W&M ScholarWorks
Dissertations, Theses, and Masters Projects Theses, Dissertations, & Master Projects
2006
Disposition of Wastewater-Associated Polybrominated Diphenyl Ethers in a Freshwater Receiving Stream
Mark Joseph La Guardia College of William & Mary - Arts & Sciences
Follow this and additional works at: https://scholarworks.wm.edu/etd
Part of the Environmental Sciences Commons, and the Hydrology Commons
Recommended Citation La Guardia, Mark Joseph, "Disposition of Wastewater-Associated Polybrominated Diphenyl Ethers in a Freshwater Receiving Stream" (2006). Dissertations, Theses, and Masters Projects. William & Mary. Paper 1539626851. https://dx.doi.org/doi:10.21220/s2-5pjx-m003
This Thesis is brought to you for free and open access by the Theses, Dissertations, & Master Projects at W&M ScholarWorks. It has been accepted for inclusion in Dissertations, Theses, and Masters Projects by an authorized administrator of W&M ScholarWorks. For more information, please contact [email protected]. Disposition of Wastewater-Associated Polybrominated
Diphenyl Ethers
In a Freshwater Receiving Stream
A Thesis
Presented to
The Faculty of the Department of Chemistry
The College of William and Mary in Virginia
In Partial Fulfillment
Of the Requirements for the Degree of
Master of Science
by
Mark Joseph La Guardia
2006 APPROVAL SHEET
This thesis is submitted in partial fulfillment of
the requirements for the degree of
Master of Science
Marlt J. La Guardia
Approved by the Committee, December 2006 TABLE OF CONTENTS
Page
List of Tables iv
List of Figures v
Abstract viii
Introduction 2
Study Goal 19
Study Site and Sample Location 21
Samples, Methodology and Quality Control 29
Results 42
Discussion 64
Conclusion 84
Appendix 87
References 101
Vita 106
iii LIST OF TABLES
Table Page
1. Technical flame-retardant (penta-, octa-, and deca-PBDEs) compositions
(%, w/w) 5
2. PBDEs detected in various environmental and biological matrixes 8
3. Sample ID, location, % lipids and %TOC 24
4. PBDE congener analysis, Relative Retention Indices (RRIs) and major
fragmentation ions (El and ECNI) 36
5. Matrix spiking solution 41
6. PBDEs (pg/kg, dry weight) in wastewater sludge, Roxboro WWTP, NC. 47
7a. PBDEs (pg/kg, %TOC) in sediments from Marlowe/Storys Creek and
tributaries, collected 2002 53
7b. PBDEs (pg/kg, %TOC) in sediments from Marlowe/Storys Creek and
tributaries, collected 2005, plus analytical blanks 54
8. PBDEs (pg/kg, %lipid) in biota collected at Roxboro WWTP outfall and
analytical blank 59
9. Matrix (NaSC> 4, sediments and fish tissue) PBDEs fortified recoveries and
replicate analysis 63 LIST OF FIGURES
Figure Page
1. Representative PBDE structure 4
2. Tetra- to hexa-BDEs in biosolids compared to DE-71 composition 16
3. Hyco River subbasin, North Carolina, U.S. (NCDENR, 2000) 22
4a. Sample locations on lower Marlowe/Storys Creek, Person County, NC. 27
4b. Sample locations on upper Marlowe/Storys Creek and tributaries (Ghent
and Storys Creeks), Person County, NC. 28
5a. Representative chromatograph of 27 PBDEs, ptclb and DCDE by
on-column injection, ECNI-SIM (DB-5HT, 30 m, 0.25mm i.d., 0.1 pm) 33
5b. Representative chromatograph of 27 PBDEs, ptclb and DCDE by
split/splitless injection, ECNI-SIM (DB-5HT, 30 m, 0.25mm i.d., 0.1 pm) 33
6. BDE-209 comparison using on-column and split/splitless injectors 35
7. Relative Retention Indices (RRIs) formula 39
8. ECNI “full-scan” spectra of octa-PBDEs illustrating differences in spectra
as a function of bromine positioning 43
9. ECNI-SIM chromatograph of PBDEs in wastewater sludge (#5BW001) 45
10. El chromatograph of PBDE major ions in wastewater sludge (#5BW001) 46
11. ECNI-SIM comparison of sludge collected in 2005 and 2002 48
12a. El spectra of hexabromocyclododecane (HBCD) in sludge 50
12b. El spectra of 1,2-bis (2,4,6-tribromophenoxy) ethane (TBE) in sludge 50
v LIST OF FIGURES (continued)
Figures Page
13. ECNI-SIM chromatograph of PBDEs in sediments (#0BS016) and
(#5BS003) 51
14. El chromatograph of PBDE major ions in sediment (#5BS003) 52
15. ECNI-SIM chromatograph of PBDEs in biota (#0BF072) 56
16a. El chromatograph of tri- trough hexa-PBDEs major ions in biota
(#0BF072) 57
16b. El chromatograph of hepta- through deca-PBDEs major ions in biota
(#0BF072) 58
17a. Roxboro WWTP sludge (2002 and 2005) tri- through hexa-PBDEs 65
17b. Roxboro WWTP sludge (2002 and 2005) hepta- through deca-PBDEs 65
18. BDE-209 transfers to Roxboro WWTP (USEPA, TRI. 2006) 67
19. Penta-formulation (DE-71) compared to Roxboro WWTP sludge
(2002 and 2005) and U.S. sludges (n=l 1) 69
20. Total PBDEs in Marlowe/Storys Creek surficial sediments, collected
from Roxboro WWTP to 10.8 km downstream 71
21a. Marlowe/Storys Creek surficial sediment congener profiles (tri through
octa-PBDEs), 2002 and 2005 73
21b. Marlowe/Storys Creek surficial sediments congener profiles (nona-
through deca-PBDEs), 2002 and 2005 73
vi LIST OF FIGURES (continued)
Figures Page
22. Tetra- through hepta-PBDEs homologue of surficial sediments
and sludge (2002 and 2005) compared to DE-71 76
23 a. PBDEs in biota (tri- through hexa-PBDEs) 77
23b. PBDEs in biota (hepta- through deca-PBDEs) 77
24. Chromatograph of PBDEs in creek chub compared to common
carp BDE-209 exposure study 81
25. Chromatograph of PBDEs in sunfish compared to rainbow trout
BDE-209 exposure study 82
vii ABSTRACT
Polybrominated diphenyl ethers (PBDEs) are commonly used brominated flame- retardants (BFRs). They are environmentally persistent and have become widely distributed. PBDEs can bioaccumulate and disrupt biological processes, e.g. the endocrine system. However, identification of their major routes of release into the environment is lacking. To date, the majority of PBDEs detected in biota contain six or fewer bromines, presumably from the recently discontinued commercial penta-PBDE product. However, the dominant PBDE product in commerce, the deca- formulation (containing >97% decabromodiphenyl ether (BDE-209), by weight), remains in commerce, as it is believed to pose comparatively minimal risk. Nonetheless, concerns exist over the possible debromination of deca- into more bioaccumulative congeners. However, this has only been shown under laboratory conditions. To evaluate the potential for debromination under realistic environmental conditions the distribution of PBDE congeners was tracked in a variety of matrices from a wastewater treatment plant (WWTP) sludge to receiving stream sediments and associated aquatic biota. Samples were collected in 2002 and 2005. GC/MS identified tri- through deca-PBDEs in these matrices. BDE-209 was the major congener in sludge (58.8 and 37.4 mg/kg, d.w., respectively), sediments (6.2 to 3150 mg/kg, %TOC) and some biota (non-detect to 21,650 pg/kg, l.w.). Also detected in these matrixes were 22 additional PBDEs, other major congers BDE- 206, -99 and -47 in sludge, -206, -207 and -99 in sediments and BDE-47, -99 and - 153 in biota. Sludge congener profiles were similar to the penta- and deca- formulations, suggesting minimal -209 debromination during wastewater treatment. Similar profiles were also observed in surficial sediments collected at the outfall and several km downstream, again indicating minimal debromination. Sludge and sediments contained additional BFRs of emerging concern, e.g. hexabromocyclododecane (< 8 mg/kg) and 1,2-bis (2,4,6-tribromophenoxy) ethane (<1.1 mg/kg). Sunfish (Lepomis gibbosus), creek chub (Semotilus atromaculatus) and crayfish (Cambarus puncticambarus sp. c) collected in 2002 near the outfall and sunfish collected in 2005, were depurated and whole body PBDE analyses conducted. Sunfish (2002, 2005) profiles were similar, but -209 (2880 pg/kg) was only detected in the 2002 sample. BDE-179, -184, -188, -201 and -202 were also detected in these biota samples, but not in sludges or sediments. A previous laboratory study identified these same BDE-209 debromination products in fish. This suggests that metabolic debromination of -209 does occur in the aquatic environment under realistic conditions. Hence risk assessments that assume no BDE-209 debromination may underestimate associated bioaccumulation and toxicity attributable to the less brominated congeners produced. Disposition of Wastewater-Associated Polybrominated
Diphenyl Ethers
In a Freshwater Receiving Stream INTRODUCTION
Polybrominated diphenyl ethers (PBDEs) [Figure 1] are widely used flame- retardant additives in polymers and textiles. The demand for polymer-based products for electrical and electronic equipment, as well as automotive equipment, construction materials, and textiles has increased over the past decade. Accordingly, the demand for brominated flame-retardants (BFRs) for these products has doubled from an estimated
145,000 metric tonnes (MT) in 1990 to 310,000 MT in 2000 (Alaee et al., 2003). These continue to be produced at high volumes (237,727 and 223,482 MT for 2002 and 2003, respectively) (http://www.bsef.com/bromine/our industry/ (BSEF, 2006)). Of the BFRs,
PBDEs were only surpassed by tetrabromobisphenol-A (TBBPA) in regards to production volume (2001 global demand 67,390 versus 119,700 MT respectively (BSEF,
2006)). PBDEs were typically produced commercially at three different levels of bromination, nominally penta-, octa- and deca-PBDEs. The deca-PBDE formulation made up 83.3% of the 2001 PBDE global market demand, followed by penta- 11.1% and octa- 5.6% (BSEF, 2006). PBDEs have been commonly used in furniture (polyurethane foam), wire and cable insulation (styrene copolymers), electronics and computers (high- impact polystyrene). The vast majority of the global penta-PBDE production was consumed in North America (e.g. 98% in 2001), mostly to flame retard polyurethane foams used in furniture padding (Renner, 2000a). However, unlike reactive BFRs like
TBBPA, PBDEs are simply blended with the polymers during their formation and thus may migrate from products more readily (Alaee et al., 2003).
2 Since the 1981 report of PBDEs in fish from Sweden (Anderson and Blomkuist,
1981), subsequent papers have documented their widespread distribution in the environment (de Boer et al., 1998, Hale et al., 2001a and Renner, 2000b). These flame- retardants have also been detected in human blood plasma, adipose tissue and breast milk
(Renner, 2000b). Noren and Meironyte (2000) reported that levels of PBDEs in Swedish breast milk doubled every five years, between 1972 -1997. Sjodin (2003) noted that levels of BDE-47 (Refer to Table 1 and 2 for compound identification and formulation.), a component of the penta-PBDE technical product, has now exceeded PCB-153 (2,2’,
4,4’, 5,5’-hexachlorobiphenyl) a persistent organic pollutant (POPs), ban in the late
1970’s, in many human blood samples. Hydroxylated metabolites of PBDEs are structurally similar to thyroxin and triiodothyronine. These have been shown to bind to related receptor proteins, suggesting that they may interfere with normal physiological functions (Meerts et al., 2000). Due to growing environmental and human health concerns, a ban on the use of penta- and octa-PBDE formulations went into effect throughout the European Union (EU) in 2004. The U.S. Environmental Protection
Agency (EPA) brokered a voluntary cessation in production with the sole American manufacturer of these two products, effective at the end of 2004 (Renner, 2004).
However, these restrictions will not eliminate PBDE releases from products currently in- service or new products manufactured with recycled PBDE-containing materials. It has also been hypothesized that the most widely used PBDEs product, the unregulated deca-
PBDE, may to some extent debrominate once released into the environment. This could result in a suite of less brominated congeners with enhanced toxicity and ability to bioaccumulate relative to the parent. PBDE release routes to the environment have been enigmatic, as they are used as polymer additives, a presumably non-dispersal application.
But recent studies have detected PBDEs in wastewater treatment plant (WWTP) sewage sludge (Hale et al., 2001b), effluents (North, 2004) and in sediments of their receiving waters (Sellstrom et al., 1998; de Boer et al., 2003). This indicates that WWTPs may be an important source of PBDEs to the aquatic environment.
Figure 1. Representative PBDE structure
Br.m Br.
