to allow the quantificationof compounds eluting in the DCM fraction, such as TBB, BTBPE,TBPH and TBBPA-DBPE. For both acid silica and Florisil cartridges,the analytes were eluted with 5 mL hexane followed by 5 mL DCM.The collected extracts were dried under a gentle stream of nitrogenand then resolubilized in 50 lL of iso-octane.

2.4. Instrument analysis

All extracts were analyzed using an 6890 Agilent (Palo Alto, CA,USA) gas chromatograph (GC) coupled to a 5973 mass spectrome-

1362 N. Ali et al. / Chemosphere 83 (2011) 1360–1365

ter (MS) operated in electron capture negative ionization (ECNI).Chromatographic separation of target NBFRs was achieved usinga DB-5 ms column (15 m � 0.25 mm � 0.1 lm). The ion source,quadrupole and transfer line temperatures were set at 250, 150and 300 �C, respectively. Helium was used as a carrier gas at an ini-tial flow of 1.0 mL min�1, at 14.5 min the flow was increased at arate of 20 mL min�1 to a final flow of 2.0 mL min�1. One micro literextract was injected in solvent vent mode with injector initial tem-perature 90 �C, kept for 0.04 min then raised at 700 �C min�1 to290 �C. Other parameters were vent flow 75 mL min�1, vent time0.04 min, purge time 1.25 min and purge flow 50 mL min�1. Theinitial oven temperature was 90 �C for 1.25 min and raised at15 �C min�1 to 300 �C and held for 12 min. BTBPE, DBDPE,TBBPA-DBPE, BDE 77 and BDE 128 were quantified by monitoringthe ions m/z 79 and 81. 13C-BDE 209 was monitored by ions m/z495 and 497. HCDBCO was monitored using the ion fragment m/z310 (quantitative) and 79 (qualitative), TBB was monitored usingion fragment m/z 357 (quantitative) and 359 (qualitative), whileTBPH was monitored using ion fragment m/z 384 (quantitative)and 515 (qualitative).

2.5. Quantification and quality assurance

All calibration standards and spiking solutions were preparedby serial dilution in iso-octane. Laboratory glassware was cleanedthoroughly with hexane prior to use. Four-point calibration plotswere created for quantification purposes and high correlation coef-ficients (R2 > 0.996) and linearity were obtained. The calibrationextracts were prepared using 100 mg Na2SO4 (as matrix surrogate)spiked with different calibration concentrations and processed asdust samples. Quantification was based on the ion m/z 79 forBTBPE, BDE 77, BDE 128, BDE 209 and TBBPA-DBPE, and ions m/z357, 310, 384 and 495 for TBB, HCDBCO, TBPH and 13C-BDE 209,respectively. Two procedural blanks were included with each batchof twelve samples. Four matrix spikes containing 100 mg of Na2SO4

spiked with target NBFRs were processed and run with the sam-ples. The target NBFRs were not detected in the procedural blanks.However, no field blanks were available for analysis.

2.6. Data analysis

Statistical analysis was performed using Minitab 15 and Micro-soft Excel 2003. To study correlations between NBFRs and previ-ously determined concentrations of PBDEs (Harrad et al., 2010a;Roosens et al., 2010), regression analysis with fitted plot line wasperformed using Minitab 15. When present, outliers were removedfor examination of correlations. To establish whether the data dis-played a normal distribution, the Ryan–Joiner normality test wasapplied, revealing concentrations of all NBFRs in each microenvi-ronment category to display a non-normal distribution (P < 0.05).Mann–Whitney test was applied to study the comparison betweenthe concentrations of NBFRs obtained from UK classrooms and Bel-gian homes dust. In all instances, where concentrations were be-low LOQ, the concentrations were assumed to equal 1/2⁄LOQ.

3. Results and discussion

3.1. Concentrations of NBFRs in Belgian and UK indoor dust

Descriptive statistics for concentrations of NBFRs (mean, med-ian, maximum, minimum, standard deviation and geometric meanvalues) are summarized in Table 1. The detection frequency for tar-get NBFRs was between 64% and 100%, except for TBB in Belgianhomes dust, where it was detected in 31% of samples. The highdetection frequencies may suggest these compounds have become

ubiquitous in indoor environments, e.g. as a consequence of theiruse as an alternative to PBDEs. Fig. 1 illustrates the existence of dif-ferent relative contributions of the target analytes to RNBFRs indust from UK classrooms and Belgian homes and offices.

