Document J3vJo1oqGdN54kq59w8wqLL4K

AR226-2710 32 AR226-2710 Aerobic Biotransformation of ^C-labeled 8-2 Telomer B Alcohol by Activated Sludge from a Domestic Sewage Treatment Plant Ning Wang*t, Bogdan Szostekf, Patrick W. Folsomt, Lisa M. Suleckif. Vladimir Capkat", Robert C. Buck, William R. Bertif, and John T. Gannonf * Corresponding author phone: (302)366-6665; fax: (302)366-6602; email: ning.wang@usa.dupont.com o f DuPont Central Research and Development, Glasgow Business Community 301, PO Box 6101, Newark, DE 19714 t DuPont Haskell Laboratory for Health and Environmental Sciences, Newark, DE 19714 ;;: '. -'? DuPont Chemical Solutions Enterprise, Wilmington, DE 19805 ..:; !! Present Address: Tandem Labs, A DivisionofNWT Inc., Salt Lake City, UT 84124 This study investigated the biodegradation potential of S-^C, 1H, IH, 2H, 2H-perfluorodecanol [CF3(CF2)614CF2CH2CH20H^, C-labeled 8-2 Telomer B Alcohol or '^-labeled 8-2 TBA] by diluted activated sludge from a domestic waste water treatment plant under aerobic conditions. After sample extraction with acetonitrile, biotransformation products were separated and quantified by LC/ARC (on- line liquid chromatography/accurate radioisotope counting) with a limit of quantification (LOQ) about 0.5% of me ^C counts applied to the test systems. Identification of biotransformation products was performed by quadrupole time of flight (Q-TOF) mass spectrometry. Three transformation products have been identified: CF3(CF2)614CF2CH2COOH(8-2 saturated acid), CF3(CF2)614CF=CHCOOH(8-2 Subject to Copyright. Do not reference. Not for further reproduction unsaturated acid), and CF3(CF2)6'4COOH(perfluorooctanoic acid, PFOA), representing 27, 6.0, and 2.1% of the initial ^C mass (^C counts applied) after 28 days, respectively. A transformation product, not yet reported in the literature, has also been observed and tentatively identified as CF3(CF2)614CH2CH2COO(H2H, 2H, 3H, 3H-perfluorodecanoic acid); it accounted for 2.3% of the mass balance after 28 days. The 2H, 2H, 3H, 3H-perfluorodecanoic acid is likely a substrate for B-oxidation, which represents one of the possible pathways for 8-2 Telomer B Alcohol degradation. The 8-2 saturated acid and 8-2 unsaturated acid cannot be directly used as substrates for B-oxidation due to the proton deficiency in their B-carbon (3 carbon) and their farther catabolism may be catalyzed by some other still unknown mechanisms. The 2H, 2H, 3H, 3H-perfluorodecanoic acid may originate either from the major transformation product, CF3(CF2)614CF2CH2COOHor from other unidentified transformation products via multiple steps. Approximately 57% of the starting material remained unchanged after 28 days, likely due to its strong adsorption to the PTFE (polytetrafluoroethylene) septa of the test vessels. No CF3(CF2)614CF2COO(Hperfluorononanoic acid) was observed, indicating that aoxidation of CF3(CF2)614CF2CH2COOHdid not occur under the-study conditions. Several "C-labeled transformation products that have not yet been identified (each less than 1% of the mass balance) were; also observed and together accounted for 7% of the total ^C mass balance after 28 days. It is not clear whether these unidentified transformation products was resulting from farther metabolism of 8-2 saturated acid or 8-2 unsaturated acid. The results suggest that perfluorinated acid metabolites such as perfluorooctanoic acid account for only a very small portion of the transformation products observed. Also, the observed volatility and bioavailability of ^C-labeled 8-2 TBA for mierobial degradation was markedly decreased as a result of the presence of a strongly adsorbing matrix such as PTFE in the experimental systems. It is apparent that the biological fate of 8-2 Telomer B alcohol is determined by multiple degradation pathways, with neither B-oxidation nor any other enzyme-catalyzed reactions as a single dominant (principal) mechanism under the study conditions. -ni , r i.