Like polychlorinated biphenyls (PCBs), 209 different PBDE congeners are feasible and the same IUPAC scheme is used for their naming. In reality PBDE commercial mixtures tend to be much simpler than those of PCBs due to the directing influence of the oxygen [Figure 1]. Commercial mixtures consist primarily of 38 individual PBDE congeners, however recent studies have identified an additional 13 (five tentatively identified based on their mass spectrometry spectra), totaling 48 congeners
(La Guardia et al., 2006) [Table 1]. The penta- formulation consists primarily of six isomers (tetra- through hexa-PBDEs) with BDE-47 and -99 contributing > 70%. The octa- mixture contains eleven primary isomers, which includes hexa- through deca-
PBDEs. Its major isomers are BDE-183, constituting > 40%, and four octa-PBDEs making up > 30% of the product [Table 1]. However, the European formulation Bromkal 1- W 73 73 *73 *0 73 ^3 S g g c c c g g g CQ oe H Q oa 7373t3 ’07 3 t37373 h g g c g c c g ^ Pi
CSi oc aJ 13, a . — i 5I S'u a «S
ph fn n T3 T) *d o o • o • C G C nd
w ■2 S ______
c o c$c$c$r3c$c$c$c$cQcQ(Q(dcdcQcd 3 w ______E ° ° s o °? ■r O' CQ i" ■a t3 fl 9 vo ■« C o o o a S o 3 W o 1 Q o “P d C 2sc^ < 2 2 2 hS o ’2' fifi6 H - ' 2 BC6 d °fl^ d fic *" o CO f " w 0 CQ jg w pH 1 E ° S s O E S O 03 oi- “P o 0 00 r- -o T3 c C5 G C3 ^ 03 cd OO as Os ^ d BT3 G G w-4 G G cO G G c c c 1 a d ri C3 -M o> CQ r - o a o ■q ir> vo (N VO •— C3 9 o Ov GC o , *OO n ■CTj;io H Q " fcd pq «n Oh u a W "2 Q U t-~ * a . "3 i o * SH os °?S Jr £ OQ f ~ #\ X® ■2 £ O' g X> 0£ § i •« c ■«§ 3 • « * 091 l"O' . O s o s m OO CN (X) o 2 a ©2 § i 6 £ CM >- Os 09 I" o 0 S» (N oo ^ £- w £“ 1/s° c a> 09 r - c a « *du cl ■<-* a> Im u £?£ a> W d' ^ S <3 cd cd cd § PQ u a JK C c C S CQ “ ^ B 3 ° . S > w EQ t*q c*3 i •p^ CQ cq Q Q Q Q Q Q Q K . oCd ^ •«—> b £= CQ cq CQ CQ CQS CQ O P*P -o w . • -C i °? K 0 a i Q i ”0 o u o o o o VO VO 01 © o o o o o o - •> 09 vo H =3 v o Vo VO Vo VO vo 'ss Ph (3 U /•“s O lo *o Vo” Vo” VO” * *s TH <-> I— 1 ^ TJ O, 'n 2 | o r.Sj& V E > V *n” *o“ ■«■2 SJ ■ o> V V r iJ =5 o V V V ^ •* § § -S Oh e U V • p H *n * n V rrl 'n 1K ^-Q1 5+2 d **1 *N However, trace levels of tri- through hepta-PBDEs have also been detected as components of the older deca-formulation (Bromkal 82-ODE, Chemische Fabrik Kalk, Koin, Germany), which may indicate improvements in manufacturing practices of the later product, fueled by growing environmental concerns over products containing trace levels of the lower brominated PBDEs [Table. 1]. Deca- and octa-formulations are primarily used to flame-retard polymers employed in electronic devices (e.g. computers and televisions). The deca-product is also used to flame-retard textiles. Researchers have reported that the less brominated diphenyl ethers (e.g. BDE-47, and -99) have become globally distributed, akin to other persistent organic pollutants such as PCBs. Although, BDE-47 and -99 are the most common congeners reported in the scientific literature, researchers have also reported nearly 40 additional congeners in various biological matrixes (La Guardia et al., 2006) [Table 2]. Of these congeners 13 have not been previously reported as components of commercial PBDE formulations and may result from debromination of higher bromine substituted PBDE or preferential bioavailability of trace congeners not yet identified as constituents of these products. Also, these may be from formulations not manufactured by American or European suppliers (e.g. China). In contrast, deca- the PBDE product in greatest demand (globally 56,418 MT in 2003 (BSEF, 2006)) has not been regulated. BDE-209 has primarily been H 'Bb a a, ^ o o O o ■v}- ■ O_ VOo VO ID 6X) £ ^ ?5 D) o o (N| (N V £ ^ Ov (N^ OvCsJ QO • Td °* OO «D V£) i« §6 JS M i2 g $ 5 s- * VO ^ VO^ CN£i (D 60 O . £ I Tj t}- tN h'r o ^ s o II -Sf vd d <=> a 6bc 'Ss> "o• pN E« | rl 'd a cq (N « « o S ° -M a 9° ^° 9 c a B a © Sh VO CO CO CN O p CO • P* O o O O p 1 6 so SO so so *2 'T 4- oo -d so a 'Bb o o o o C rr p c p a» & & o o o o o o o o cn VV V V V V V V 3 •PNO ASm VO "2 (S nd nd nd O C N d © -w <0.24-2.5 nd nd-0.009 2 210 nd 4-4791.0 0.23- 0.021-36 0.26- 270- 80-2300 1,640- tq 9.1-220 w L32 1_33 16000 10,000 °9 ■21 ■21 21 ■21 ■21 » X Q ^ OQwQ3 .Q " CdwCQ^CQ^CQ^ .Q^.Q ,Q -' s t3 VO Vo d "■5 'i ‘S £ is Q ^ Q > Q Q 59 «5 OQ OQ s ^ ^ ^ S >' ''I' 2 2 -2 *y -T (» Vo ^ V V > OQ V > OQ v . - K » - Si - - Si .. - Si “ * Si- - Si hexa-BDE hexa-BDE fcd X 'fr hexa-BDE *N V > 5^5 -V) «N 2,2', 4 ,4 5 ’, ,5 ’- Q O O (N I f V a a m 6j0 !L ^ h > d ST r- ST) * d > • pN D. 3m -w CQ sr £ . ° . ° 5 T3 o g o g - o T3 TJ •O C SO P , SO P rj G G V On V CN G —< (N • opN c> •"o p* ■O T 3 0 CQ 2'M - J fl T 'S i 43 L CT E a o 3m • pN ► S3 a> Iz> • opM Sm CQ - a ^j- VO VO CO CN OOI i n CO Os i n CN r - Tf~ n T Os o * n o CN r- in c o o o "Cf C" o o o I CN CN o H S O ^ CN C O O s os'1 sf Os Q> I a. o „ C O 'SS •ih ^O r7 O s O s *R so in PQ ^ Oo" rR o CN ^ 53 1 /—N tlQ O O in W5 ^ CN £ ° mT S ^ f os o P-H ° 2 (N ^ § v o .S ‘c O PQ R ^ S3 si S3 O g « 2 ^ •§ o »-• as ,'- 4 d , in s Ql VO J* «R © o O & 2 * n 33 O o Cj CN .a ^ c a r 3 4 OO W> . R s <=* ; - ! £ O tJ) O £ $ cK cd . a . © ; S.S* 3 ^ R rt . •"■* o tt 0 •“» & T3 wT o © kj: ^ •h o M S' s* ffi «* 8 s CN t§ g 3 .2 -MCd CZS fc-H a a s 1 2 ,= ^ * M o s - o" £ g. -rh3 • r ro 2 £* O K T W ^ si 5 J U • mT >1 o o ^ cd Sm ^ t i • pN cd > O bB c r j S as p c i O ' a s H ^3 2 Q ,- m • r Q • m t O S3 r cd >—1 • r O cd Cd MM Sm C S ^ d a s " J o 3m e d cd 2 s^ . mh*' H ~ o $ Q O S £ . § - £ 00 • r b HH ‘ r m3 cn . r TJ £ £ « ) •SMa s - ® X 2 PQ 3 oo o d 3 rj C a s § -"W a s SM 2 5 T Icd a s a s ^3 m3 13 o o « . d 0 T3 • r o o - CA Q cd w ^■r^3 £ ^ Si Q H ^ 1 3 PQ 2 o Sm ■o B £ Pm a s , o a s a s 2 > © 3P cn j3 o 3 o o o 2 s N o - r o £ ffi cd" Ph * " • ~ P h X> K „ C/3 cd cd _ r . W> >< O r Sm bX) r H r Q cd O a s O pQ a s - - O CA oS h a s t 3 cd pT CA o a s o § S 2 a ^ s a s 'T ’’“! •>—1 cd •’-l si a >< PQ p H M-l H W a s Sm cn co 'd" in so oo on O£m a s C* 11 reported from sites near points of initial release and levels in most biota to date have been comparatively low. This has been attributed to its reportedly low bioavailability and tendency to strongly bind to sediment and soil. However, BDE-209 has been detected with increasing frequency as more laboratories improve their analytical methods. Watanabe and Tatsukawa (1987) reported that BDE-209 debrominates when dissolved in certain solvents and subjected to UV irradiation. Major homologues observed were tri- through octa-PBDEs and a host of brominated dioxins and furans. Two additional studies (Eriksson et al., 2004 and Soderstrom et al., 2004) also observed photolytic debromination of BDE-209 dissolved in solvents and associated with artificial and natural sediment, soil and sand. Both studies observed an increase in the lower brominated (hexa- through nona-) PBDEs. However, most of the octa- through hexa- BDEs observed were not specifically identified. Soderstrom identified BDE-47, -99 and -154 on silica gel originally amended with BDE-209. Soderstrom also reported -209 dissolved in toluene and exposed to UV light generated BDE-99, -100, -153, and -154. This route of degradation remains in question. Approximately 3% of the deca- product is used to flame-retard textiles through broad application back coating (WHO, 1994), which may make it more prone to environmental release. The remaining 97% of the deca- product is used in manufacturing high-impact polystyrene (HIPS) for electronic equipment, e.g. computer and television housings. This incorporation and the chemical properties of BDE-209, namely low water solubility (< 20-30 pg/L) and vapor pressure (< 1 O'6 mmHg at 20 °C) (WHO, 1994) are believed to minimize the amount of BDE-209 released from the HIPS matrix. This will limit its photo-degradation and bioavailability 12 potential. However, recent studies have shown that household dust can contain PBDEs, including BDE-209, at mg/kg levels (EWG, 2005). Also, industrial discharges may contribute to deca- releases. According to the U.S. EPA Toxic Release Inventory (http://www.epa. gov/tri/ (TRI, 2006)), total industrial releases on and off-site to land (e.g. air, surface water, and landfills) of BDE-209 from 1988 to 2004 averaged over 500 MT per year. Approximately 90 MT more per year were released to wastewater treatment plants (WWTPs). Industry with household waste (containing PBDE laden dust) transferred to WWTPs may sequentially release deca-PBDEs and associated degradation products to the environment through their aqueous effluent or land disposal through sewage sludge. Few studies have examined the dietary uptake and biotransformation of BDE-209. However, BDE-47, -99, -153 and several nonspecified hexa- to nona-PBDEs were detected in juvenile rainbow trout (Oncorhynchus mykiss) fed cod chips spiked with BDE-209 (Kierkegaard et al., 1999). PBDE concentrations in the liver and muscle tissue increased with length of exposure. Preferential uptake of less brominated congeners could not be ruled out because some of these congeners were present as minor constituents in the spiked solution. However, BDE-153, -154 and an unidentified octa- PBDE were not detected in the original deca- mixture, indicating likely transformation of BDE-209. More recently, juvenile carp (Cyprinus carpio) were fed BDE-209 (>98% purity) spiked food for 60 days (Stapleton et al., 2004a). At the end of this study BDE- 209 was not detected in the carp tissues, however seven other PBDEs were observed and accumulated over time. These possible metabolites were identified as penta- through 13 octa-PBDEs and two metabolites were positively identified as BDE-154 and -155. Dietary exposure of carp to BDE-183 or -99-spiked food, resulted in apparent metabolic debromination of both congeners (Stapleton, et al., 2004b), i.e. conversion of BDE-99 to —47, and -183 to -154 and an unidentified hexa-BDE. Approximately 10% of the original exposure weights of BDE-99 and -183 were detected as BDE-47 and -154, respectively. This metabolic debromination might also be species dependent, which would account for the low concentrations of BDE-99 (<0.1% of the total PBDEs) reported in Virginia carp (Hale et al., 2001a). In contrast, BDE-99 constituted 30.3% and 35.4% of total PBDEs in channel catfish and surficial sediments collected from the same location. In a follow-up to the carp exposure study, juvenile rainbow trout were exposed to -209 via the diet for five months (Stapleton et al., 2006). BDE-209 concentrated in the liver. Several hepta-, octa-, and nona-BDEs congeners also accumulated in the trout’s liver. To determine whether the observed debromination was metabolically driven, liver microsomes were prepared from both carp and rainbow trout and incubated with BDE- 209. In the trout liver as much as 22% of the BDE-209 mass was biotransformed, primarily to octa- and nona-BDEs. About 65% of -209 was transformed to hexa-BDEs in the carp, indicating species dependent metabolic debromination. PBDEs have also been detected in human blood; including the higher brominated PBDEs (e.g. hepta- through deca-BDEs). In a study from Japan, 156 blood samples were analyzed and 18 individual PBDEs were detected in most samples, totaling from 5,400 - 15,000 pg/g l.w. (Takasuga et al., 2004). Congeners detected were primarily the lower brominated PBDEs (mono- to hexa-BDEs), but BDE-183 and -209 were also observed, ranging from 56-2,800 and 1,300-31,000 pg/g lipid weight (l.w.), respectively [Table 2]. Of the 18 congeners detected in these samples, four have not been previously reported as components of common commercial mixtures, possibly indicating metabolic debromination. In another study, which included blood from a presumably non- occupational PBDE exposure group (i.e. clerical workers), hexa- through nona-PBDEs were detected in each of the donors (Sjodin et al., 2001). Deca-, penta- and tetra-PBDEs were only detected in a few samples. BDE-99 was found at the highest concentration (37,290 pg/g l.w.), but was only reported in 8 of the 12 samples. An unnamed nona- and octa-BDE (nona2- and octa2-), along with BDE-183 and BDE-153, were detected in each of the samples, ranging from 84-to 2112 pg/g l.w. BDE-209 concentrations exceeded all but -99 in 5 of 12 samples. Another study analyzing blood serum from Swedish electronic-dismantling workers reported penta- through deca-PBDEs, with BDE-183, a hepta-PBDE, being the predominant blood contaminant (Sjodin et al., 1999). This Swedish study also estimated the half-lives of BDE-209 and -183 as 6.8 and 86 days, respectively. These studies suggest that the higher brominated diphenyl ethers are indeed bioavailable. Microbial reductive debromination of the octa- and deca-technical products has recently been demonstrated. Jianzhong et al. (2006) exposed these to anaerobic bacteria commonly found in wastewater treatment. Hepta- and octa-PBDEs were produced by the Sulfurospirillum multivorans culture from deca-. Hepta- through di-PBDEs were produced by Dehalococcoides-containing cultures exposed to the octa-mixture. BDE- 154, -99, -49 and —47 were identified among the debrominated products. PBDEs have also been detected in U.S. biosolids (n= 11) (Hale et al., 2001b). The term ‘biosolids’ refers to wastewater sewage sludge that has been further processed for land application, due to their nutrient contents. This practice is becoming more common at agricultural and mine reclamation sites. Biosolids are also used in home gardens and public spaces such as parks. The major PBDE congeners observed in the above study were BDE-47, - 99 and -209. BDE-209 was detected up to 4890 pg/kg, dry wt (d.w.). Penta- related PBDE concentrations exceeded European sludge values by 10- to 100-fold, mimicking the relative commercial demands for this BFR in Europe and the U.S. (Hale et al., 2001b). Interestingly, few differences in biosolid concentrations were apparent as a function of WWTP sludge stabilization process used or geographical location for the tetra- through hexa-PBDEs detected. The commercial penta-formulation (e.g. DE-71, Great Lakes Chemical, West Lafayette, IN) exhibits a PBDE congener pattern closely resembling the pattern detected in each of the biosolids, suggesting this product is likely the source of contamination and not microbial debromination of the higher brominated PBDEs (La Guardia et al., 2004a) [Figure 2]. Although industrial contributions to the WWTPs were not specifically characterized, it is believed that sources of BDE-209 may be industrial, e.g. due to potential releases from back-coating textiles, based on the high and variable concentrations seen in these samples. However, studies have shown that household dust, which may be transferred to WWTPs through household plumbing, can contain PBDEs (including BDE-209), at concentrations similar to those observed in sewage sludge (Hale et al., 2006). PBDEs have also been detected in WWTP effluents in the Netherlands (de Boer et al., 2003). Both BDE-47 and -209 were observed in the filtered solids of nine effluents; mean concentrations were 22 and 350 pg/kg (d.w.), respectively. This suggests that particle-associated PBDEs are not being completely removed from the waste stream during the treatment process (de Boer et al., 2003). In another study, river sediment in proximity to WWTPs along the Viskan River, Sweden, contained elevated levels of BDE-47, -99, -100 and -209 (Sellstrom et al., 1998). The concentration of BDE-209 was almost 300 times higher than that of BDE-47 in these sediments, 16,000 and 54 pg/kg (d.w.), respectively. These finding suggest that some PBDEs entering waste streams may eventually be released to receiving waters via wastewater effluents. The higher brominated PBDEs, once released, may also come in contact with conditions favoring their debromination, enhancing their toxicity and ability to bioaccumulate. Figure 2. Tetra- to hexa-BDEs in biosolids compared to DE-71 composition 1000 □ Biosolids, n=11 ■ *DE-71, % Composition BDE47 BDE100 BDE99 BDE154 BDE153 (*) Values were derived by multiplying the mean (n =11) of the "penta-PBDE" totals (BDE47, -100, -99, - 154, -153) by the % composition o f each congener in the commercial formulation DE-71, as reported by La Guardia et al., 2004. To inform communities and citizens of chemical hazards in their areas, the Emergency Planning and Community Right-to-Know Act (EPCRA) was enacted in 1986. 