Levels of BTBPE in all dust samples were an order of magnitudelower than the other flame retardants except TBB. BTBPE was pres-ent uniformly at around 5% of the total NBFRs in Belgian and UK in-door dust. BTBPE is typically added to acrylonitrile butadienestyrene polymers (ABS), which are used in electronics e.g., tele-phones, dashboards, equipments for refrigerators, toys etc. (Harjuet al., 2008). The broad range of applications may explain its occur-rence in Belgian homes, offices and UK classrooms dust samples.Despite the uniform presence of BTBPE across all type of dust,the concentrations of BTBPE in UK classrooms and Belgian homeswere significantly different from each other (p < 0.05). BTBPE wasdetected at lower concentrations (median = 2 ng g�1) in Belgianhomes dust than in UK (median = 5.3 ng g�1) and US (med-ian = 30 ng g�1) homes dust (Harrad et al., 2008b; Stapleton et al.,2008). However, BTBPE detected in Belgian offices dust (med-ian = 19 ng g�1) and UK classrooms (median = 9 ng g�1) dust wasrelatively higher than Belgian homes dust, indicating specificapplications in these microenvironments (Tables 1 and 2).

DBDPE was the major NBFR in Belgian homes and offices dust,comprising on average 43% and 50% of RNBFRs, respectively, com-pared to UK classroom dust, in which DBDPE comprised an averageof 19% of RNBFRs. Even though the profile of DBDPE across UKclassrooms and Belgian house was different, their concentrationswere not significantly different (p > 0.05). DBDPE concentrations(median 153 ng g�1) in Belgian house dust were comparable tothose reported in UK (270 ng g�1) and US (201 ng g�1) homes dust(Harrad et al., 2008b; Stapleton et al., 2008). However, the medianconcentration of DBDPE (721 ng g�1) in Belgian offices (Table 2)dust exceeded 7-fold the median concentration of DBDPE in UKoffices dust (Harrad et al., 2008b). The median concentration ofDBDPE 98 ng g�1 dust in UK classrooms dust was a magnitude low-er than Belgian offices dust observed in present study and the re-ported levels of DBDPE in US houses dust elsewhere (Stapletonet al., 2008).

The first report of the presence of TBB and TBPH in indoor dustwas in US homes dust (Stapleton et al., 2008). To our knowledge,this study is the first report on TBB and TBPH in European indoormicroenvironments. Both TBB and TBPH are constituents of theFM-550 (a flame retardant introduced in 2003 as an alternativeto Penta-BDE (Stapleton et al., 2008)). TBB was detected at the low-est percentage among all studied NBFRs at 3% in UK classrooms,2.8% in Belgian homes dust and 0.8% in Belgian offices dust. Thismay imply little current use of FM-550 in Europe. However, TBPH,another important component of FM-550, was present at on aver-age 25% RNBFRs in UK classrooms dust, 30% RNBFRs in Belgianhomes dust, but only 6% RNBFRs in Belgian offices dust. Even withsimilar profile pattern of TBPH and TBB in UK classrooms and Bel-gian homes dust, their concentrations were statistically differentfrom each other (p < 0.05). The principal use of TBB is thought tobe in FM-550 (in which the TBPH:TBB ratio is 1:4 (Stapletonet al., 2008)). However, higher TBPH:TBB ratios are observed inour dust samples. These ratios, coupled with the greater relativeabundance of TBPH, indicate the existence of other sources ofTBPH, e.g. TBPH has been used as plasticizer in polyvinyl chloride(PVC) and in neoprene (Anderson et al., 2006). Another explanationfor the elevated TBPH:TBB ratios could be more facile degradationof TBB relative to TBPH (Davis and Stapleton, 2009). Yet, concentra-tions of TBB and TBPH in dust were significantly correlated forboth UK classrooms (R2 = 0.572; p < 0.05) and Belgian homes(R2 = 0.967; p < 0.05). TBB and TBPH occurred at median concentra-tions of 1 and 13 ng g�1 in Belgian home dust, 7 and 64 ng g�1 inBelgian office dust, and 25 and 96 ng g�1 in UK classrooms dust,

Table 1Descriptive data of NBFRs in Belgian house, office and UK classroom dust (ng g�1 dust).