* i^ Subject to Copyright. Do not reference. Not for further reproduction Introduction There is growing interest in the environmental fate and effects of fluorinated chemicals in the environment. As a result of the identification and ubiquitous global presence of perfluorooctane sulfonate [F(CF2)8S03"1 in the environment and humans and to a lesser extent perfluorooctanoic acid IF(CF2)7COOH], the environmental fate and distribution of perfluorinated chemicals (PFCs) is of growing interest in order to understand the sources of these chemicals in the environment (1-6). Carbon-fluorine bonds are extremely strong (7). As a result, it is generally accepted that '? perfluorocarbon chains do not readily biodegrade and any biodegradation may be limited to hydrocarbon functionality to which a perfluorocarbon moeity is attached. Interest extends beyond perfluorooctane sulfonate and perfluorooctanoic acid to the broader class of perfluorinated chemicals which includes fluorotelomer alcohols [FTOHs]. 8-2 Telomer B Alcohol [8-2 TBA, F(CF2)gCH2CH20H, CAS# 678-39-7] is one of the raw material intermediates used in the manufacture of fluorotelomer-based products. Fluorotelomer' alcohols are comprised of an even number of fluorinated carbons appended to an ethanol moiety.' >The fltioratelomer functionality in products delivers unique surface modification properties including water'and oil repellency (polymeric products) and wettingand leveling (surfactants) (8-10). i" The environmental fate and routes into the environment of fluorotelomer-based substances is still largely unknown. Biodegradation is one transformation route that will determine the fate of 8-2 TBA in the environment. A limited number of experimental investigations have been conducted to investigate microbial biodegradation of FTOHs and telomer-based substances (11-16). The same is true regarding metabolism in living systems and pharmacokinetics (17, 18). Potential transformation pathways have been reported from predictive models based upon known abiotic and biotic functional group transformations and their probabilities (5). After submission of the initial draft of this study, a new paper was published (19), which reports the biodegradation of non-radioisotope labeled 8-2 Telomer B Alcohol by microbial enrichment culture. In this study, the authors quantified three known transformation products, 8-2 saturated acid, 8-2 unsaturated acid, and PFOA and found that 8-2 Subject to Copyright. Do not reference. Not for further reproduction unsaturated acid is the major metabolite and PFOA accounts for about 3% of total mass at day-81. The authors conclude that the fi-oxidation is the principal pathway that determines the fate of 8-2 Telomer B Alcohol. In Ihis study, PFOA concentration also decreased significantly from about 6% of total mass at day 51 to 3% at day 81. Because the above studies in general were limited to identification and quantification of transformation products already known from prior literature or predictive models, potential unknown transformation products were not identified or quantified. Hence, the transformation pathway(s) and stable transformation products remain unclear. In addition, while widely believed to be the major transformation product from microbiological transformation of 8-2 TBA, perfluorooctanoic acid, F(CF2)7COOH, was only identified in small quantities. 8-2 TBA is highly surface active, sorptive (Unpublished results by Telomer Research Program), and both hydro- and oleophobic (20). Additionally, 8-2 TBA has exhibited unique physical-chemical properties which make it a challenging test substance. As a result, test substance bioavailability and partitioning in the test system presents real challenges, Headspace migration, strong, potentiallytnt&versible sorption, and rapid migration to test system surfaces have been reported (20, 21). Final't% the'iisolafal, identification and accurate quantitation of the transformation product(s) as well as-obtaiMigriaothentic standards of them has been limited thus far by testing ofnon-radiolabeled material. We report here the first experimental biodegradation study to achieve mass balance which more clearly defines the biological fate of a fluorotelomer alcohol. To achieve mass balance and overcome identification and quantitation problems, a custom synthesized ^C-labeled 8-2 TBA [F(CF2)714CF2CH2CH2QHw]a, s used to investigated the