17 EPCRA Section 313 requires EPA and the States to annually collect data on releases (e.g. air emissions and surface water) and transfers (e.g. landfills and WWTPs) of over 316 toxic chemicals from industrial facilities. These data are available to the public in the Toxics Release Inventory (TRI). Deca-PBDE is a "high-use" chemical in the U.S. and is the only PBDE product listed on the TRI. In 2001 the TRI listed 27 facilities releasing more than 4.5 MT of BDE-209 in the U.S (TRI, 2006). The total 2001 releases from these facilities were approximately 500 MT. The top releases came from deca-PBDE chemical manufacturing and chemical waste management facilities. The fifth largest reported amount from an individual operation, 45 MT, was from a facility located in Roxboro, North Carolina. This facility (Collins and Aikman Products Company, Cavel Plant), according to the TRI, is the only reported discharger of BDE-209 in the Hyco River watershed. This plastic goods manufacturer is not reported to discharge directly to surface waters, but sends its wastewater to the Roxboro WWTP. Preliminary findings (October, 2002) revealed BDE-209, at 12 |ag/L, with trace levels (< 20 ng/L) of BDE-47 and -99 in the Roxboro WWTP effluent (La Guardia et al., 2003). PBDEs have also been reported in fish and sediment samples collected from the Roanoke and Dan River basins, Virginia, U.S. The Hyco River is a tributary of the Dan River. However, major sources of PBDEs contamination within these basins were not apparent (Hale et al., 2001a). Major congeners of the penta-formulation (BDE-47, -99, - 100, -153, -154 and -49) were detected. Again, BDE-47 was the most abundant congener reported, followed by -100, -99, -49, -154 and -153. BDE-47 was detected in 89% of the 332 fish sampled and 22% of the 133 sediment samples (quantitation limits, 5 pg/kg, (l.w.), and 0.5 pg/kg, (d.w.), respectively) from these basins. BDE-47 contributed 40-70% of the total PBDEs observed. At 16 sites, the BDE-47 concentration in fish samples exceeded 1000 pg/kg, (l.w.). These sites were distributed throughout the Dan and Roanoke River basins indicating the likelihood of multiple point sources. The highest total PBDE concentration detected was 47,900 pg/kg, (l.w.), in a carp from the Hyco River. This exceeds the previously reported highest PBDE concentration in fish tissue (perch, 36,900 pg/kg, l.w.), from the Viskin River, Sweden (Sellstrom et al., 1993). However, unlike the Viskin, which flows through industrialized and urbanized centers, the Hyco is situated in a rural agricultural setting. No readily identifiable local source of PBDE contamination was apparent except for the Roxboro WWTP, located approximately 40 kilometers up stream from the carp sample site. This indicates that this facility may be a major source of the observed Hyco River PBDE contamination. The high-observed concentrations of PBDEs in this watershed provide an opportunity not only to document a release source, but also to study the potential bioavailability and degradability of BDE-209 under realistic environmental conditions. STUDY GOAL Some PBDEs are recognized as persistent, bioaccumulative, toxic pollutants. Rising environmental concentrations have been acknowledged as a human and ecological health concern. Historically, studies of PBDE in the environment have primarily focused on ambient concentrations of the major congeners of the penta- formulation (e.g. BDE-47 and -99). However, following restrictions on the use of the penta- and octa-formulations attention has shifted to the historically more used and still unregulated deca-formulation (>97%, BDE-209). However, the sources and routes of its release to the environment have been poorly investigated. Some have argued that fish exposed to -209 can metabolize it to lower brominated PBDEs, resulting in increased toxicological burdens. Also, -209 has been reported to undergo photolytic debromination in the laboratory, producing lower brominated PBDEs, with enhanced toxicological potentials. However, these degradation routes have not been observed under realistic environmental conditions. This study will assess the extent to which BDE-209 and other PBDEs, associated with a wastewater effluent (WWTP Roxboro, NC), are being distributed within the surrounding aquatic environment. It will assess if these PBDEs are bioavailable to aquatic species related to this outfall and if PBDE debromination by products are likely being formed. Gas chromatography (GC) coupled with electron- capture negative ionization (ECNI) selective ion monitoring (SIM) mass spectrometry (MS) will be used to screen samples for mono- through deca-PBDEs and tentatively identity them based on their relative retention index (RRI). Detected PBDE congeners will be confirmed by “full-scan” electron ionization (El) and ECNI based on molecular 19 structure spectra. WWTP sludge will be analyzed to estimate the PBDE burden of the waste stream. PBDEs are hydrophobic compounds and have been reported to partition (>90%) to wastewater solids (North 2004; Song et al., 2006). Sediments collected near the wastewater outfall will be screened and PBDEs therein compared to those in sludge and to those previously reported in the technical products. PBDEs congeners detected within these samples, for which no reference standards are available, will be quantified against available reference compounds of equal bromine substitution. Biota samples collected near the associated outfall will also be analyzed for PBDEs similarly. This study will provide valuable data for assessing the persistence and bioavailability of specific PBDE isomers related to WWTP effluents and associated sediment. Sediment samples will also be collected from several sites downstream from the outfall (Roxboro WWTP outfall to the Hyco River), which will be valuable in determining the pattern and extent of contamination of the river system. All samples will be screened for PBDE debromination products. The latter process may provide evidence of debromination of the widely used deca-product or the other highly brominated PBDEs. STUDY SITE AND SAMPLE LOCATION STUDY SITE: The Hyco River flows northeast from North Carolina and enters the Dan River downstream from South Boston, Virginia. The Hyco River watershed, in North Carolina, includes Hyco and South Hyco Creeks and Hyco Lake, as well as Mayo Creek and Mayo Reservoir watershed [Figure 3]. Other waters within the Hyco River sub-basin are Ghent, Storys and Marlowe Creeks, Roxboro Lake and Lake Roxboro. All major streams generally flow northward into Virginia. The headwaters of the Hyco are dammed to form the After Bay Reservoir located approximately 35 kilometers upstream from the Dan River. The After Bay Reservoir forms Hyco Lake (also known as Carolina Power Lake) and covers 3750 acres. Carolina Power and Light Company (CP&L) constructed the reservoir in the early 1960's, to be used as a cooling reservoir for their steam electric generating plant. Water released from the reservoir flows down the Hyco River through rural and agricultural areas. Lands within this sub-basin are mostly low rolling hills, characteristic of the piedmont region. Land use is dominated by forest (71 percent) and agriculture (22 percent) (NCDENR, 2000). Population according to the 1990 census was 9903 and is expected to increase 6 to 14 percent from 1998 to 2018 (census 2000, City of Roxboro NC. population 8698). There are seven permitted discharges to this sub-basin, three are major effluents, discharging greater than one million gallons per day (MGD) rhttn://h2o.enr.state.nc.us/NPDES/documents/docs/nermits2.ndf|. CP&L, Roxboro and Mayo generating plants discharge tens of millions of gallon per day (MGD) of cooling water to the Hyco Lake and Mayo Reservoir. The Roxboro WWTP holds two permits 21 Figure 3. Hyco River subbasin, North Carolina, U.S. (NCDENR, 2000) \ \ ^ ^ \ \ \ ' u &V & \ ) !% a hU( / ~ k * slflS (J3°0^HN)->0 0 q ' ^ 22 and is the other large discharger, releasing an average of 2.1-MGD into Marlowe Creek, down stream from the City of Roxboro, NC. During dry months, this discharge is more than 99 percent of the total flow in Marlowe Creek. Cogentrix, a power generating facility provides electricity to CP&L, discharges cooling water to Mitchell Creek (a tributary to Marlowe Creek). The other two outfalls are for Caswell and Person County Public Schools, which discharge treated sewage into the North and South Hyco Creek upstream from Hyco Lake. All facilities that release to surface waters in North Carolina are required to monitor their effluent’s toxicity. Historically, a few discharges have been out of compliance with permit limits and have failed toxicity tests. Since 1994, Marlowe Creek water quality has improved due to improvements at the Roxboro WWTP. However, during the 1999 basin monitoring cycle Marlowe Creek’s bioclassification rating was only upgraded from poor to fair, leaving the waterway listed as impaired (NCDENR, 2000). SAMPLE LOCATION: Samples were collected in the fall of 2002 [Table 3] to determine if PBDEs used in local manufacturing were transferred for further treatment to Roxboro WWTP and then subsequently released to the environment. The extent of PBDE contamination from Roxboro’s outfall to where Marlowe/Storys Creeks enters the Hyco River was also investigated at this time, along with possible alternative PBDE sources. A follow-up was conducted in the fall of 2005, by collecting samples at the outfall and several locations downstream [Table 3]. These were analyzed to ascertain if additional PBDE exposure Table 3. Sample IDs, location, %lipids and %TOC £ £ O C 3 T H P 55 ) x o a> O) ? V m o c A v v ’A U « ? ^ 5 l C5 "S flj o o Q £ % a d N= o H u o o _ ) X in o 2 J Ui 3 - T3 T3 •X373 PQ PQ £ £ o o o o t—cn i !O ! 10 CN pP J3J C/3V3 CN CN CN in oo CN O C 'S ) 0 s a> cs c» u m s s a, % CQ QJ O G u H o a> 8 (NtN(N(NN(SMMMMN^JQ^ •5 o o o PQ PQ o PQ PQ o PQ o GO o 0 0 0 0 0 0 0 ooooooooooo^^<^< GOGO *—H cn o o CM o o o o o o o o o o o o o o o o ooooooooooas^oooooooooooooooooo tj o o o o o o o o o o o o o o o o CN Co Co Co CN ^ to cnCN fnTj-vooots^v^ONr^'Ov^mrj-oor^ON Z£Z£ZZ£ZZ.£££££££ 0 0 0 0 0 0 0 0 0 0 0 00 00 00 00 00 00 00 00 0 0 o o o GO .y —i r in o vi vio ro in « o o o fO o o d •3 SO GOGO _d OO OS rf D(U *0 .5 — — 1—1 1—4 1—1 t-~ <\ c< m oo os 1M IO o *3 o o o CO o GO GO CO pp QP PP QPQ PQ PQ o PP PQ PQ pp o pp o o OS oo GO _d o so •n ^-HOOr^OCNCNOQvC^-OO 4-» D( CQ ■ c o « - j C/3 u *3® a> 3 + a, % -*2 qj w O 0 s i > o ‘a 2 H CQ a a> cm S£ T S'-g^ 'S S C3 i+4 o G-i in o m o o o o o o o Ifl rH m (N O Ooo O oo *-H ooO o O(N O O (N (N in H P iM n i PQ PQPQ b pH PQ h ms in _ m j-J 2 (D d C T cd r-H CN C C 1 P m M l l M cn m D m d y m cd d O 3 h rv 24 25 occurred, since samples were last collected three years earlier, and for potential environmental PBDE debromination products within this river system. Roxboro WWTP is an activated sludge-type secondary treatment plant. Its treated effluent is discharged on site to Marlowe Creek, and its activated sludge (~2% solids) is pumped to dewatering lagoons located within the facility’s property. Approximately, twice a year these lagoons are cleared of remaining solids (-30% dry weight). This material is trucked off location to landfills or land applied for disposal. PBDEs are persistent lipophillic compounds ((Log KoW 4-8, > 8 BDE-209) (WHO, 1994)) and once introduced into waste streams concentrate in residual wastewater solids. Some of these solids (particulates) are released through treated wastewater effluent, introducing PBDEs to receiving waters. However, most solids are removed from the waste stream and retained within the treatment plant, as are >90% of the PBDEs that enter the WWTP (North, 2004; Song et al., 2006). Therefore, the analysis of the retained solids may be a better indicator of temporal wastewater exposure to persistent lipophillic compounds than grab or composite influent sampling. Also, these solids may concentrate trace PBDE components of the technical formulations within the waste stream or potential PBDE degradation products that may occur during the treatment process. These constituents might be below detection in the effluent. To monitor PBDE introduction to the Roxboro WWTP and potential exposure to PBDEs in its receiving stream, activated sludge (4-L) was collected prior to transfer to drying lagoons in 2002 and 2005. Also, to determine the impact of PBDEs on the receiving stream, sediments and biota were collected near the outfall in these same years. This receiving stream is effluent dominated. Therefore the sample site chosen, approximately 15 meters downstream from the outfall, was presumed well mixed. Surficial sediments (1-L) were collected at this site and minnow traps for the opportunistic collection of biota (baited with PBDE-free squid) were placed on both sides of the stream near this site. Traps were emptied approximately 24 hours after deployment and samples separated by species and transferred to holding tanks for depuration. In order to determine the extent of PBDE contamination of this river system, sediment samples (1-L) were taken upstream and several locations downstream from the outfall at the confluence of Marlowe/Storys Creek and the Hyco River. (Once introduced into aquatic environments, wastewater particulates with associated hydrophobic PBDEs settle out and contribute to sediments (Oros et al., 2005).) Sample sites were distributed along approximately 11 kilometers of river system and several locations were re-sampled in 2005 [Figure 4a and b]. Also, to investigate possible alternative PBDEs sources to the