Microenvironment/Analyte BTBPE DBDPE TBB TBPH TBBPA-DBPE BDE 47 BDE 99 BDE 183 BDE 197 BDE 209

UK classrooms n = 36 n = 36 n = 36 n = 36 n = 36 n = 43a n = 43a n = 43a n = 43a n = 43a

Detection (%) 86 75 92 97 645th PERCENTILE 0.25 10 1 10 10 4 5.3 <2 <3 14095th percentile 204 1282 126 1424 2891 85 140 23 31 24 000Mean 78 293 45 381 729 32 54 5.1 5.6 8500St. Dev. 296 493 55 1151 1863Median 9 98 25 96 107 26 36 1.2 3.1 5000Minimum <0.5 <20 <2 <2 <20 1.6 1.1 <2 <3 49Maximum 1741 2467 289 6175 9961 120 270 48 35 88 000Geomean 8 95 22 83 104

Belgian homes n = 39 n = 39 n = 39 n = 39 n = 39 n = 43b n = 43b n = 43b n = 43b n = 43b

Detection (%) 85 100 31 97 855th Percentile 0.25 64 1 3 10 1.1 1.5 0.5 0.5 6095th Percentile 79 903 75 450 402 63 110 16 10 1537Mean 33 303 20 212 144 21 37 11 4 604St. Dev. 163 389 73 867 221 49 115 43 9.5 994Median 2 153 1 13 78 8 9 2 1.4 317Minimum <0.5 55 <2 <2 <20 0.5 0.6 0.1 0.5 15Maximum 1019 2126 436 5004 1286 307 748 262 53 5295Geomean 2 193 2 19 70 8 11 2 2 306

Belgian offices n = 6 n = 6 n = 6 n = 6 n = 6 n = 10b n = 10b n = 10b n = 10b n = 10b

Detection% 100 100 67 100 835th Percentile 3 214 1 24 18 10 22 2 1 10295th Percentile 300 1617 30 228 1852 61 133 3090 1200 6679Mean 80 789 12 95 608 26 58 578 225 1513St. Dev. 150 587 13 89 838 19 41 1718 667 3540Median 19 721 7 64 306 21 45 24 9 443Minimum 2 170 <2 16 <20 10 19 0.6 0.5 69Maximum 384 1846 31 265 2211 67 141 5464 2121 11 574Geomean 18 612 6 67 192 21 49 25 11 468

Data for PBDEs congeners (BDE 47, 99, 183, 197 and 209) reported elsewhere.a UK classrooms (Harrad et al., 2010a).b Belgian homes (Roosens et al., 2010).

UK classroomsBelgian homesBelgian offices

BTBPE DBDPE TBB TBPH TBBPA-DBPE

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

%

Fig. 1. Percentage distribution of novel BFRs in indoor dust from UK classrooms,Belgian homes and offices. Error bars represent standard deviations.

Table 2Comparison of median values of novel BFRs studied in indoor dust (ng g�1 dust).

N BTBPE DBDPE TBB TBPH TBBPA-DBPE Reference

Belgianhomes

39 2 153 1 13 78 Present study

Swedishhouses

5 4.8a 47a n.m. n.m. n.m. Karlsson et al.(2007)

UK houses 30 5.3 24 n.m. n.m. n.m. Harrad et al.(2008b)

US houses 37 30 201 133 142 n.m. Stapleton et al.(2008)

Belgianoffices

6 19 721 7 64 306 Present study

UK offices 18 <dl 99 n.m. n.m. n.m. Harrad et al.(2008b)

UK schools 36 9 98 25 96 107 Present study

‘n.m.’ Not measured.‘<dl’ below detection limit.

a Mean values.

N. Ali et al. / Chemosphere 83 (2011) 1360–1365 1363

respectively. The obtained median concentrations were an order ofrelatively lower than the median concentrations reported in UShouse dust 133 and 142 ng g�1 dust for TBB and TBPH, respectively(Stapleton et al., 2008).