biodegradation potential and transformation products after incubation with activated sludge inoculums from a domestic waste water treatment plant The use of ^C-labeled material allowed us to quantify and identify unequivocally potential transformation products down to a LOQ of 0.5% of initial ^C applied to the experimental system. The objective of this study is to assess whether 8-2 TBA can be readily transformed aerobicaUy and if so, discern the likely biotransfonnation pathway(s) and products. Subject to Copyright. Do not reference. Not for further reproduction Experimental Section Materials and Methods. Non-radioisotope labeled 8-2 TBA (CF3(CF2)6CF2CH2CH20H, 8-2 TBA) was from TRP (Telomer Research Program) and had a 99.9% purity. The ^C-label 8-2 TBA (CF^(CF2)614CF2CH2CH20Hw)as custom synthesized with a radiochemical purity of virtually 100%. No ^C impurity was detected with a LOQ of- 0.02% as analyzed by HPLC coupled with a ^ flow through detector. The specific activity was 54.3 mCi nmiol'1 based on ^C counts on a mass basis and was confirmed by gas chromatography-mass spectrometry (GC/MS) analysis to be within 2% difference. The chemical purity was 93% with CF3(CF2)6CH2CH:20H as the major impurity. The re label 8-2 TBA was dissolved in absolute ethanol (purity = 99.97%, Aldrich Chemical Co., Milwaukee, Wisconsin, USA) to about 2.2 mCi mL'1 ethanol as the stock solution and was stored at ~ -10C. The positive control reference chemical aniline-HCl (CfrHpN-HCl) was from Sigma Chemical Co. (St. Louis, USA) with a purity of 99.6% and uniformly labeled aniline-HCl (^C^N-HCl or '^(Uhaniline-HCl) was from Moravek Biochemicals (Brea, USA) with a specific activity of 78 m(3ii;mmol"1 and radiochemical purity of 99.1%. Three milliliters of 1 g aniline-HCl L'1 aqueous solution wasi added itoi" 0.28 mCi ^CO^-aniline-HCl solids to give a final concentration of approx. 1.2 g L"1 as the stock solution. All other solvents used were purity > 99% and all other chemicals were reagent grade or higher. The water used throughout me experiment was deionized water ("-17.6 MQ-cm) from Bamstead E-pure system. Activated sludge (2 L in a 4-L container) was collected the same day the experiment was initiated from the City ofWilmington (Delaware, USA) Municipal Waste Water Treatment Facility - Aeration Basin #2. After arriving at the lab, the sludge was mixed by shaking to suspend me microorganisms. After settling for approx. 14 min, the upper aqueous phase oftihe sludge was used as the inoculums or was autoclaved and then used as abiotic control. Experimental System. Because of the semi-volatile nature of the starting material (20), a sealed test system was used to prevent loss of "C-labeled 8-2 TBA due to volatilization. Glass serum bottles (120-mL volume) were used as the test vessels. The mineral medium used in this study was composed of 8.5 mg L-' ofKH2P04,21.8 mg L-1 ofiyHPO^, 33.4 mg L-1 ofNa2HP04-2H20, 0.5 mg L-1 Subject to Copyright. Do not reference. Not for further reproduction of NHLtCl, 36.4 mg L-1 ofCaCk-'iW, 22.5 mg L-1 of MgSOrTHzO, and 0.25 mg L-1 of FeGb^HaO, and the pH was adjusted to 7.0 (22). The test and appropriate control media were prepared in 0.5 to 1 L polypropylene containers and then dispersed with glass pipettes into individual serum bottles. Five different treatments were included in the test design to assure study integrity. 1) Biodegradation test vessels (4 replicates at each sampling time point): contain 30 mL test medium composed of activated sludge (0.5%), 0.48 uL ^C-label 8-2 TEA stock solution, and about % saturated (--80 ug L'1) 8-2 TBA solution made in mineral medium. 2) Abiotic control vessels (4 replicates at each sampling time point): contain autoclaved sludge (0.5%) plus 0.5 roM NaCN were substituted the live sludge and the rest of the components were identical as in biodegradation test vessels. 3) Spike recovery vessels (4 replicates at each sampling time point): contain 30-ml test medium with activated sludge (0.5%) in mineral medium, 5 mL of which was withdrawn for fluoride analysis before being spiked (dosed) at each sampling time point (days 0, 7, 14, and 28) with 50 uL of 226 mg L-1 ^C-labeled 8-2 TBA and 0.5 mL of 4 mg L-' freshly prepared standard fluorinated acids, CF3(CF2)6CF2CH2COOH (8-2 saturated acid), -- .'