river system, additional sediments were taken at accessible locations on tributaries (Ghent Creek (#0BS018) and Storys Creek (#0BS019)), which feed into Marlowe Creek, downstream from Roxboro WWTP outfall [Figure 4b]. These samples were only taken during the initial 2 0 0 2 collection. Figure 4a. Sample locations on lower Marlowe/Storys Creek, Parson County, NC. iue b Sml lctos n pe MroeSoy Cek tiuais Get n Soy Cek) Pro Cut, NC. County, Person Creeks), Storys and (Ghent tributaries Creek, Marlowe/Storys upper on locations Sample 4b. Figure HO*rot* SAMPLES, METHODOLOGY AND QUALITY CONTROL SAMPLES: Samples (WWTP sludge, surficial sediments and biota) were collected during the summer and fall of 2002 and fall 2005 (See Sample IDs, Location, Table 3. for type and number of samples.) Sludge grab samples were collected in 4L glass amber bottles. Sediment samples were collected in 1L glass amber jars. Amber glassware was used throughout to minimize potential photolytic debromination of BDE-209. Amber glassware were used throughout to reduce the possibility of photolytic debromination. BDE-209 in particular has been reported vulnerable. Minnow traps were placed approximately 15 meters downstream from the outfall and specimens collected the following day. Sludge and sediments were placed on ice in the field. Once in the laboratory sludge samples were kept at ~ 4 °C, sediments < 0 °C. All biota samples were depurated for 72 hours prior to being sacrificed to eliminate PBDEs associated with gut contents. During the depuration portion of the study a wire mesh was placed on the bottom of the tanks to separate biota from their feces. Tanks were cleaned and water replaced daily during the depuration period. METHODOLOGY: Whole fish were homogenized and then freeze-dried. Sediments were freeze- dried and then sieved (2000 pm) to remove large debris. Sewage sludge was centrifuged and excess water removed. Remaining solids were then freeze-dried. All samples were stored in glass jars with Teflon® lids at <0 °C until analyzed. (A sub-set of the first 29 30 group of samples ( 2 0 0 2 ) was initially analyzed to determine contaminant content and assess the analytical method. These data have been presented elsewhere (La Guardia et al., 2003 and 2004b). All sample data presented here were analyzed in 2006.) Sediment and sludge sample data were normalized based on Total Organic Carbon (TOC) [Table 3]. TOC analysis was conducted by thermal conductivity detection (Exeter CE440, Chelmsford, MA); inorganic carbon was removed by addition of hydrochloric acid. Biota concentration data were expressed on a % lipids basis. For PBDE determinations, samples (0.5-g sludge, 9 to 10-g biota, 20-g sediment, d.w.) were subjected to enhanced solvent extraction (Dionex ASE 200, Sunnyvale, CA). ASE-200 conditions: two extraction cycles, pressure @ 1000 psi, temperature @100 °C, heat 5 minutes, static 5 minutes, 60% flush, purge 180 seconds. Approximately 30 mL of dichloromethane (DCM) was used per sample. A surrogate standard (1-pg) 2,2’,3,4,4’, 5,6,6’-octachlorobiphenyl (PCB-204) (Ultra Scientific, North Kingstown, RI) was added prior to the extraction. For biota % lipid determinations (dry weight basis), 10% of the ASE extract was dried and the residual weighed. Remaining extracts were reduced to 5 mL under nitrogen, and purified by size exclusion chromatography, (SEC, Envirosep- ABC®, 350 x 21.1 mm. column; Phenomenex, Torrance, CA). The column was eluted with DCM at 5 mL/min. The first 50 mL, containing high molecular weight lipids, were discarded. The next 60-mL, containing the compounds of interest, were collected and solvent exchanged to hexane. Each post-SEC extract was added to a 2-g silica glass column (Isolute, International Sorbent Tech., Hengoed Mid Glamorgan, UK) and eluted with 3.5-mL hexane, followed by 6.5 mL of 60:40 hexane/DCM. The second fraction, which contain the halogenated compounds of interest, was reduced in volume and solvent exchanged to hexane. Pentachlorobenzene (PtClb) and decachlorodiphenyl ether (DCDE) (Ultra Scientific, North Kingstown, RI) were added. DCDE was also used as an internal quantitation standard. Compounds of interest in the purified extracts were separated by gas chromatography (GC), (6890N, Agilent Tech., Palo Alto, CA) equipped with an on- column and a split/splitless injector. Ion fragmentation spectra used for compound identification were produced by electron-capture negative ionization (ECNI) and electron ionization (El) (JMS-GC Mate II, JEOL, Peabody, MA). A DB-5HT (30 m, 0.25 mm i.d., 0.1 pm) column (J&W Scientific, Agilent Tech.) was used to resolve compounds of interest. (Historically, two analytical columns have been suggested for PBDE analysis: a longer column (> 30m) to chromatographically separate the complex suite of less brominated PBDEs and a shorter (< 15m) one for the thermally labile BDE-209. Recently, it has been reported that mono- through deca-PBDEs can be reliably analyzed using a single 30-m DB-5HT column (La Guardia et al., 2006).) All samples and calibration curve analytical standards were first introduced (1-pL) into an on-column injector at an initial injector temperature of 65 °C, carrier gas helium. Temperature was then increased to 150 °C at 30°C/minute, then 10°C/minute to 300 °C, and held for 15 minutes. It was then increased to 350 °C at 30°C/minute and that temperature held at for 5 minutes to bake-out the injector. The GC column oven was programmed to follow the injector temperature ramp program. Poor chromatography (unresolved peaks) was observed for the biota and sludge samples, perhaps due to column overloading by co extractives. Accordingly, these samples (biota, sludge and an additional set of calibration standards) were reanalyzed using a pressure pulse split/splitless injector, which greatly improved the chromatography. Samples were introduced (1-pL) into the split/splitless injector, equipped with a glass liner (1 mm, ID), injector temperature 300 °C and pressure 50-psi, carrier gas helium. The split vent was opened and pressure was reduced to 15.2 psi (1.5 mL/min.) after 4 minutes, following sample injection. Thereafter column flow rate (1.5 mL/min) was kept constant throughout the remaining portion of the run. Initial column oven temperature was 90 °C, held for 4 minutes, then increased to 150 °C at 30°C/minute, then 10°C/minute to 300 °C, and held for 15 minutes. It was then increased to 350 °C at 30°C/minute and held at 350 °C for 5 minutes, as a bake-out procedure. Figures 5a, 5b show chromatographs of a multi-component PBDE mixture, plus internal and surrogate standards, produced using the on-column [5a] and split/splitless [5b] injectors in the ECNI-selective ion monitoring (SIM) mode m/z 79 ([ 79 Br]'), 81 ([ 81 Br]') for PBDEs and m/z 35 ([35C1]"), 37 ([37C1]') for PtClb, PCB-204 and DCDE. A low temperature on-column injection technique has been reported to reduce the likelihood of PBDE debromination due to prolonged exposure to high injector temperatures common with split/splitless injectors (Covaci et al., 2003). However, split/splitless injectors have the capacity to accommodate dirtier samples and thus are more robust for routine analysis than on-column injectors. When optimized, it has been reported that split/splitless injectors perform similarly to on-column injectors for the lower brominated PBDEs, but response declines for the octa-, nona and deca-BDEs (Bjorklund et al., 2004). An improvement in PBDE response with minimum degradation 33 Figure 5a. Representative chromatograph of 27 PBDEs, ptclb and DCDE by on- column injection, ECNI-SIM (DB-5HT, 30 m, 0.25 mm i.d., 0.1 pm) 1,600,000- 00 00 1.400.000- 1.200.000- 1.000.000- Tt©\ r-r-> No t" 800,000- 0\ No 600,000- ON 400.000- o n 200.000- 81,79 . iLiL DL J l i i L 560.000 480.000- ptclb 400.000- 320.000- DCDE 240.000- 160.000- 80.000 37,35 'I r t i r ~i—i—i—r r p - r m —r*r~ r~r ~j—r ■T ) r-T -i i i | r~f rrn-p •r-r—i—i— t—r Min. 5 10 15 20 25 30 35 Figure 5b. Representative chromatograph of 27 PBDEs, ptclb and DCDE by split/splitless injection, ECNI-SIM (DB-5HT, 30 m, 0.25 mm i.d., 0.1 pm) 320,000- 00 160,000- fS Min. of the thermally labile PBDEs has been seen with pressure pulse split/spitless injectors capable of temperature programmable gradient (Bjorklund et al., 2004). To compare the two injection techniques used in this study, 50 ng of BDE-209 (purity, > 97%) were introduced via both injectors, employing identical GC/MS conditions (e.g. column and ECNI mode). The area ratio between -209 and potential debromination products were monitored. The response for -209 using the on-column injector was 3.4-times greater than with the split/spitless injector. However, the area ratios between -209 and the sum of two nona-PBDEs (BDE-206 and -207) were 1.3% (0.08%, RSD) for the on-column and 2.4% (0.08%, RSD) for the split/splitless injector [Figure 6 ], indicating slight debromination of -209 with the split/splitless injector. However, BDE-206 and -207 were the only other PBDEs observed and their contribution represented only 1% of the total -209 present. The target analytes were first detected and quantified by ECNI-SIM of m/z 79 ([79 Br]'), 81 ([SIBr]') for the PBDEs and m/z 35 ([35C1]), 37 ([37C1]‘) for the internal and surrogate standards. Electron energy of 62.1 eV was used for ionization of the reagent gas (methane). The GC/MS interface was maintained at 300 °C, filament delay of 3 min, ion source temperature at 300 °C, pressure of 3.3 mPa, and scan time of 1 sec. The electron magnet and optical path were calibrated using perfluorokerosene (PFK) and 1,2,4-trichlorobenzene at a 1000:1 mixture ratio. El and ECNI conditions were optimized by monitoring m/z 381 and for ECNI-SIM m/z 85. Quantification curves were generated by analyzing dilutions of a PBDE standard (Wellington Laboratories Inc., Ontario, Canada) containing 27 PBDEs ranging from mono- to deca-PBDEs [Table 4]. 35 Figure 6. BDE-209 comparison between on-column and split/splitless injectors NCI_SIM_092906_13 BDE-209 50ug/mL on-column 700,000- On-column Injector 630.000- BDE-209 560.000- BDE-206, -207 (area) = 1.3% BDE-209 (area) 350,000- 280.000- 210,000- BDE-206, -207 70,000- 81,79 nil nri'iri n'lTpiTrrrnrrr 111 ii 11111111 j111111111111111111111111111111 Min. NCI_SIM_092906_08 BDE-209 50ug/mL S/S 200,000- Split/Splitless Injector BDE-209 180.000- BDE-206, -207 (area) 120.000- = 2.4% BDE-209 (area) 100.000- 80,000- 60.000- ^10. 000 - BDE-206, -207 20 , 000 - *n im rnr u n i n n | n u n n 1111111 n'iT|rii'iTTi'ii'HnnTffifpn n r(l» ii ii ii 11 n |i Min. 22 24 26 2E oo oo oo oo 5 3 53 36 oo oo oo oo oo No" o o o o o oo 5 3 53 Tf Tt 53 o ' o' o" o' o" oo oo oo oo oo oo cn cn cn cn cn o o o o o o 1—H * t * 5 3 5 3 r^- C— -H cn m cn cn cn 53 53 53 00 OO OO OO OO OO 00 00 00 00 OO OO OO OO OO 00 OO 3 00 OO c n c n oo oo oo .-h 1— 1 o O' o' W O n"O n"O n"O n"O n"O n"O n" on "O n"O n" O n" O n"O n"O n" O n"O n" ON O n" O n" (N (N M OO OO CO oo cn oo 3 " 3 - C— t— in" 3 3 — 3 - 3 in" 3- 3- 3- c n 3- 3~ 3~ ~ ' t t— 3c n 3 3 cn cn cn „ cn ~ cn O n O n - On ~ On 3 3 oo 3 oo 3 oo oo oo oo oo CN CN cn O n o n On On On 3 3 3 3 3 O n ON 3 3 OO oo NONONONONO 53- 53- 53- 53- 53- + o o NONONONO NO CNCN CN CNCN C7 ■'T i n m m m i n r-~ r- r- g i o OOOO NONONONO NONO NO-cf ■ct ■53- 53- 5 }- 5 t 5 t 53- 53- 5 f CN CN CN CN CN CN CN CN CN CN CN CN i n CN CN o o OO OO OO OO OO NO NO no NO NO 53- 53- f-H 53- 53- CN CN CN CN CN 0 OOOOO O CN cn cn ■'3- o ■^3- -cl- i n i n i n i n i n NO NO m NO NO r- r- r~- r - r - 0 0 OO OO 00 OO OO PQ i n CN oo oo oo oo oo NO NO "Ct NO NO NO NO NO NO NO NO 53- 5 f 53- 5 f 5 t 53- 53- 53- 53- 53- CN CN CN CN CN CN N" NO NO -ct CN CN CN CN CN o o o O O OOOO 53- OOOO NO NO NO NO NO 53- 51- 53- 53- 51- 53- + CN i— i »—i CN CN cn cn cn cn cn • s f " s f ■5J- 53- 53- 5 f 5 f 53- 5 t 53- •n i n i n i n m NO NO NO NO NO NO w NO NO TT T j- T f •5 f 5j- CN CN CN CN CN O o o O O NO NO NO NO NO CN CN -n- '53- •53- 53- 53- i n m m i n i n O 0 On On 0 in O O 0 0 On 53 NO r^ _ NO r-~ On 0 0 On NO in NO CN O 5 3 On 53 m 53 0 0 O 0 NO 0 On 53- 53 53- NO On cn ON 0 cn ON 0 CN 0 0 0 CN 53- in NO OO O On O CN CN 53 •n CN cn 53- 5 3 NO NO NO NO NO 3 - r - 0 0 0 0 OO ON ON O n OOO t-H CN CN CN CN CN O r-H 0 O 0 0 ■“11—11—1 1—1 r_l 551 1 ' ~ l' 5“* ,-H CN CN CN CN CN CN CN CN CN CN CN CN CN CN IM . W 0 0 r-~ in cn cn m CN m r- O 3^ t> NO 0 0 m cn cn 0 0 r- cn CN CN t> cn in in 3- 0 0 CN 3 ~ CN CN 0 0 NO On NO i> On CN 5 3 oo NO 0 NO 0 0 cn NO 0 0 O CN c ^ CN 3^ 3 -; cn § .g s T^" CN cn O n mi rQ rQ On O n WWW w CN cn cn 53 53- in in in NO NO NO rQ fQ 3 Q O n »—i On ON O 0 5—1 1 15“' 1 *“ 1 5-1 i—1—1 1 1 —1 H H 5-' 5 - 1 - i 1—1 1 CN CN O NO CN 0 0 cn CN 0 0 cn ON 5 3 CN 5 3 On 3 O in 5 3 0 0 cn O cn _ l CN 3 On On ON 5 3 0 0 3 NO NO O O 3 CN 3 O 5 3 in 53 3 O 5 3 ON »-H5 3 0 CN cn On O 5 3 m NO 3 ON On CN O CN cn cn NO 3 O i-H CN cn 5 3 m NO NO NO NO 3 3 3 OO OO On On On On O 0 0 O OOO »-HCN CN CN CN CN CN ^H ^H 1—1 1—11—1 ’~l 5-1 1—1 ”H11—1 1—1 1 5“' CN CN CN CN CN CN CN CN CN CN CN 3 CN CN CN m 3 CN p cn CN OO cn cn in cn 3 OO in 0 0 CN OO CN O n m OO m 0 0 cn CN 3 cn in cn 3 cn »—H H p cn in 5 3 ON °° P 0 0 N—< 5 3 On in 3^ 0 t3 O OO in in NO 0 0 O 3Q 5 3 p On cn in •n ON 5 3 in 3 Q 0 0 O W w W CN CN CN cn cn 5 3 in in in NO NO NO 3 ^ 3 3^ On On On ON O r-H r-H r—1 rH ^H ^H *—• H 1 H-1 H 1 r-H r—C rH H rH rH r-H H rH rH rH »“HCN 0 On 5 3 cn OOVOOO'tONmrHNr-i 5 3 3 cn no in 3 0 0 On 5 3 3 NO 3 On m cnmoooo3oooNOOOONOON H *—! CN 53- 3 0 53 NO 3 2 On 0 0 9 CN 1 1 PQ PQ W PQ PQ PQ w PQ PQ w w PQ W Q Q Q Q QU Q Q Q Q Q QQ QQQq PQPOQQQQQOPQQ PQ PQ PQ PQ PQP h PQ PQ PQ PQ PQ PQ PQ pq pq pq pqpqpqpppqpqpqpqpqpqpqpqpq po m m co m m co co CO CO CO g g w w w w w W >,w w w w w W w PQ PQ S3 PQ PQWW WWW w w WWW -in ■<-> B B ’O 'S .b -S a ss a a c u a f t a a a o o o o a o CN CN CN oo oo oo oo oo oo o o o o CN oo oo oo vo r- H h oo oo oo Os Os wM CN CN CN o CN CN CN CN o w-u r- r-~ r-~ oo OO :< P> o o o CN CN CN "© C"- t-~ K •2SP CN N- r- o o c vo Os m o o © tT N" in o o .2 h CN CN CN cn cn 5 •tS-3 -<->o £ & 8 CN O oo oo © Ov Tf in CN CN § ’M cn Tf’ in cn cn © ’a , cn CN CN cn cn C/3 f O oo CN o o ►S3 t" O r- o o © ss Tl- in m o o bp § »- CN CN CN cn cn =3 © ^2 -© o© “ CN t> r- CN CN CN in vq CN CN S fi .S' cn cn Tt- a o CN CN CN cn cn ii ro s oo o VO Os u o o o O s—^ CN CN CN CN On w w w w ^ K Q Q Q Q 2 £ PQ pQPQ pQ g a PQ £ m m m cNi w w W W -© Q Q t; PQ PQ PQQ QPQ U5i cU i§ a o"W Q Q CJ § -ePQ vo vq vo VO 03 vo" vo" „VO" •> T3 u -o in *>' a a in so- o "i ^ ~ vo a N-" Tf S m © rq eq u m" cn" O cq CN cn" cn" cn - m_ o Mono-through nona-BDE concentrations ranged from 10 to 1000 pg and BDE-209 ranged from 50 to 5000 pg on-column for both injection programs. The minimum r 2 value for acceptance of the five-point linear quantification curve was 0.998. To assist in sample compound identification, relative retention indices (RRI) were calculated for each of the analytical standards using marker compounds [Figure 7]. In complex chromatographic runs with close eluting