Median concentrations of TBBPA-DBPE were 78, 306, and107 ng g�1 dust in Belgian house, Belgian office, and UK classroomdust respectively (Table 2). Interestingly, TBBPA-DBPE was the ma-jor NBFR detected in UK classrooms dust comprising on average48% RNBFRs. In Belgian house and office dust, TBBPA-DBPEcontributed 20% and 38% RNBFRs, respectively. DBDPE andTBBPA-DBPE are the two predominant NBFRs in dust from all threestudied microenvironments, while BTBPE and TBB were only minorcomponents.

Although definitive conclusions are hampered by the relativelysmall number of samples analyzed in this preliminary study; dif-

ferences between the NBFR patterns observed in dust from differ-ent microenvironment categories may be attributable todifferences in the numbers and types of flame-retarded goodsand materials (e.g. textiles, foams, furniture etc.) used in domestic,offices, and classrooms environments. For example, higher levels ofDBDPE in homes and offices dust than classroom dust are probablydue to its use in a wide range of consumer products e.g., high im-pact polystyrene (HIPS), ABS, polycarbonate/ABS polymer (PC/ABS), HIPS/polyphenylene oxide. Such products are mostly usedin homes and offices than in classrooms, which agrees with thelower contribution of DBDPE to the NBFR profile in UK classroomdust (Harju et al., 2008). TBPH is used in polyvinyl chloride (PVC)and neoprene which are high demand and use in commercialproducts like wires, cables, floor mats and industrial sheets (Harju

Table 3Assessment of human exposure to NBFRs via dust ingestion, using mean and high dust intake rates for working adults, non working adults and toddlers. All values are inng kg�1 bw d�1. We assumed 100% absorption of intake dust (Jones-Otazo et al., 2005).

Toddlera Non working adultb Working adultc

5th Percentile 95th Percentile Median 5th Percentile 95th Percentile Median 5th Percentile 95th Percentile Median

BTBPEMean dust ingestiond 0.00 0.35 0.01 0.00 0.02 0.00 0.00 0.04 0.00High dust ingestione 0.00 1.39 0.05 0.00 0.06 0.00 0.00 0.08 0.00

DBDPEMean dust ingestion 0.18 3.26 0.47 0.02 0.26 0.04 0.03 0.30 0.08High dust ingestion 0.71 13.06 1.89 0.05 0.65 0.11 0.06 0.71 0.18

TBBMean dust ingestion 0.00 0.28 0.02 0.00 0.02 0.00 0.00 0.02 0.00High dust ingestion 0.01 1.14 0.08 0.00 0.05 0.00 0.00 0.05 0.00

TBPHMean dust ingestion 0.02 2.15 0.10 0.00 0.13 0.00 0.00 0.12 0.01High dust ingestion 0.06 8.61 0.40 0.00 0.32 0.01 0.01 0.28 0.02

TBBPA-DBPEMean dust ingestion 0.03 3.01 0.28 0.00 0.12 0.02 0.00 0.20 0.04High dust ingestion 0.13 12.03 1.12 0.01 0.29 0.06 0.01 0.46 0.08

a Assuming toddler dust ingestion is 79.9% home and 20.1% classroom.b Assuming non working adult dust ingestion is 100% home.c Assuming working adult dust ingestion is 78.9% home and 21.1% office.d Mean dust ingestion rate for adults = 20 mg d�1; for toddlers = 50 mg d�1.e High dust ingestion rate for adults = 50 mg d�1; for toddlers = 200 mg d�1.

1364 N. Ali et al. / Chemosphere 83 (2011) 1360–1365

et al., 2008), so its presence at higher concentrations in Belgianhomes and UK day classrooms dust samples is understandable,while in case of Belgian offices dust, the sample size (n = 6) is small,which might not be the true indicator of its concentrations in offi-ces dust. BTBPE is typically used in ABS which may explains theuniformly occurrence of this compound across homes, offices andclassrooms dust samples.