isomers, RRIs improve compound identification by compensating for between-run variations in retention times. RRI [Table 4] for 27 congeners contained in the multi-component calibration standard, plus an additional seven PBDEs previously identified as potential debromination products (Stapleton et al., 2006), were calculated for both the on-column and split/splitless injector programs. PtClb, DCDE and BDE-209 were used as retention time markers for RRI calculations and assigned index values of 1000, 2000 and 3000, respectively. If the sample did not contain BDE-209, C 13- BDE-209 (Cambridge Isotopes Labs., Andover, MA), which exhibited an identical retention time, was added and the sample re-analyzed. The RRI for each peak in the samples was calculated and compared to the analytical standards RRI. Tentative identification was presumed if the RRIs were within ±3. Along with RRI, fragmentation patterns and isotope intensity spectra produced by ECNI (scan range 10-550 m/z) and El (scan range 50-1000 m/z, scan time 0.30 sec., electron energy 70 eV) were used to identify PBDEs. (The previously stated GC conditions were used for both ECNI and El analyses.) The predominant ions generated in ECNI spectra of PBDEs are 79 and 81 m/z. However, cleaving at the ether bond has Figure 7. Relative Retention Indices (RRIs) formula RRI = * 1000 + RRI x mp 4 m f 4 m p where: t* = retention time of peak X tnv = retention time of last marker compound preceding X tM" = retention time of next marker compound fallowing X R R U = KK1 defined far the last rrmker preceding X also been observed for hepta-, octa-, nona- and deca-PBDEs (La Guardia et al., 2006). These produced spectra with ion clusters centered around 328 and 330 m/z for [C6Br3H20]', 408 m/z for [C 6Br4HO]' and 486 and 488 m/z for [C 6Br50]'. For PBDE identification, bromine distribution between the two benzene rings can be determined for hepta- through deca-PBDEs by observing these fragments. In El mode, the dominant ion clusters are centered on the molecular ion [M]+ and the loss of two bromines [M-Br2]+ (La Guardia et al., 2006). Table 4 contains the predominant ions products produced by ECNI and El modes for compound identification used in this study. QUALITY CONTROL: In addition to correcting all data based on surrogate recovery, analytical blanks (NaSC>4, oven baked at 400 °C, >4 hours) were analyzed with each batch of samples (10 samples/batch). Method performance was also evaluated by the analysis of replicate samples and matrix (NaS04, sediments and biota) spikes. Spike solutions (1 mL) contained mono- through deca-PBDEs [Table 5] in hexane and were introduced to the matrix prior to enhanced solvent extraction. 41 Table 5. Matrix spiking solution* Compound IDs ng/mL 3-mono BDE BDE-3 100 2, 4-di BDE BDE-7 100 4,4’-di BDE BDE-15 100 2,2’, 4-tri BDE BDE-17 100 2, 4,4’-tri BDE BDE-28 100 2,2’, 4,4’-tetra BDE BDE-47 100 2,2’, 4,5’-tetra BDE BDE-49 100 2,3’, 4,4’-tetra BDE BDE-66 100 2,3’, 4 ’, 6-tetra BDE BDE-71 100 3,3’, 4,4’-tetra BDE BDE-77 100 2,2’, 3, 4,4’-penta BDE BDE-85 100 2,2’, 4,4’, 5-penta BDE BDE-99 100 2,2’, 4,4’, 6-penta BDE BDE-100 100 2, 3’,4,4’, 5-penta BDE BDE-119 100 3,3’, 4,4’, 5-penta BDE BDE-126 100 2,2’, 3,4,4’, 5’-hexa BDE BDE-138 200 2,2’, 4,4’, 5,5’-hexa BDE BDE-153 200 2,2’, 4,4’, 5,6’-hexa BDE BDE-154 200 2, 3,3’, 4,4’, 5-hexa BDE BDE-156 200 2,2’, 3, 4,4’, 5’, 6-hepta BDE BDE-183 200 2,2’, 3, 4,4’, 6,6’-hepta BDE BDE-184 200 2, 3,3’, 4,4’, 5’, 6-hepta BDE BDE-191 200 2,2’, 3,3’, 4,4’, 5,6’-octa BDE BDE-196 200 2,2’, 3,3’, 4,4’, 6,6’-octa BDE BDE-197 200 2,2’, 3,3’, 4,4’, 5,5’, 6-nona BDE BDE-206 500 2,2’, 3,3’, 4,4’, 5, 6,6’-nona BDE BDE-207 500 2,2’, 3,3’, 4,4’, 5,5’, 6,6’-deca BDE BDE-209 500 * stock solution (BDE-MXE) supplied from Wellington Laboratories, Ontario, Canada RESULTS RESULTS (identification and quantification overview): ECNI-SIM mode produced lower detection limits for PBDEs vs. “full-scan” ECNI or EL However it yielded less structural fragmentation information, which can lead to the misidentification of co-eluting compounds (e.g. BDE-154 co-elutes with BB- 153 (2,2’, 4,4’, 5,5’-hexabromobiphenyl) (Korytar et al, 2005). PBDEs tentatively detected by ECNI-SIM and RRI were re-analyzed by El. (See Appendix A-H, for representative El spectra for tri- through deca-PBDEs). Also, it has been demonstrated that ECNI “full scan” can offer unique additional spectra information for congeners within some PBDE homologue groups (e.g. those with > six bromines), by indicating the bromine ring substitution pattern (La Guardia et al., 2006). For example, BDE-197 and - 204 are both octa-BDEs, but the former has four bromines per benzene ring while the latter has a 3-5 substitution [Figure 8 ]. When necessary for compound identification, samples were re-analyzed by ECNI (scan range 10-550 m/z) to determine bromine ring substitution. (See Appendix I-N, for representative ECNI spectra for hepta- through deca-PBDEs.) ECNI-SIM also allows for semi-quantitative estimation of analytes present in the absence of reference standards, as in the case of potential debromination byproducts identified within the sample set. When chromatographed, PBDEs generally elute in order of increasing bromination. By comparing the response of the uncharacterized-PBDEs in ECNI-SIM mode to a reference PBDE standard with an assumed equal number of bromines, an estimation of the unknown concentration can be provided. Calibration curves were constructed from the analysis of serial dilutions of a 42 43 Figure 8. ECNI “full scan” spectra of octa-PBDEs illustrating differences in spectra as a function of bromine positioning 79 79 BDE-197 2,2',3,3'4,4>,6,6,-octaBDE [4,4] octa-BDE BDE-204 2,2\3,4,4\5,6,6,-octaBDE [3,5] octa-BDE O' O' 0 " 408 328 486 408 m /zm/z multiple component standard containing 27 PBDEs. An additional six calibration curves were also derived from the multiple component standard. Response curves for BDE-188 and -179 were estimated from BDE-184, BDE-202, -201 and -203 from BDE-197 and BDE-208 from BDE-207. Each sample was first analyzed by ECNI-SIM, using m/z 79 ([79 Br]') and 81 ([ 81 Br]'). PBDEs observed were tentatively identified by RRI and were then reanalyzed by EI. ECNI “full scan” was also used for compound conformation as needed. Compound quantification was determined by ECNI-SIM (m/z 79 ([ 79 Br]') and 81 ([81 Br]-)). WASTEWATER SLUDGE RESULTS: Sludge samples collected in 2002 (#0BW003) and 2005 (#5BW001) were analyzed by ECNI-SIM. Due to the presence of co-extracted material, split/splitless injection was used. Both sludges produced similar chromatograms and revealed tri- to deca-PBDEs. The major PBDEs observed in the ECNI-SIM chromatogram were BDE- 209, -99 and -47 [Figure 9]. All reported PBDEs were confirmed by EI [Figure 10]. A total of 17 PBDEs were identified in the 2002 sample and 18 in the 2005 sample. However, two additional PBDEs (BDE-17, -28) were detected in the 2005 and BDE-71 was only detected in the 2002 sludge sample [Table 6 ]. The major PBDE congener in both samples was BDE-209. It was present at 58,800 and 37,400 pg/kg (d.w.) for the 2002 and 2005 samples, respectively. All sludge results are reported based on dry weight. Total PBDEs in sludge from the two years were 97,400 and 42,900 pg/kg, respectively. Two additional brominated compounds were also identified in the 2005 sludge (RRI #2022 and #2129), but RRI #2022 was not detected in the 2002 sample Figure 9. ECNI-SIM chromatograph of PBDEs in wastewater sludge (#5BW001) D C J V C 90Z esi- LOl CD T c CD CD L6 SOZ 1 vo O n D C CCD CD 1 0 0 D C D C CvJ o Figure 10. EI chromatograph of PBDE major ions in wastewater sludge (#5BW001) 2a 000,000- TIC 6 0 , 0 0 0 - tri-BDEs 3 0 , 0 0 0 - /J' 4 0 1 , 4 0 6 X tetra-BDEs CS 1 ,6 0 0 ,0 0 0 penta-BDEs 000,00 0- 5 6 6 , 5 0 4 J l^ 1 6 0 0 , 0 0 0 - hexa-BDEs , 3 0 0 ,0 0 0 ______y u 6 4 4 , 4 0 4 ______L l imQQQ- hepta-BDEs 000,000; 7 2 4 , 7 2 2 ft ,A ------a A L_ ____ ...... L octa-BDEs 1600,000^ nona-BDEs 8 0 0 ,0 0 0 ^ 0 8 2 , 8 8 0 ■A IV 1,6Q0,0Q0h deca-BDEs 47 Table 6. PBDEs pag/kg, dry weight) in wastewater sludge, Roxboro WWTP, NC. SAMPLE IPs 0BW003 5BW001______SLDGBLK01 BDE-3 nd nd nd BDE-7 nd nd nd BDE-15 nd nd nd BDE-17 nd 14 nd BDE-28 nd 78 nd BDE-49 98 95 nd BDE-71 51 nd nd BDE-47 1670 822 nd BDE-66 125 55 nd BDE-77 nd nd nd BDE-100 422 305 nd BDE-119 nd nd nd BDE-99 1660 918 nd BDE-85 255 108 nd BDE-126 nd nd nd BDE-154 295 136 nd BDE-153 488 187 nd BDE-138 nd nd nd BDE-156 nd nd nd BDE-188 nd nd nd BDE-184 nd nd nd BDE-179 nd nd nd BDE-183 310 89 nd BDE-191 nd nd nd BDE-202 nd nd nd BDE-201 nd nd nd BDE-197 993 171 nd BDE-203 1190 220 nd BDE-196 1600 202 nd BDE-208 726 295 nd BDE-207 1340 276 nd BDE-206 27400 1490 nd BDE-209 58800 37400 nd Total PBDEs 97423 42861 nd %REC, PCB-204 73% 62% 79% %TOC 28.2 25.3 0.0 Figure 11. ECNI-SIM comparison of sludge collected in 2005 and 2002 fN Os CN CN OO cn _o I o o o fN CN cn N f fN »—i CN OS CN CN o m o o oo o [Figure 11]. These compounds were later identified by EI spectra comparison using the NIST Mass Spectral Search Program (NIST, 2002) [Figure 12a and 12b] as the flame- retardants HBCD (hexabromocyclododecane) and TBE (1,2-bis (2,4,6-tribromophenoxy) ethane). These have recently been detected in environmental matrixes (Remberger et al., 2004 and Hoh et al., 2005). The HBCD concentration was calculated from the response factor of BDE-153 and estimated to be 2750 pg/kg, for sample (#5BW001). TBE levels were also estimated, assuming a response equivalent to BDE-153, as 243 and 284 pg/kg, for samples #5BW001 and #0BW003, respectively. SEDIMENT RESULTS: Sediments collected in 2002 and 2005 were first analyzed by GC/MS equipped with an on-column injector in the ECNI-SIM mode. Chromatographs for the two sets (2002 and 2005) were similar at the same sample locations [Figure 13]. Major PBDE congeners detected were BDE-209, -206, -99 and -47. A total of 20 congeners were detected in the 2002 samples (14 for 2005), ranging from tri- to deca-BDE. These were also confirmed by EI analysis [Figure 14]. BDE-209 was the predominant congener detected in each sample, ranging from 6.2 to 3,150 mg/kg (%TOC based) [Table 7a, b]. All sediments are reported on a % TOC basis. BDE-47 and -99 ranged from non-detect to 3580 pg/kg and non-detect to 5540 pg/kg, respectively. These two congeners were detected in all samples except that (#0BS018) collected from the tributary (Ghent Creek) that feeds into Storys Creek, approximately 11 kilometers miles down stream from the Roxboro WWTP outfall. BDE-206 was the second most abundant PBDE detected, up to 84,000 pg/kg and concentrations exceeded both BDE-47 and -99 in 15 out of 17 50 Figure 12a. EI spectra of hexabromocyclododecane (HBCD) in sludge 321 100-t 561 237 80- 5WB001. RRI#2022 60- 40- 336 350 NIST MS 3 of 40 (3194-55-6) #ions=294 1.2,5,6,9.10-Hexabromocyclododecane 100-, 239 319 80- 60- 561 20 - m/z200 250 300 350 400 550500 Figure 12b. EI spectra of 1,2-bis (2,4,6-tribromophenoxy) ethane (TBE) in sludge 100-i 357 80- 5WB001. RRI#2129 60- % 40- 328 20 iLi uL NIST MS 1 of 40 (37853-59-1) #ions=176 Benzene, 1,1'-[1,2-ethanediylbis(2,4,6-tribromo- 357 100n 80- 60- 40- 329 688 20- m/z 350 500 550 600 650 Figure 13. ECNI-SIM chromatograph of PBDEs in sediments (#0BS016) and (#5BS003) fN €81 961- o fS _o _o _o fS n ON O n ' n O _o 51 Figure 14. EI chromatograph of PBDE major ions in sediment (#5BS003) 52 o ° w 53 o o cn vo VO m o g ® OO vO ■d - d x ) 0 \ oo J t 3 73 co 73 On oo Td Td ”0 Td Td 7 3 Td vo c a a ss o a on o VO 3 S3 a 33 s3 ss s3 ^ g O O o o i n —i o o Tt T) 73 "O m vo "O -rt Tfr 2 ; tg VO Tf 2737373732;'g73 73^^2^m2o?®^cn cn o C G (N 3 7f 3 m » S3 ^ S3 S3 33 33 ^ S3 S3 S^^S^vO^Mvo *4a CN fOm^cn —i m CN ^ 7j- On cn vo in CN N Tf 0 0 CN vo vo r * o o o 00 o o o o ON CN t> o «n tj- iH o o o o o 0 0 o o o •O *0 ""O "O 'O 73 on vo 73 2? Td cn ’’O ?,0 *"0 Td Td Td Td sd ss ss s3 s3 S3 N- oo S3 C cn -—I cn S3 33 S3 S3 CN o O n s3 a a a a 2 1 6 2 3 5 4 1 7 4 > VO CN N- 4 1 9 e' m O n CN cn 3 0 1 ( en 1-H s * C3 -*■* 3 X! o ON o • o o o c n 73 73 73 73 T3 Td •O Td Td to Td Td Td 7 3 Td Td •73 vo o o CN m on 73 d- 73 73 2 "O O & cn o o O n ^ oo TS S3 S3 sd a t> S3 in S3 S3 N" 33 cn S3 S3S5S3S3S5S3S3S3S3S3 Z 'O Tf in Tfr cn C N- ^ vo a ^ cn OO CN 03 rn cn X O VO O O On ^o o O On O O O VO o' -o ^ *"0 *"0 "C5 *"0 ’’O OO 73 73 7 3 cn r- 7 3 t3 737d737d7d7d7d737d73^gcN m h O H O ^ s3 s3 s3 a a s3 a S3 S3 33 oo oo S3 S3 3 3 3 3 3 3 5 3 5 3 N" cn w S ^ cn h 2 ^ m vo vo Is a a > a oo 2 - 5? * 3 *“0 *0 ’’O *"0 Td Td Td cn ^7dTdTd73 73TdT3T3TdT373Td73TdTd O o m o 3 5 3 5 5 s3 s3 a 00 33 m 2E3S5S3S3S3S3S3S3S5S3S3S3S3S3S3 vo 0 0 CA G N- cn cn u v/^ ”0 in 73 n 73 ’x3 Td Td Td Td Td Td Td *0 Td Td Td Td Td Td Td Td Td N- cn o 5 5 5 3 3 t 3 C cn S3 S3S3s3 s3 s3 s3 s3 s3 S 3 s3 s3 e3 s3 s3 S 3 s3 s3 5 3 cNVO H vo vo N ® o N cv W) o o X Td Td Td T3 T3 73 T3 Td Td T3 Td Td Td 73 T3 T3 T3 Td T3 73 Td T3 Td T3 Td T3 Td Td 73 Td "O 'O o o S3S3S3S3S3S3S3S3S3S3S3S5S5S3S3S3S3S3S3S3S3S3S3S3S3S3S3S3S2S3S3S3 m in g * cn m 6)0 CN CN 3 CA W mMflf'80ONHbVOf> ovONunvoTrmoovoooTtONcnc-i cn vo 00 I" vo On I I rH i-H CN Tf T}- NO t~~- »— on oo cn in m cn m oo oo r- oo on _ On O On O O O O Q Cx3 Cid i l I I I l I I CN n-H CN CN CN CN § offl c q GP c o a - ^^pqpqpqpQpqpqpqpq 0h § o- « 03 1■w * 9 r - O e s H S* x o 03 H Table 7b. PBDEs (jig/kg, %TOC) in sediments from Marlowe/Storys Creek, collected 2005, plus analytical blanks ______SAMPLE IDs______5BS001______5BS002______5BS004______5BS003______5BS005A______5BS005B SEDBLK01 SEDBLK02 I BDE-3 nd nd nd nd nd sd sd "O id a TO "O sd sd ! nd TO I i s Td ca d Tdnd s d sosi'0’OT3'O'd’T3T3’d d d s d T d a c T d d c T d d s T d d c T d d s T d d s T d d a T s d d ’T3t3’a T s d d T s d d s T d d c T d d s T d ’d'd'dT3'd’d d c T d d s T d ’dT)’d',aT3T3'd'd’dT3’dT3T3’0 d c T d d j d T d c d T s d d T s d d T s d d T c d d T c d d T e d d T s d d T c d c d T d c d T d ’a T ’i3 d T d nd T d T i T d T d T d T d O d d nd nd ndfO ddd’d'd'd’d ’d ’doo’dTJoo'd'— ’■O’OTOTd'O’UcnTO'O t d d d d dd d C d d cc c c cc daccccocc CCCCflCnflflHCTf CCCCflCnflflHCTf O d d d TO Td 0 * Td Td C rNrl u b rN u , bN w r E N u r N H N N N H N r W H N N H w r N w N N rH i H H H H N H H r H H N H H H SSSSS i-< r' v ' - n nfoorOOON®®ON®ON®®®o ® ® ® N O ® N O ® ® N O O cn»f>oooor'O w onw O N O i-H r'® rv© rr''T T -i»-i q n c CN o o o sd m r-H -O O T d T QQWH O TO TO d d ■—1 Td Td Td C d sd sd a h Sd a o\ o q *—1 m T d — ' n - T d T d tj O N O i f i i C T f n i a o i o w ^ O N n i H d d d d o d sd r- ci 1 m m ^’d ^ "d CN cn o ONOONiri»oninxxC'aoo\ CN o S S O qq m 1 s id sd c TO TO in O TO TO d d n c ed td c 3 a TO TO q c q c q p s q c c q c q o o c q c q WWWWWWWWWWWWWWWWWWH d SdSd 1 0 * O C t"- t"- d d d d d d d d d d d d d d d d 1 d n ON O C 0 0 1 O " " 0 ’ ""O"O ""O ’’O "O ’’O *0 d T ’’O "O *0 2 r 2 2 d T d T 0 T 0 T d T d T d T d T d T " *0 " *0 *0 *0 TO *"0 *”0 *”0 ""O *"0 '"O 0 * 0 * 0 * 0 * 0 * sdsdsdsdsdcdsdsdsdsdsdsdsdsdsd aaaaaccicicicitia c c c a c s c a a c c s s d }£ j} £ £ d S d S d S d S d S d S d C d S d S a c a c i t a a d a a a a 1 fO 1 "O "O "O TO *0 Td "O *0 "O *0 *0 1 "O *0 1 *0 "O d n 1 d n 1 *"0 1 d n 1 d T 1 d T O © O O n O « O n n O h 1 *0 d T 1 *0 *"0 O O O 1 *0 6 0 fO m n c 0 1 " - n C" >n -rf 1 *0 n O « h m 1 V) n ” -H cn o ^ o S § o 0 N C 0 n c T— m r*“H t-Hr-" O O ncn cn m m O O -H n- O O n-O o o 1 1 O 0 0 n c n c n c O 0 0 0 N C Tj" - f CN cn 0 0 0 0 CN O 0 0 0 0 N CN CN o o 1 n ^ n O fiU ® O V H t3H i? 03 g g m t T ® 0 0 N C t-- N C U CN cn Tt 0 0 t" CN CN Tf Tf t" co 00 O tt in r- gg g o « * <-> £3 05 3 n 3 O N N C CN ON O °o —I ON o'' m I ft, r- r i 54 samples. Total PBDEs for the 2002 samples ranged from 1710 mg/kg, at the outfall (#0BS014) to 3250 mg/kg (#0BS026), approximately 1.3-km down stream. For the 2005 sample set, total PBDEs ranged from 195 mg/kg (#5BS002) at the outfall and maximum concentration of 2450 mg/kg (#5BS003), approximately 5.6 km down stream. TBE was also detected in each sample set. The highest concentrations were located 5.6 km miles down stream of outfall, 1020 and 897 pg/kg in 2002 and 2005, respectively. HBCD was only detected in the 2005 sample set [Figure 13]. The maximum concentration (7600 pg/kg) was detected at the outfall. BIOTA RESULTS: Two species of fish (sunfish ( Lepomis gibbosus, n=13) and creek chub (Semolilus atromaculatus, n=6)) and a crustacean (crayfish (Cambarus puncticambarus sp.c, n=5)) were collected at the outfall in 2002. In 2005, only a group of sunfish (n=22) was available at the outfall collection site. Biota extracts were introduced to the GC by split/splitless and PBDEs detected by ECNI-SIM. The chromatographic approach, monitoring bromine ions ( m/z 79 and 81), produced sharp peaks with good separation for the tri- through deca-PBDEs [Figure 15]. A total of 23 PBDEs were detected by ECNI- SIM and confirmed by EI analysis [Figure 16a and b]. The most abundant congener in the biota samples was BDE-209, 21,600 pg/kg, (l.w.) in the crayfish sample #0BF075 [Table 8]. However, -209 was not detected in the chub (#0BF073) or 2005 sunfish (#5BF001) composites. All biota results are reported based on lipid weight. BDE-47, the second most dominant congener reported, was detected in all the samples and ranged from 4,110 to 17,200 pg/kg. Also detected in each sample were BDE-153, four octa- Figure 15. ECNI-SIM chromatograph of PBDEs in biota (#0BF072) o o 881 o o o oo oe o 80Z 90Z o o o 001 o n o o C\J esi o oc ro _ Lf) o _ CD C\J 56 57 t-> o Lfi to ffl o os o • p* .o tg CA c o u 0 E 00 w Q M Oh1 03 X o Jah uOS ox o ■HJ OS _o rNcm os Q Sh IL X m u o HH w 03 o o o o acz O O CD a o a a a o o vo a a o C ° CD o o 'T o o o O V o o o o o § g XI C . CD TT 4> o' o' o ' | - o' D c o rr o' o' o' o o' o' O 15 1h o o o CD T c o o o o ^ CO CM os a a o o' V CM <£> T CM o ' OXox I CM co V ZD CD fe UJ ^r LD Figure 16b. EI chromatograph of hepta- through deca-PBDEs major ions in biota (#0BF072) El_082406_1 6 | 0BF072 O O O O O O O O CM O O O CM M CM CM ] O O O O O O O O O O O O JL = CM csj 59 Table 8. PBDEs (jig/kg, % lipid) in biota collected at Roxboro WWTP outfall and analytical blank SAMPLE IDs 0BF073 0BF075 0BF072 5BF001 BIOBLK01 BDE-3 nd nd nd nd nd BDE-7 nd nd nd nd nd BDE-15 nd nd nd nd nd BDE-17 nd nd 32 29 nd BDE-28 285 nd 246 179 nd BDE-49 618 nd 1300 855 nd BDE-71 nd nd nd 541 nd BDE-47 17200 4110 11600 7600 nd BDE-66 nd nd 1670 952 nd BDE-77 nd nd nd nd nd BDE-100 3460 nd 2340 1820 nd BDE-119 nd nd nd nd nd BDE-99 nd 3560 13300 522 nd BDE-85 725 nd 496 229 nd BDE-126 nd nd nd nd nd BDE-154 2610 nd 2290 1880 nd BDE-153 918 767 3110 2420 nd BDE-138 nd nd nd nd nd BDE-156 nd nd nd nd nd BDE-188 1450 nd 884 289 nd BDE-184 nd nd 360 164 nd BDE-179 1140 nd 166 137 nd BDE-183 nd nd 83 77 nd BDE-191 nd nd nd nd nd BDE-202 895 87 747 243 nd BDE-201 129 78 773 335 nd BDE-197 nd 43 193 86 nd BDE-203 117 132 74 20 nd BDE-196 45 200 65 28 nd BDE-208 103 143 201 67 nd BDE-207 79 1920 276 73 nd BDE-206 94 2650 411 133 nd BDE-209 nd 21600 2880 nd nd Total PBDEs 29868 35290 43497 18679 nd %Rec. PCB-204 60% 81% 90% 72% 88% type chub crayfish sunfish sunfish NaS04 BDEs (BDE-202, -201, -203 and -196) and three nona-BDEs (BDE-208, -207 and -206) [Table 8]. Three additional hepta-BDEs (BDE-188, -184 and -179) were also observed in the biota samples. These, along with BDE-202 and -201, were confirmed by ECNI “full-scan” and were not detected in either the sludge, or sediment sample sets. Overall the two sunfish composites from 2002 (#0BF072) and 2005 (#5BF001) produced similar chromatograms. However, the total PBDE level for the 2002 composite was approximately twice that of the 2005 composite, 43,500 and 18,700 pg/kg, respectively [Table 8]. HBCD and TBE were not detected in either biota samples. QUALITY CONTROL / QUALITY ASSURANCE: Quality control measures employed included the analysis of surrogates and analytical batch blanks. Problems during sample preparation and analysis, which can alter the results, may arise from matrix effects, equipment or analyst error. These problems may be assessed by the addition of “surrogate” compounds to samples prior to extraction. The surrogate, having similar chemical properties to the compounds of interest, exhibits similar analytical behavior. The success of the analytical process can be tracked based on the surrogate recovery. Surrogate (PCB-204) was introduced to each sample and blank prior to their extraction. Samples were separated into four analytical batches (one sludge, two sediment and one biota). One blank was processed with each batch. Surrogate recoveries ranged from 62 to 73% for the sludges [Table 6], 68 to 106% for the two sediment batches and 60 to 90% for the biota samples [Table 7a, 7b and 8]. All results were corrected for surrogate recoveries. Surrogate recoveries for the blanks ranged from 88 to 98%. Blanks were also checked for potential cross sample and laboratory contamination by monitoring for all target analytes by GC-ECNI-SIM using m/z 79 ([79Br]') and 81 ([slBr]'). PBDEs were not observed above the detection limits in any of the blanks. Results for each blank (SLDGBLK01, sludge blank, SEDBLK01, -02 sediment blanks and BIOBLKOl biota blank) are located with their respective sample sets [Tables 6, 7a,b and 8]. The extraction process and the reproducibility of the analytical procedure were also tracked by the analysis of PBDE-fortified matrices (NaSC> 4, sediment and fish tissue). The spiking solution contained 27 PBDE analytes ranging from mono to deca- BDEs [Table 9]. Fortified (1 mL of the spiking solution [Table 5]) and unfortified sodium sulfate (NaS 0 4 ) aliquots (10 gms each) were analyzed in triplicate. There were no observable PBDEs within the unfortified NaSC >4 analysis. Good mean surrogate recovery 92% (13 % STD (percent standard deviation)) was observed for the three fortified NaSC >4 samples. Spiked PBDEs ranged from 10 to 50 ng/g, d.w. and mean recoveries ranged from 61 to 122% (overall average 92%). The mono and di-BDEs (BDE-3, -7 and -15) had the lowest recoveries < 65%, perhaps due to their higher volatilities. Fortified and unfortified sediments (10 gms each) were also analyzed in triplicate. Surrogate recoveries were >70%. Mean spiked recovery was 65% (range 45- 85%). Again recoveries were better (>60%) for the higher (>5) brominated PBDEs. The %STD was generally higher (+10%) for each of the PBDEs in the fortified sediment samples than the NaSC >4 samples, means 24 and 13%, respectively. The unfortified sediment samples did contain some low levels of PBDEs (BDE-47 (0.8 ng/g, dry wt.), - 99 (1.8 ng/g) and -85, -100, -153, -154 (< 0.2 ng/g)), and 27.9 ng/g o f-209. These were accounted for when calculating final recoveries. Dried fish tissue was used for the fortified fish matrix experiment. This matrix has been previously analyzed and contained no detectable PBDEs. Fortified (10 to 50 ng/g, d.w.) and unfortified aliquots (10 gms, d.w.) were extracted. Mean surrogate recovery was 72% and individual congener recoveries ranged from 73 to 147%. BDE-3 had the lowest recovery and BDE-206 and -196 the highest. The recoveries for the other 24 PBDEs ranged from 81 to 129%. Replicate analysis was also completed on an unfortified sediment sample, (#5BS005a and b). Two aliquots were assigned to different analytical batches and extracted separately. Surrogate recoveries were similar, 68% and 69%. The %STD was < 28% for the nine PBDEs detected in the sample [Table 9]. Although, there was some variability (%STD > 30%, primarily for the lower brominated PBDEs (mono- and di-PBDEs)) and high recoveries for BDE-196 and -206 in the spiked fish matrix, this method generally preformed well for the majority of the PBDEs. Results were well within QC accepted criteria for ongoing precision and recovery (OPR) (50-150% OPR) outlined in USEPA Draft Method 1614. (Brominated diphenyl ethers in water, soil, sediment, and tissue by HRGC/HRMS), for each target analyte and matrix tested. Q Os F™H r—* os i n o m n- Os 63 H Os C/5 t - c^ ° ) Os i-H CN OO CN CN 1 >_H 1 i ? "w M o a V) r- in o o a o ^3 "d 'd T3 T3 'd T3 t3 Xl CN T! X3 X3 "O X3 X3 X3 T3 T3 OO Os o o Os v o a a a c a a so a C3 On C3 a a a a a a a a •—i oo o ^ SO +-> i/5 xr tj- ei 02 CN ju in 73 a sq NO t ^ oo oo in cn in ^t in so t^; a i n oo o i n o CN xr n x vo i n i n o CN Xj- oo Os oo f2 Os in CN © CN i n cn OO /**s i n CN r*H -M VmP Jh © p oo OO sq -H cn o sq C"; in in oo rf o Os xf p p p p p °° p p p p in o o r-H CN xr Xt" oo cn i—i o 1 © T}- so CN oo cn n - os in in oo in Os in so m r- © m SO m in so so in m so in m in SO SO so SO r~- r- r- oo t"- so oo >* 73 © Crp-C -4-1 Q p p p oo n- CN sq CN Os cn cn Os in in p p p cn p p p p 31.2 In H cn °5 °5 P P sd NO CN oo sd cn CN cn CN cn 00 CN n- H ^H n- CN CN sd CN r-H cn in CN 1 cn .© C/5 oo CN CN CN CN CN CN CN CN CN CN CN CN CN CN CN CN CN CN CN cn cn CN CN 8.3 w Q _ tt 0J5 Jcf in CN 00 cn in OO o in SO in oo p O Os p p p p O p p p r-H p o Plh > bn sd in in cn CN cn Xt- xf’ id in in o CN /—s Tt1 in sd in in so sd id in in cn m n- CL» < C © W3 O xf cn cn in id p oop p p P p oop CN sf so SO oo oo x tv- CN CN CN in cn oo Os Os cn ooHOh'tOsod??^ Os Os 73 SO so so oo r- o OO oo oo 2 2 ooo \ ososos OsO\~5:osoo CN SS C3 cn ■+•< C © Q so p m in in Tt H p p p p p p P s CNCN o o O tj- o cn mi C/5 cn cn 1—1 so sd rt Os C^ oo t— © TfCs o os cn oo cn '—' iCNn ifr \i sosu I—I -—1 CNv n h-xcN'nosOssosoxos^n i - -si vn *« i w' wi CZ5 w> -$®P p p P p CN >. W) oososcncnossdt^oor^ododosodosos^HOin „ n - Os A *< fl so so sd oo > MOOOOOSOOOOOlHrHfHHHHHHHCNCNTfSflO & • PMX sm Q -4-1 ci *a .^^«-»«v^i^.vnr^oos-s ,^soTrcnoosorrcni-ii>sor^soos v-1 c3 ■ a os is s H w W) s £ P< u o © o WWpQPQPQCQPQPQeQPQpQp3CQCQCQPQPQPQpqpQPQPQPQPQPQPQpq3§9999§§§S«® 9§® Q®®®®QQQQO®9 ^ Ph b 3 U 3 cs C/5 H DISCUSSION WASTEWATER SLUDGE: Tri- through deca-PBDEs were detected in Roxboro WWTP sludge (collected in 2002 (#0BW003) and 2005 (#5BW001)), indicating potential environmental exposure through PBDE-laden effluent particulates. It has been estimated that a kilogram per year of PBDEs are introduced into the San Francisco Estuary through WWTP effluent discharge (North, 2004). For the major congeners of the penta- and octa-formulation, the tri- through hexa-PBDEs, a 46% lower concentration was observed in the 2005 compared to the 2002 samples [Figure 17a]. This reduction may relate to the December 2004 discontinuation in the manufacture of these formulations in the U.S. (sample #5BW001 was collected in November, 2005). For this WWTP substantial inflow of deca- has been hypothesized from a major manufacturer of plastic goods. It is possible that penta- and octa- were also utilized. It has also been hypothesized that the PBDEs in wastewaters may be derived from releases from finished products in use rather than manufacturing. For example, it has been reported that indoor dust can contain substantial (mg/kg) burdens of PBDEs (Hale et al., 2006), which can eventually enter household waste streams (e.g. washing clothes). Also, as products containing PBDEs age their release has been reported to increase. For example, in an indoor air study replacement of an older •5 personal computer yielded a nearly 50% reduction in office air levels (431 pg m' to 253 pg m'3, sum of BDE-47 and -99) (Hazrati et al., 2006). Environmental releases of PBDEs and subsequent sludge deposit may continue until household and office products 64 65 figure 17a. Roxboro WWTP sludge (2002 and 2005) tri- through hexa-PBDEs 1800 1600 1400 ■O) £ 1200 o % 1000 £ □ 2002(#0B\MX)3) 800 ■ 2005 (#5B\MX)1) O) O) 600 IL 400 200 u .EMJZL “i r Figure 17b. Roxboro WWTP sludge (2002 and 2005) hepta- through deca-PBDEs 60,000 50,000 □ 2002 (#0BW 003) -C 40,000 U) ■ 2005 (#5BW 001) i—i— i i— •..■...... r - J = l ------1 (e.g. electronic devices, polyurethane foam furniture) containing these formulations are replaced with PBDE-ffee products. According to the Toxics Release Inventory (TRI) 2287 kg in 2002 and 1692 kg in 2004 (TRI last reported year) of -209 were transferred to the Roxboro WWTP. Prior to 2002, the TRI suggested that Roxboro received over 34,000 kg of -209 per year. Maximum transfer, 113,826 kg, was in 1994 [Figure 18]. However, since 1999 transfers decreased 10-fold and by 26% between 2002 and 2004. This reduction may have contributed to the pattern in -209 sludge values, which decreased 41% between 2002 and 2005 [Figure 17b]. Also noted in the samples was a decline of the three nona- (BDE-206, -207 and -208) and three octa-BDEs (BDE-196, -203 and -197) [Figure 17b]. It has been previously reported that -209 can debrominate under anaerobic conditions. Gerecke et al. (2005) examined sewage sludge collected from an anaerobic digester. A 15% reduction of -209 was observed after incubation for 238 days. About 5% of the initial - 209 was reduced to two nona-PBDEs (BDE-207, -208) and several octa-PBDEs (three tentatively identified as BDE-196, -198/203 and -197). They also conducted similar experiments on BDE-206 and -207 and observed debromination, as evidenced by the formation of octa-PBDEs. However, no other debromination products were reported. It also has been reported that deca- technical products can contain nona- through octa- PBDEs (La Guardia et al., 2006). Saytex-102E (Albemarle Corp., Louisiana, USA) contained BDE-206, -207 and -208 totaling 3% (w/w). An older European deca-product (Bromkal 82-ODE, Chemische Fabrik Kalk, Koin, Germany) contained three nona- 67 Figure 18. BDE-209 transfers to the Roxboro WWTP (USEPA, TRI. 2006) 120,000 100,000 80,000 60,000 Ui 40,000 20,000 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Years 68 PBDEs (BDE-206, -207 and -208) and octa-PBDEs (BDE-196, -203 and -197) totaling 10% (w/w). PBDEs have been previously detected in other WWTP sludges. Hale et al. (2001b) reported a maximum -209 concentration of 4890 pg/kg, mean 1010 pg/kg (d.w.), in 11 sludges (biosolids) collected from four different regions of the U.S. They also reported five additional PBDEs (BDE-47, -100, -99, -153, and -154) ranging from 1100 to 2290 pg/kg (d.w.). These congeners contributed an average 75% of the total PBDE sludge concentration, with the remainder predominantly from BDE-209. In one sludge -209 contributed 70% of the total PBDEs detected. For the Roxboro sludges, BDE-209 was the major PBDE congener detected (37,400 and 58,800 pg/kg (d.w.), 2002 and 2005, respectively), constituting 60 and 87% of the total, respectively [Figure 17a and 17b]. Also, the -209 concentrations exceeded the previous survey’s maximum reported -209 values by 10-fold, likely due to high -209 usage by local plastics manufacturers. However, the sum of BDE-47, -100, -99, -153 and -154 for Roxboro’s 2005 sample (2,370 pg/kg, #5BW001) was similar to this survey’s maximum value and the 2002 sample (4,540 pg/kg) was only twice this concentration. The profiles of these congeners for Roxboro sludges were similar to those reported by Hale et al. (2001), approximating that seen in the intact penta- technical formulation [Figure 19]. This suggests that -209 debromination might not be a major contributor to the lower brominated PBDEs seen in Roxboro sludges. 