A comparison was drawn between the concentrations of stud-ied NBFRs and concentrations of PBDE congeners earlier reportedin the same sets of samples by Harrad et al. (2010a) and byRoosens et al. (2010). Used an alternative to the Octa-BDE mixture,BTBPE levels were compared with its main congeners i.e. BDE 183and BDE 197, while DBDPE and TBBPA-DBPE levels were comparedwith BDE 209. TBB and TBPH concentrations were studied againstthe major congeners of Penta-BDE mixture i.e. BDE 47 and BDE 99.No significant correlations were found between NBFRs and PBDEcongeners (p > 0.05), except between BTBPE and BDE 197 in UKclassroom dust (R2 = 0.455; p < 0.05). Concentrations of TBBPA-DBPE and DBDPE in both Belgian homes dust and UK classroomsdust were not correlated (p > 0.05), indicating different sources ofemission. For both TBBPA-DBPE and DBDPE, the values in UK class-rooms and Belgian homes dust were significantly different andwere not statistically correlated with each other (R2 = 0.003 forDBDPE and R2 = 0.009 for TBBPA-DBPE, p > 0.05), hinting to a diver-sity of contamination sources.

3.2. Exposure assessment via dust ingestion

In order to make a preliminary evaluation of the exposure viadust ingestion to NBFRs, we assumed 100% absorption of intakein line with other studies (Jones-Otazo et al., 2005). We also as-sumed average adult and toddler dust ingestion figures of 20 and50 mg d�1, and high dust ingestion figures for adults and toddlersof 50 and 200 mg d�1 (Jones-Otazo et al., 2005). We then estimatedlow-end, ‘‘typical’’ and high-end dust ingestion exposure scenariosfor homes, offices, and classrooms separately, using 5th percentile,median, and 95th percentile concentrations in the dust samples re-ported here. Overall dust ingestion exposure estimates were thencalculated taking into account ingestion of dust in each of the rel-evant microenvironments (homes and offices for adults; homes

and classrooms for children). Dust ingestion is assumed to occurpro-rata to typical activity patterns (i.e. for adults 78.9% home,21.1% office; for toddlers 79.9% home and 20.1% classroom)(Harrad et al., 2008b).

Typical (mean) exposure with high dust ingestion estimates foradults were 0.01, 0.2, 0.01, 0.02 and 0.08 ng kg�1 bw d�1 and fortoddlers 0.05, 1.9, 0.08, 0.4 and 1.12 ng kg�1 bw d�1 for BTBPE,DBDPE, TBB, TBPH and TBBPA-DBPE, respectively. For NBFRs, thecalculated mean and high exposure values for both toddlers andadults (Table 3) were significantly lower than their correspondingreference dose (RfD) values. For TBB and TBPH, no RfD values wereavailable which made a case to use surrogate RfD values. The sur-rogate RfD values were used based on the structural comparisonsof TBB and TBPH with bis(2-ethylhexyl)-phthalate (DEHP) (Hardyet al., 2008). The used RfD values are: BTBPE = 243,000 ng kg�1

bw d�1, DBDPE = 333,333 ng kg�1 bw d�1, TBB = 20,000 ng kg�1

bw d�1, TBPH = 20,000 ng kg�1 bw d�1 described elsewhere (Hardyet al., 2008). These RfD values have been established on the basis ofrather old toxicological studies with a lack of robust or recent dataon NBFRs and therefore the health impacts of these exposures can-not be fully evaluated at the moment. Hence, the presence ofNBFRs in our microenvironments demands thorough toxicologicalstudies, which should lead to a revision of these RfD values.

The samples analyzed in the present study were collected in2007 and 2008. Hence, the relatively low concentrations of NBFRscompared to PBDEs in these samples, may reflect the fact that atthe time of sample collection, the ban on the PBDE mixtures andthe corresponding increase in the use of NBFRs as their replace-ment was in its infancy. Subsequent and future increases in useof NBFRs may therefore lead to higher levels in indoor environ-ments. Furthermore, characterization of exposure via other path-ways such as diet and inhalation are required. Finally, while thisstudy demonstrates the presence of NBFRs in indoor environments,more detailed studies are required to elucidate the specific sourcesof this contamination in individual microenvironments.

4. Concluding remarks

This study adds to the growing weight of evidence for environ-mental contamination by NBFRs and provides evidence that both

N. Ali et al. / Chemosphere 83 (2011) 1360–1365 1365

adults and young children are exposed to these chemicals viaingestion of indoor dust. Against this backdrop, there is a clearneed for further investigations into the origins of such contamina-tion and into the human health implications arising from exposureto NBFRs.