69 Figure 19. Penta-formulation (DE-71) compared to Roxboro WWTP sludge (2002 and 2005) and U.S. sludges (n=ll) 100% 90% 80% 70% □ BDE-153 60% □ BDE-154 0 BDE-99 50% ■ BDE-100 40% □ BDE-47 30% 20% 10% 2 0 0 2 2 0 0 5 U.S. Sludge* DE-71** (#0BW003) (#5BW001) (n=11) *Hale, et al. (2001b), **La G uardia et al., (2006) SEDIMENTS: Sediments were collected in 2002 and 2005. These were from the Roxboro WWTP outfall, several kilometers downstream from the outfall, and from tributaries, which feed into Roxboro, effluent receiving (Marlowe and Storys Creek). PBDE levels therein suggest that this WWTP was the major contributor of PBDEs to the Marlowe/Storys Creek system. Although, PBDEs totaling 6.60 and 23.5 mg/kg (% TOC), were detected in two tributaries (Storys Creek (#0BS019) and Ghent Creek (#0BS018), respectively) which feed into Marlowe/Storys Creek downstream form the WWTP, these values were only 2% those at the WWTP outfall, i.e., 1710 mg/kg (#0BS014). Total PBDEs levels were greatest several kilometers downstream from the outfall and were still detectable at the furthest collection point (10.8 km downstream) from the outfall, where Marlowe/Storys Creek meets the Hyco River [Figure 20]. Concentrations were lower upstream from the outfall (0.1 and 0.2 km), , 38.0 and 37.5 mg/kg, 2002 (#0BS013) and 2005 (#5BS001), respectively [Table 7a and 7b]. These values may indicate PBDE releases from the City of Roxboro, NC. via urban runoff, or other sources. Previously, high PBDE burdens (tetra- through hexa-PBDEs, totaling 47,900 pg/kg (l.w.) were reported in carp (Cyprinus capio) collected from the Hyco River. This site was just upstream with the Hyco’s confluence with the Dan River, an additional 30 km downstream from where Marlowe/Storys Creek enters. This report exceeded all previously reported levels in edible fish tissue (Hale et al., 2001a). Examination of the watershed, in addition to supplemental efforts using passive samplers (unpublished data) point to the Roxboro WWTP as the likely source. Maximum sediment concentrations in 2002 (#0BS026) and 2005 (#5BS003) samples were detected Figure 20. Total PBDEs in Marlowe/Story Creek surficial sediments, collected from Roxboro WWTP to 10.8 km downstream co in o co o m m cm M CM o o CM o o (OOJ.%) MCM CM o o in |/6 5 6 i u saaad leiox saaad m o o o o in in in in in in o> in in © co between 1 and 6 km from the outfall (3250 mg/kg and 2450 mg/kg, respectively). Further down stream (10.8 km) levels dropped (322 and 247 mg/kg (2002 (#0BS020) and 2005 (#5BS005), respectively), but still exceeded background levels observed upstream of the outfall by 10-fold [Figure 20]. There appeared to be a 3-fold greater concentration in 2005 (5.6 km) sediment (#5BS003) compared to the 2002 sampling (#0BS016) [Figure 20]. This may indicate variations in WWTP discharge or relocation of historic PBDE reservoirs in the river system. This may also be reflected in the HBCD sediment values. The highest HBCD concentration (7600 pg/kg, % TOC) was reported at the outfall. It was only observed in the 2005 samples, perhaps indicating fairly recent release to the river system. BDE-209 was the major congener detected in each of the sediments, contributing >89% of total PBDEs, followed by the nona-PBDEs which constituted 3 to 10%. Most of the sediments also contained tri- through octa-PBDEs. However their total contributions were an order of magnitude lower than -209 [Figure 21a and 21b]. BDE- 209 has been reported to undergo photolytic (Watanabe and Tatsukawa, 1987, Eriksson et al., 2004 and Soderstrom et al., 2004), and microbial debromination (He et al., 2006) under laboratory conditions. This is believed to increase its toxicological potential. Covaci et al. (2005) previously detected up to 8400 pg/kg (d.w.) of -209 in sediment cores from the Scheldt River (Belgium), which drains a very densely populated and highly industrialized area of northern France. BDE-209 concentrations from Marlowe/Story Creek were similar to the Scheldt River and ranged from 109 to 2205 pg/kg, on a dry weight busies). Also detected in these cores were tri- through hexa- 73 Figure 21a. Marlowe/Storys Creek surficial sediment congener profiles (tri- through octa-PBDEs), 2002 and 2005 16000 octa- 14000 ■ hepta 12000 hexa- w penta- ^ )000 tetra- N® 0s *000 '§D >000 -0.1 0.0 0.5 0.8 1.3 5.0 5.3 5.6 10.8 -0.2 0.0 1.3 5.6 10.8 kilometers (km) from outfall (collected 2002) km from outfall (2005) Figure 21b. Marlowe/Storys Creek surficial sediment congener profiles (nona- through deca-PBDEs), 2002-2005 3.500.000 3.000.000 2.500.000 □ deca- ■ nona- U 2.000.000 O H oN 1.500.000 WD 'Si a. 1,000,000 500,000 I] n n -0.1 0.0 0.5 0.8 1.3 5.0 5.3 5.6 10.8 -0.2 0.0 1.3 5.6 10.8 kilometers (km) from outfall (collected 2002) km from outfall (2005) PBDEs (sum of 1.4 to 270 pg/kg) and, like -209, they decreased in concentration with depth. This indicates minimal degradation of -209 to lower brominated PBDEs in these sediments. Congener profiles of the tetra- through hexa-PBDEs were also observed to be similar to the penta- technical mixture, indicating likely derivation from products containing this flame-retardant [Figure 22]. Mai et al. (2005) also reported that BDE-47 and -99 constituted greater than 50% of the nine tri- to hepta- PBDEs detected in sediments from the Zhujiang and Dongjiang Rivers, which flow through two urbanized regions (Guangzhou and Dongguan) of China. These sediments also contained -209, ranging from 21.3 to 7340 pg/kg (d.w.). Although, the congener profiles for the lower brominated PBDEs were similar to the commercial penta- formulation, debromination of -209 could not be ruled out as octa- and nona- PBDEs were also detected in the sediments at levels exceeding those expected in unaltered deca-technical mixture. Sediments collected from Marlowe/Storys Creek contained similar concentrations of tri- through octa-BDEs in the 2002 and 2005 sample sets. An exception was the most downstream sample site (10.8 km) in 2005, where a decrease of 92% was detected [Figure 21a]. Maximal concentrations for these congeners in 2002 and 2005 in sediments 5.6 km downstream of the outfall; 14,200 (#0BS016) and 13,700 (#5BS003) pg/kg (%TOC), respectively. Concentrations of tri- through octa-BDEs also increased with distance from the outfall [Figure 21a]. In contrast, the amount of nona- and deca-BDEs decreased [Figure 21b], indicating possible -209 debromination as a function of time and environmental exposure. However, this was not observed in the 2005 samples set. At mid-stream (5.6 km, #5BS003) -209 levels increased. However, the tri- through octa- PBDEs concentrations were similar to 2002 levels. Further downstream (10.8 km, #5BS005) values decreased for tri- through octa-BDEs, while BDE-209 remained the same compared to 2002 levels [Figure 21a and 21b]. Regardless, tri- through octa- PBDEs only accounted for 3.3% or less of total PBDEs. Others (e.g., Covaci et al. (2005); Mai et al., (2005)) have also reported similar PBDE contributions and attributed them to products containing the penta- formulation. Congener profiles for the tri - through hepta-PBDEs were comparable for Marlowe/Storys sediments collected in 2002 and 2005, at the outfall and 1.3, 5.6 and 10.8 km downstream. BDE-47 and -99 were the major PBDEs detected and profiles generally resembled the penta- technical formulation [Figure 22]. These same congener patterns have also been observed by others (Covaci et al. (2005); Mai et al., (2005)) in -209 rich sediments. Congener profiles were also similar to sludge collected from the Roxboro WWTP (the probable source of PBDEs to Marlowe/Storys Creek) [Figure 22], indicating that even after 13 years of -209 residence (according to the TRI) minimal -209 debromination has occurred in surficial sediments along the 10.8 km creek system. BIOTA: Biota collected near the Roxboro WWTP outfall contained 24 PBDEs, ranging from tri- to deca- PBDEs [Figure 23 a and 23b]. Dominant congeners detected were BDE-47 and -99. PBDEs have been previously reported in 89% of edible fish tissue collected from two large Virginia watersheds (Roanoke and Dan Rivers) (Hale et al., 2001a). This watershed is located downstream from Marlowe/Storys Creek. In the Virginia samples (332 fish fillets), BDE-47 was the most abundant congener detected. Figure 22. Terta- through hepta-PBDEs homologues of surficial sediments and sludge (2002 and 2005) compared to DE-71 o © uojinqujsip a uojinqujsip 00 o 0s (D © sP © 6 |uo % o|Ouioq CM o O (800M90#) Z'l 8 9 201 oMs) ® UooMas#) 9'S 801 'i Z 300? 800? 0) _3 ? ? 1 ■a € (0 £ UJ 5 O +5 Q o £ S2 = c ^ o O) 0 o E ^ 0) > 0 a> o c 3 0 o E ^ 0 o E*"S o E <2 = *■* 5 o 3 o CM o o in CD O CD o 76 77 Figure 23a. PBDEs in biota (tri- through hexa-PBDEs) 20,000 1 8 , 0 0 0 □ CRAYFISH 1 6 , 0 0 0 E 2 C H U B ■o 1 4 , 0 0 0 □ Sunfish (2002) 'EL 12,000 ■ Sunfish (2005) 10,000 o> *3) 8,000 6,000 4 , 0 0 0 2,000 0 # ^ & Figure 23b. PBDEs in biota (hepta- through deca-PBDEs) 3 5 0 0 21,600---- ► 3 0 0 0 1 CRAYFIS H 2 CHUB 2 5 0 0 ~o □ Sunfish (2002) .9- 2000 .. ■ ourmsn d> 1 5 0 0 i! 1000 1 5 0 0 i I i I Pi 1 nfj I m !~L [T5H. ru_ ffl \Ji ] 0 k JV The sum of BDE-47, -99, -100, -153 and -154 ranged from < 5 to 49,900 pg/kg (lipid basis). Fish collected in the current study near the Roxboro outfall ranged 14,200 to 32,600 pg/kg (l.w.) for the same five PBDE congeners. BDE-47 and -99 were the major congeners detected. Although these did not exceed the previous Virginia record level, concentrations surpassed over 90% of those previously reported in Virginia, and are among the highest reported in fish in the world. BDE-209 was also detected in two of the fish from Roxboro and was only exceeded by the —47, -99 and -153 concentrations [Figure 23a and 23b]. This indicates that -209 was bioavailable and metabolic debromination could occur in these samples. BDE-209 was also detected in the crayfish composite (21,600 pg/kg, l.w.). Interestingly, it was an order of magnitude higher than the sunfish sample (2880 pg/kg, l.w.) [Figure 23b]. A literature search determined that this was the first report of PBDEs in crayfish, however PCBs (polychlorinated biphenyls) have been previously reported from the river Meuse (Netherlands) in crayfish (Orconectus limosus). PCB profiles were reported to follow those in fish, favoring tri- through hexa-PCBs (Schilderman et al., 1999). However, Holmqvist et al., (2006) compared PCBs in crayfish (Pacifastacus leniuscukus) from Swedish streams and lakes and found a higher variability in total PCBs in stream crayfish. This was attributed to crayfish being omnivorous and hence influenced to a greater degree by diverse contaminant sources within their catchment. BDE-209 accounted for 95% of the PBDEs in sediments where the Roxboro crayfish were collected and may contribute to the high concentrations, constituting 59% of total PBDEs. Hence invertebrates may be an important route of -209 exposure to aquatic and terrestrial predators. BDE-47 has been previously reported as the most abundant congener in fish ranging from 45% of total PBDEs in channel catfish to 74% in carp [Hale et al., 2001a]. However, congener profiles for tetra- through hexa-BDEs detected in the Roxboro sunfish samples were more comparable to the penta- technical mixture, i.e. similar BDE-47 and -99 contributions., This may be due to factors such as the organism’s food source, age or trophic level. A distribution favoring the hexa- PBDEs (43%, BDE-153, -154, 20% -47 and 25% -99), was previously reported in sunfish ( Lepomis macrochirus) from Hadley Lake, Indiana, U.S. This unusual PBDE distribution may have been related to the presence of a nearby PBDE research facility (Dodder, et al., 2002). In the VA chub sample, BDE-99 was not detected and -154 concentrations were elevated compared to -153. This same congener profile was also seen in common carp ( Cyprinus carpio) collected from the Dan and Roanoke Rivers (Hale et al., 2001a). Stapleton et al. (2004) exposed common carp to BDE-99 and -183 via the diet and reported significant conversion of BDE-99 to -47 and -183 to -154. Neither congener (BDE-99 or -183) were detected in the Roxboro chub, but were observed in surrounding sediments and in other fish species at this site. One explanation could be that both creek chub (Semotilus atromaculatus) and common carp {Cyprinus carpio) metabolize PBDEs similarly, both belong to the same family {Cyprinidae). Hence comparative metabolic capacities may contribute to the different PBDE congener profiles observed between the various fish species, as well as the crayfish. Although -209 was only detected in the 2002 sunfish (#0BF072), both sunfish composites did contain three nona-PBDEs. They also contained five octa-PBDEs and four hepta-PBDEs. The chub composite contained three nona-, four octa- and two hepta- PBDEs. Of these, two octa- (BDE-201, -202) and three hepta- (BDE-188, -184 and - 179) congeners were not detected in either the sludge or sediment samples. Therefore these conceivably could be metabolic debromination products of the higher brominated PBDEs (e.g. BDE-209). In a follow-up to the carp exposure study, Stapleton et al. (2006) conducted a BDE-209 dietary exposure experiment on juvenile common carp and rainbow trout (Oncorhynchus mykiss). After 60 days of exposure (112 days for the trout), carp whole body homogenates were extracted and analyzed for PBDEs. BDE-209 was not detected in the carp sample. However, one octa- (BDE-202), two hepta- (BDE-179 and 188) and three hexa-PBDEs (BDE-154, -155 and one un-identified hexa-BDE) were detected, indicating that carp can metabolize -209 to lower brominated diphenyl ethers. As noted previously, common carp may metabolize PBDEs in a manner similar to creek chub. Chromatograms of the lab-exposed carp and the chub from Marlowe/Storys Creek exhibit similar the hepta- through deca- congener patterns [Figure 24]. PBDEs present were identified as one