Acknowledgements

A.C. acknowledges gratefully the provision of a postdoctoral fel-lowship from the Research Scientific Foundation of Flanders(FWO). N.A. thanks the University of Antwerp for financially sup-porting his PhD studies, while EG acknowledges studentship fund-ing from NERC and Unilever (Ref. NER/S/U/2006/14255). LaurenceRoosens and Wendy D’Hollander are acknowledged for the collec-tion of dust from Belgian houses and offices.

References

Ali, N., Harrad, S., Muenhor, D., Neels, H., Covaci A., accepted for publication.Analytical characteristics and determination of major novel brominated flameretardants (NBFRs) in dust. Anal. Bioanal. Chem.

Anderson, P.L., Öberg, K., Örn, U., 2006. Chemical characterization of brominatedflame retardants and identification of structurally representative compounds.Environ. Toxicol. Chem. 25, 1275–1282.

Betts, K.S., 2008. New Thinking on flame retardants. Environ. Health Persp. 116.Birnbaum, L.S., Staskal, D.F., 2004. Brominated flame retardants: cause for concern?

Environ. Health Persp. 112, 9–17.Bromine Science Environmental Forum (BSEF). <http://www.bsef.com> (accessed

15.11.10).Covaci, A., Gerecke, A.C., Law, R.J., Voorspoels, S., Kohler, M., Heeb, N.V., Leslie, H.,

Allchin, C.R., de Boer, J., 2006. Hexabromocyclododecanes (HBCDs) in theenvironment and humans: a review. Environ. Sci. Technol. 40, 3679–3688.

Covaci, A., Harrad, S., Abdallah, M.A.E., Ali, N., Law, R.J., Herzke, D., de Wit, C.A., 2011.Novel brominated flame retardants: a review of their analysis, environmentalfate and behaviour. Environ. Int. 37, 532–556.

Darnerud, P.O., 2003. Toxic effects of brominated flame retardants in man andwildlife. Environ. Int. 29, 841–853.

Davis, E.F., Stapleton, H.M., 2009. Photodegradation pathways of nonabrominateddiphenyl ethers, 2-ethylhexyltetrabromobenzoate and di(2-ethylhexyl)tetra-bromophthalate: identifying potential markers of photodegradation. Environ.Sci. Technol. 43, 5739–5746.

de Wit, C.A., 2002. An overview of brominated flame retardants in the environment.Chemosphere 46, 583–624.

de Wit, C.A., Herzke, D., Vorkamp, K., 2010. Brominated flame retardants in theArctic environment – trends and new candidates. Sci. Total. Environ. 408, 2885–2918.

European Court of Justice, 2008. Cases C-14/06 and C-295/06, Judgement of theCourt, 1 April 2008, Directive 2002/95/EC and Commission Decision 2005/717/EC. <http://curia.europa.eu> (accessed 07.10).

European Union, 2003. The Ban of Marketing the Penta- and Octa-BDE Mixtures.Directive 2003/11/EC of the European Parliament and of the Council of 6February 2003 Amending for the 24th Time Council Directive 76/769/EECRelating to Restrictions on the Marketing and Use of Certain DangerousSubstances and Preparations.

Frederiksena, M., Vorkampb, K., Thomsenc, M., Knudsena, L.E., 2009. Humaninternal and external exposure to PBDEs – a review of levels and sources. Int. J.Hyg. Environ. Health 212, 109–134.

Hardy, M., Biesemeir, J., Banasik, M., Stedeford, T., 2008. Brominated flameretardants in house dust: detection does not equate to risk. Environ. Sci.Technol. 42, 9453–9454.

Harju, M., Heimstad, E.S., Herzke, D., Sandanger, T., Posner, S., Wania, F., 2008.Emerging ‘‘new’’ Brominated Flame Retardants in Flame Retarded Products andthe Environment. Report 2462, Norwegian Pollution Control Authority, Oslo,Norway.

Harrad, S., Wijesekera, R., Hunter, S., Halliwell, C., Baker, 2004. A preliminaryassessment of UK human dietary and inhalation exposure to polybrominateddiphenyl ethers. Environ. Sci. Technol. 38, 2345–2350.

Harrad, S., Hazrati, S., Ibarra, C., 2006. Concentrations of polybrominated diphenylethers in indoor air and dust and polychlorinated biphenyls in indoor air inBirmingham, United Kingdom: implications for human exposure. Environ. Sci.Technol. 40, 4633–4638.

Harrad, S., Ibarra, S., Diamond, M., Melymuk, L., Robson, M., Douwes, J., Roosens, L.,Dirtu, A.C., Covaci, A., 2008a. Polybrominated diphenyl ethers in domesticindoor dust from Canada, New Zealand, United Kingdom and United States.Environ. Int. 34, 232–238.

Harrad, S., Ibarra, C., Abdallah, M.A.E., Boon, R., Neels, H., Covaci, A., 2008b.Concentrations of brominated flame retardants in dust from United Kingdomcars, houses and offices: causes of variability and implications for humanexposure. Environ. Int. 34, 1170–1175.

Harrad, S., Goosey, E., Desborough, J., Abdallah, M.A.E., Roosens, L., Covaci, A., 2010a.Dust from UK primary school classrooms and daycare centers: the significanceof dust as a pathway of exposure of young UK. children to brominated flameretardants and polychlorinated biphenyls. Environ. Sci. Technol. 44, 4198–4202.

Harrad, S., de Wit, C.A., Abdallah, M.A.-E., Bergh, C., Björklund, J.A., Covaci, A.,Darnerud, P.O., de Boer, J., Diamond, M., Huber, S., Leonards, P., Mandalakis, M.,Östman, C., Småstuen Haug, L., Thomsen, C., Webster, T.F., 2010b. Indoorcontamination with hexabromocyclododecanes, polybrominated diphenylethers and perfluoroalkyl compounds: an important exposure pathway forpeople? Environ. Sci. Technol. 44, 3221–3231.

Ismail, N., Gewurtz, S.B., Pleskach, K., Whittle, D.M., Helm, P.A., Marvin, C.H., Tomy,G.T., 2009. Brominated and chlorinated flame retardants in Lake Ontario,Canada in lake trout (Salvelinus namaycush) between 1979 and 2004 andpossible influences of food-web changes. Environ. Toxicol. Chem. 28, 910–920.

Jones-Otazo, H.A., Clarke, J.P., Diamond, M.L., Archbold, J.A., Ferguson, G., Harner, T.,Richardson, G.M., Ryan, J.J., Wilford, B., 2005. Is house dust the missing exposurepathway for PBDEs? An analysis of the urban fate and human exposure toPBDEs. Environ. Sci. Technol. 39, 5121.

Karlsson, M., Julander, A., van Bavel, B., Hardell, L., 2007. Levels of brominated flameretardants in blood in relation to levels in household air and dust. Environ. Int.33, 62–69.

Renner, R., 2004. In US, flame retardants will be voluntarily phased out. Environ. Sci.Technol. 38, 14A.

Roosens, L., Cornelis, C., d’ Hollander, W., Bervoets, L., Reynders, H., vanCampenhout, K., van den Heuvel, R., Neels, H., Covaci, A., 2010. Exposure ofthe Flemish population to brominated flame retardants: model and riskassessment. Environ. Int. 36, 368–376.

Shi, T., Chen, S.J., Luo, X.J., Zhang, X.L., Tang, C.M., Luo, Y., Ma, Y.J., Wu, J.P., Peng, X.Z.,Mai, B.X., 2009. Occurrence of brominated flame retardants other thanpolybrominated diphenyl ethers in environmental and biota samples fromsouthern China. Chemosphere 74, 910–916.

Stapleton, H.M., Allen, J.G., Kelly, S.M., Konstantinov, A., Klosterhaus, S., Watkins, D.,McClean, M.D., Webster, T.F., 2008. Alternate and new brominated flameretardants detected in US house dust. Environ. Sci. Technol. 42, 6910–6916.

Stapleton, H.M., Klosterhaus, S., Eagle, S., Fuh, J., Meeker, J.D., Blum, A., Webster, T.F.,2009. Detection of organophosphate flame retardants in furniture foam and UShouse dust. Environ. Sci. Technol. 43, 7490–7495.

Thomas, G.O., Wilkinson, M., Hodson, S., Jones, K., 2006. Organohalogen chemicalsin human blood from the United Kingdom. Environ. Pollut. 141, 30–41.

Zhou, Z.M., 2006. Implement of administrative measure on the control of pollutioncaused by electronic information products and the exemption of deca–BDEmixture. Flame Retard. Mater. Technol. 4, 15–16 (in Chinese).