octa- (BDE-202) and two hepta- (BDE-188 and -179). These have not been previously detected in the sediments or sludge. Two nona-BDEs (BDE-208, - 207) were also detected in the chub sample. In the Stapleton et al. rainbow trout dietary -209 exposure study, -209 was detected, as well as several hepta- (BDE-188, -184, 179 and -183), octa- (BDE-202, - 201, -204/197, -203 and -196) and nona- PBDEs (BDE-208, -207 and -206) . Uptake of -209 from food was estimated at only 3.2% (Stapleton et al., 2006). Although rainbow trout are from the Family Salmonidae and sunfish are Centrachidae , the same PBDEs (hepta- through deca-PBDEs) were detected [Figure 25]. Similar nona- through hepta- Figure 24. Chromatograph of PBDEs in creek chub compared to common carp BDE-209 exposure study oc oc vo 5 > ■— ci 13 S-9 c CO D C d c CD CD CD ° hj o m ' to GO CD CD CD CO CD D 155 BDE Hexa-3. D 154 BDE ’ST CD CD CD CD O' BDE 202 D 188 BDE DC CD CD CNJ CD CD CD CD CD CD o CO CD CD' CD CD o CD CD CO CD CD' CD ^r CD d c CD 10Z ' CD CD C\J CD d c CD Z,6l l O L 90Z ' r^- CO C\J I O I_C co = i _ C\J C\J C\J C j o C\J co CD n J 81 Figure 25. Chromatograph of PBDEs in sunfish compared to rainbow trout BDE-209 exposure study 00 00 BDE 202 BDE 00 CM o o CO BDE 208 BDE 209 BDE BDE 188 BDE 639 ----- Z6L309 £03 309 —= 309 £03 961 D 154 BDE BDE206 - 9 0 3 CD CD 00 CO CD CD NO ON CD CSJ CNJ CD 82 PBDEs were also detected for the 2005 sunfish sample, but -209 was not observed. Although it is likely exposure conditions varied between the control -209 exposure study and the Roxboro outfall, congener profiles for the trout and sunfish (carp and chub) were similar, suggesting similar metabolic pathways. Stapleton et al. (2006) also confirmed - 209 metabolic debromination using a preparation of both trout and carp liver microsomes. After incubation with -209 PBDE debromination products and profiles were similar to those seen in the whole fish lab exposure study (Stapleton et al., 2006). This substantiated a metabolic pathway for BDE-209 debromination. This also supports the current study’s conclusion that -209 is bioavailable and can undergo metabolic debromination in the field, generating lower brominated PBDEs. Ultimately this may lead to increased PBDEs burdens in organisms living in aquatic environments and terrestrial animals, which rely on aquatic environment as a food source. CONCLUSION PBDEs with four to six bromines have been recognized to be of human and ecological health concern. They are persistent, bioaccumulate and have been shown to disrupt the endocrine system. These concerns have led to discontinuations in the use of the two flame-retardant products (penta- and octa-) containing these PBDEs in Europe and the U.S. However, the widely used and unregulated deca- product may continue to add to environmentally relevant PBDE burdens due to its potential to debrominate, as previously reported under laboratory conditions. Environmental analysis of BDE-209 has also been hindered by limitations of commonly employed laboratory practices and equipment. Using recent technical improvements (e.g. on-column injectors, extended range magnetic mass spectrometers, chemical ionization) these analytical problems were overcome in this study, producing a method capable of analyzing mono- through deca-PBDEs. This method was also shown to be useful in identifying speculative environmental debromination products (e.g. BDE- 201 and -202) and is a useful screening tool for other brominated flame-retardants (e.g. HBCD and TBE). Unfortuantly, results from controlled laboratory experiments are not usually acted upon until they can be proven in a real-world setting. Leading to substantial damage occurring in the interim, as seen with PCBs. Using the USEPA TRI report, a potential environmental point source of BDE-209 was identified (Roxboro, NC. WWTP effluent) 84 and further verified by a preliminary study. On further evaluation, wastewater sludge was collected on two occasions (2002 and 2005) and analyzed for mono- through deca- BDEs. Results indicated that the source of PBDEs detected in both sludges were from products containing these flame-retardants. Even though the levels of -209 exceeded previous sludge reports by 10-fold, debromination was not believed to be a major contributor to the lower brominated PBDEs detected. Sediments collected at the outfall and approximately 11 kilometers downstream also contained PBDEs, with the major congener being -209. Previous reports have indicated debromination of -209 under some photolytic and anaerobic conditions. However, PBDE sediment profiles observed throughout this river system were consistent with those of the sludge PBDE profiles, indicating minimal -209 debromination in these surficial sediments. PBDEs were also detected in biota collected at the outfall. BDE-209 was detected in two (crayfish and sunfish) of the four composite samples. (A 10-fold -209 increase was detected in the crayfish composite compared to the fish composite, indicting a potentially important exposure route to predators.) Potential debromination products (nona- and octa- PBDEs) were observed in each sample, including two octa- (BDE-201, -202) and three hepta- PBDEs (BDE-188, 184 and -179) in the fish samples. These congeners were not detected in either the sludge or sediment from this site. In a previous laboratory dietary - 209 exposure studies, -209 was reported to undergo metabolic debromination (in the liver) of juvenile rainbow trout and common carp, resulting in two characteristic metabolic PBDEs profiles containing nona- through hexa-PBDEs. The same debromination PBDE profiles were also observed in the sunfish and chub samples collected from the Roxboro WWTP outfall, respectively. These findings confirm that 86 environmentally derived BDE-209 can undergo species-specific metabolic reductive debromination in-vivo, ultimately increasing toxicological burdens and potentials in feral biota. 87 APPENDIX A El SPECTRA, tri-BDE BDE-28 (2A4-TRI-BDE) 100n 246 [M-2Br] [M] 8 0 - 6 0 - 4 0 - 20- m/z 250 300 350 88 APPENDIX B El SPECTRA, tetra-BDE BDE-47 (2,21A 4'-TETRA-BDE) 100-1 326 [M-2Br] [M] 80- 6 0 - 4 0 - 20- m/z 300 350 500 89 APPENDIX C El SPECTRA, penta-BDE B D E-99 (2,2'A A 5-PE NTA-B D E) 100-. [M-2Br] 80- [M] 40- 20- m/z 300 400 500 600 90 APPENDIX D El SPECTRA, hexa-BDE B D E -154 (2,21A 45,6'-HEXA-BDE) 100-, ABA [M-2Br] 80- [M] 60- 20- m/z 600500 700 91 APPENDIX E El SPECTRA, hepta-BDE B D E -18 A (2,2',3,4',6,6-HEPTA-BDE) 100-, 722 [M-2Br] 8 0 - [M] 6 0 - 4 0 - jM. m/z 500 600 700 92 APPENDIX F El SPECTRA, octa-BDE BDE-197 (^'IS'^-me'-OCTA-BDE) 100-i [M-2Br] 802 8 0 - [M] 60- 4 0 - 20 - 535 m/z 600 700 800 93 APPENDIX G El SPECTRA, nona-BDE BDE-206 (2,2', IS'A ^B'^-N O N A -BD E) 100-, 720 [M-2Br] 8 0 - 880 [M] 6 0 - 560 20- 613 m/z 600 700 800 900 94 APPENDIX H El SPECTRA, deca-BDE BDE-209 (decabromodiphenyl ether) 100-i 800 [M-2Br] 80- 6 0 - 960 [M] 20- 693 727 m /z 600 700 800 900 1000 95 APPENDIX I ECNI SPECTRA, hepta-BDE BDE-175 (^W AS'.e-HEPTA-BDE) 100n 809 8 0 - [3-4 substituted] 6 0 - 4 0 - O' O' Br Br 20 - 408.7 160.9 328.8 “1—1—r i | I 11 I I | I I | i i i i | i r m/z 100 200 300 400 500 96 APPENDIX J ECNI SPECTRA, hepta-BDE BDE-190 (2,3,3',4.4',5,6-HEPTA-BDE) 100- 78 9 80- [2-5 substituted] 60- 40- 20 - 160.9 ‘V — i ------1— “I— 1 i T ' 'i— ■]— ‘-I— 11 “ i— ------1------1------r 4— I-----1------1------1------1------1------f — |-----1------r m/z 100 200 300 400 500 97 APPENDIX K ECNI SPECTRA, octa-BDE BDE-197 (^'JX^'^G'-OCTA-BDE) 100-i 809 Br Br 80- [4-4 substituted] 408.7 60- 0 ‘ 40- 20 - 328.8 159.9 T 1---1-- -i------r—— i---- 1------1------i r r t 1------1------r m/z 100 200300 400 500 98 APPENDIX L ECNI SPECTRA, octa-BDE BDE-204(2,2',3,4,4',5,6,6,-OCTA-BDE) 10O-i 789 80- Br Br [3-5 substituted] 60- /o O' 40- O' Br Br 486.6 20- 328. 408.7 jJL JU— .u— U, -I T 1 " T |iir i 1 1 | m i—■—V---- i 1111 |----H—r m/z 100 200 300 400 500 99 APPENDIX M ECNI SPECTRA, nona-BDE BDE-206 (2,23,34,45,56-N0 NA-BDE) 100-. 78.9 80- 6 0- O" 4 0 - 20- 408.7 160.9 m/z 100 200 300 500 100 APPENDIX N ECNI SPECTRA, deca-BDE BDE-209 (DECA-BDE) 100—i 78.9 80- O" 60- 20 - 328.! m/z 100 200 300 500 101 REFERENCES Alaee, M.; Arias, P.; Sjodin, A.; Bergman, A. Environment International 2003, 29, 683- 89. Anderson, 0.; Blomkuist, G. Chemosphere 1981, 10, 1051-1060. Bjorklund, J.; Tollback, P.; Hiame, C.; Dyremark E.; Ostman, C. J. Chromatogr. A2004, 1041, 201-210. BSEF, Bromine Science and Environmental Forum, Brussels, Belgium. Data report at http://www.bsef.com/bromine/our industry/ 2006. Covaci, A.; Voorspoels, S.; de Boer, J. Environment International, 2003, 29, 735-756. Covaci, A.; Gheorghe, A.; Voorspoels, S.; Maervoet, J.; Redeker, E. S.; Blust, R.; Schepens, P. Environ. Int., 2005, 31, 367-375. de Boer, J.; Wester, P. G.; Klamer, H. J. C.; Lewis, W.E.; Boon, J. P. Nature 1998, 28-29. de Boer, J.; Wester, P. G.; van der Horst, A.; Leonards, P. E. G. Environ. Poll. 2003,122, 63 - 74. Dodder, N. G.; Strandberg, B.; Hites, R. Environ. Sci. Technol. 2002, 36, 146-151. Eriksson, J.; Green, N.; Marsh, G.; Bergman, A. Photochemical decomposition of 15 polybrominated diphenyl ether congeners in methanol/water. Environ. Sci. Technol. 2004, 38, 3119-3125. EWG, Environmental Working Group, In the dust toxic fire retardants in American homes, Washington, D.C., U.S. http://www.ewg.org 2005. Gerecke, A. C.; Giger, W.; Hartmann, P. C.; Heeb, N. V.; Kohler, H. E.; Schmid, P.; Zennegg, M.; Kohler, M. Chemosphere 2006, 64, 311 -317. 102 Hale, R. C.; La Guardia, M. J.; Harvey, E. P.; Mainor T. M.; Duff, W. H.; Gaylor, M. O. Environ. Sci. Technol. 2001a, 23, 4585-4591. Hale, R C.; La Guardia, M. J.; Harvey, E. P.; Gaylor, M, O.; Mainor, T. M.; Duff, W, H. Nature 2001b, 412, 140-141. Hale, R. C.; La Guardia, M. J.; Harvey, E. P.; Gaylor, M. O.; Mainor T. M. Chemosphere 2006, 64, 181-186. Hazrati, S.; Harrad, S. Environ. Sci. Technol. 2006, article in press He, J.; Robrock, K. R.; Alvarez-Cohen, L. Environ. Sci. Technol. 2006, 40, 4429-4434. Hoh, E.; Zhu, L.; Hites, R. Environ. Sci. Technol. 2005, 39, 2472 - 2477. Holmqvist, N.; Stenroth, P.; Berglund, O.; Nystrom, P.; Graneli, W.; Larsson, P. Chemosphere, 2006, article in press Jianzhong, H.; Robrock, K. R.; Alvarez-Cohen, L. Environ. Sci. Technol. 2006, 40, 4429 - 4434. Kierkegaard, A.; Balk, L.; Tjamlund, U.; De Wit, C. A.; Jansson, B. Environ. Sci. Tenchnol, 1999, 33, 1612 -1617. Korytar P.; Covaci, A.; de Boer, J.; Gelbin, A.; Brinkman UAT. J Chromatogr. A 2005, 1065:239-249. La Guardia, M.J.; Hale, R.C.; Harvey, E. Abstract, SETAC 24th Annual Meeting in North America, Austin, Texas, 2003. La Guardia, M.J.; Hale, R.C.; Harvey, E. P.; Bush, E. O.; Mainor, T.M.; Gaylor, M.O. Residuals Sci. Tenchnol, 2004a, 2, 111 - 122. La Guardia, M.J.; Hale, R.C.; Harvey, E. Abstract, Fourth SETAC World Congress, 25th Annual Meeting in North America. Portland, Oregon 2004b. 103 La Guardia, M.J.; Hale, R.C.; Harvey, E. Environ. Sci. Technol. 2006, 40, 6247 - 6254. Mai, B.; Chen, S.; Luo, X.; Chen, L.; Yang, Q.; Sheng, G.; Peng, P.; Fu, J.; Zeng, E. Environ. Sci. Technol. 2005, 39, 1861-1867. Meerts, I, A, T, M.; van Zanden, J.J.; Luijks, E.A.C.; van Leeuwen-Bol,I.; Marsh, G.; Jakobsson,E.; Bergman, A.; Brouwer, A. Toxicol, Sci. 2000, 56, 95-104. NCDENR, North Carolina Department of Environment and Natural Resources, Division of Water Quality, Basinwide Assessment Report - Roanoke River Basin, May 2000, 46 - 54. NIST, The NIST Mass Spectral Search Program, NIST/EPA/NIH Mass Spectral Library, Version 2.0a, July 1, 2002. Noren, K.; Meironyte, D. Chemosphere 2000, 40, 1111 -1123. North, K. D. Environ. Sci. Technol. 2004, 38, 4484-4488. Oros, D. R.; Hoover, D.; Rodigari, F.; Crane, D.; Sericano, J. Environ, Sci. Technol. 2005, 39, 3 3 -4 1 . Remberger, M.; Stembeck, J.; Palm, A.; Kaj, L.; Stromberg, K.; Brorstrom-Lunden, E. Chemosphere, 2004, 54, 9 — 21. Renner, R. Environ. Sci. Technol. 2000a, 34, 452A-453A. Renner, R. Environ. Sci. Technol. 2000b, 34, 223A-226A. Renner, R. Environ. Sci. Technol. 2004, 38, 14A. Sellstrom, U.; Jansson, B.; Kiekegaard, A.; de Wit, C. Chemosphere 1993, 26, 1703- 1718. Sellstrom, U.; Kiekegaard, A.; de Wit, C.; Jansson, B. Environ. Toxicol. Chem. 1998, 17 (6), 1065-1072. 104 Schilderman, P. A. E. L.; Moonen, E. J. C.; Maas, L. M.: Wella, I.; Kleinjans, J. C. S. Ecotox. Environ. Safety, 1999, 44, 241-252. Sjodin, A.; Hagmar, L.; Klasson-Wehler, E.; Kronholm-Diab, K.; Jakobsson, E.; Bergman, A. Environ. Health. Perspect. 1999, 107 , 643 - 648. Sjodin, A.; Patterson, D, G.; Bergan, A. Environ. Sci. Technol. 2001, 35, 3830 - 3833. Sjodin, A. Environ. Sci. Technol. 2003, 37, 384A. Song, M.; Chu, S.; Letcher, R. J.; Seth, R. Environ, Sci. Technol. 2006, 40, 6241 - 6246. Soderstrom, G.; Sellstrom, U.; De Wit, C.; Tysklind, M. Photolytic debromination of decabromodiphenyl ether (BDE 209). Environ. Sci. Technol. 2004, 38, 127-132. Stapleton, H. M.; Alaee, M.; Letcher, R. J.; Baker, J. E. Environ, Sci. Technol. 2004a, 38, 112-119. Stapleton, H. M.; Letcher, R. J.; Baker, J. E. Environ, Sci. Technol. 2004b, 38, 1054 - 1061. Stapleton, H. M.; Brazil, B.; Holbrook, D.; Mitchelmore, C. L.; Benedict, R.; Konstantinov, A.; Potter, D. Environ, Sci. Technol. 2006, 40, 4653 - 4658. Takasuga, T.; Kurunthachalam, S.; Hiroaki, T.; Ohi, E.; Tsuji, H.; Nagayama, J. Chemosphere, 2004, 57, 795-811. TRI, Toxics Release Inventory (TRI) Program, USEPA., http://www.epa.gov/tri/ 2006. USEPAs Draft Method 1614. Brominated diphenyl ethers in water, soil, sediment, and tissue by HRGC/HRMS. Watanabe, I. and Tatsukawa, R. Bull. Environ. Contam. Toxicol. 1987, 39, 953-959. WHO, Environmental Health Criteria 162: Brominated Diphenyl Ethers; 1994, World Health Organization: Geneva, Switzerland. Zennegg, M.; Kohler, M. Environ. Sci. Technol. 2 0 0 5 , 39, 1078-1083. 106 VITA Mark Joseph La Guardia Mark was bom in Smithtown, New York on June 1, 1963. He graduated from Ward Melville High School, East Setauket, New York in June 1982. Served on active duty in the United States Navy 1983 to 1987 and earned an Associate in Science degree January 1988 from The University of the State of New York. He continued his education at Old Dominion University, Virginia, where he received a Bachelor of Science degree in Chemistry, minor concentration in Oceanography December 1992. After graduating, Mark was employed as a Quality Control Chemist for a chemical manufacture Allied Colloids, Suffolk, Virginia and later joined an environmental engineering firm Environmental Engineering and Technology, Newport News, Virginia as Laboratory Manager. In 1998, he joined the Virginia Institute of Marine Science (VIMS), The College of William and Mary, as a Marine Scientist. Accepting his present role as Senior Marine Scientist in 2001. The author enrolled in the Master of Science program at The College of William and Mary in January 2001 and defended his thesis December 2006. After graduation, the author plans on continuing to work on his research interest with his current employer, VIMS.