.-sAteriip CF3(CF2)6CF=CHCOOH (8-2 unsaturated acid), CFstCF^COOH (PFOA), and CF3(CF2)4C00H ..;<, (PFHA). The '^-labeled spike solution (226 mg L'1) was prepared by mixing 2 mL of 150 mg '8-2 TBA L'' in ethanol and 8 uL original "C-label 8-2 TBA stock solution (specific activity s= 54.3 mCi mmol"1). 4) Sample matrix vessels (2 replicates at each sampling time point): contain 30-ml test medium with activated sludge (0.5%) in mineral medium to be used as a blank solution for LC/MS/MS analysis of spiked fluorinated acids. 5) '^(t^-aniline-HCl positive control vessels (2 replicates at each sampling time point) to assess the inoculum microbial activity over time: contain 30-ml test medium with activated sludge (0.5%) in mineral medium plus 30 uL of 1.2 g "C^-aniline-HCl L'1 stock solution. Test Conditions aad Sampling. After each test vessel was filled with appropriate test medium, the bottle was crimp-sealed with a PTFE septum (pre-washed with methanol and sterile water)/aluminum cap. It should be noted that the biodegradation system in this study contained only about 13 mg L"1 organic carbon, which included the microbial inoculum and the ethanol as a co-solvent Subject to Copyright. Do not reference. Not for further reproduction to disperse the starting material. This low amount of organic carbon plus enough headspace (90 mL air) assured aerobic conditions in sealed serum bottles. The diluted sludge (0.5%, 200-fold dilution) contributed to about 1 mg organic carbon L"1. The inoculum level was also low (estimated to be approx. 105 bacterial cells per bottle), so that the test conditions were comparable to the "Ready Biodegradation" test according to the study guide provided by Organization for Economic Co-Operation and Development (22). All the glass serum bottles were shaken at 250 rpm in an environmental incubator in the dark and at room temperature (ranged 25 to 27.7 C monitored by a Dickson chart recorder during 28 days). The 90-mL air space inside a vessel ensured an aerobic environment. The test vessels from different experimental treatments were sampled at days 0, 7, 14, and 28 for extraction and analysis. Sample Extraction and Analysis. To quantify the initial concentration of the starting material, triplicate 30 mL medium from the 1 L polypropylene bottles for treatments 1 and 2 were extracted with 50 mL MTBE-0.1 M H2S04 for 1.5 h and the MTBE phase (top phase of the extract) was used for <-H quantification of 8-2 TBA and ^C-labeled 8-2 TBA by GC/MS; as described (23). On days 0, 7, 14, ' ' and 28, 5 mL of the test medium from treatments 1, 2, 3, and>4 above was withdrawn from each of the serum bottles with a syringe through a 26-gauge needle and was treated with 50 uL of 5N NaOH first and then neutralized with 42 pL of6NH2S04 for fluoride analysis. The remaining 25 mL medium in each bottle was extracted with 45 mL ofacetonitrile (injected into the sealed bottle) for approx. 1 h at room temperature (first extraction). Due to possible injection wound (piercing damage) to the septa, some of the bottles were then re-sealed with fresh PFTE septa/aluminum cap to assure an air-tight seal during long-term storage at - -10 C. The acetonitrile extract from each sample bottle was used for liquid scintillation counting of ^C, for quantification and identification ofl C-labeled transformation products using LC/ARC, and for LC/MS/MS analysis of fluorinated acids. To measure the mineralization of ^CCU^-aniline-HClby the inoculum (Treatment 5), 0.1 mL aliquot from the serum bottles was mixed with 0.2 mL of 0.1 M HCl to remove ^COz remaining in the medium and 5 mL of Subject to Copyright. Do not reference. Not for further reproduction scintillation cocktail was added to the 0,3 mL mixed solution for "C counting by a liquid scintillation counter. It was found later that initial ^C, counts of the acetonitrile extract (from first extraction) did not account for all ^C applied as ^C-labeled 8-2 TBA at day 0. For this reason, a second sequential extraction was carried out for the serum bottles and septa separately. Each sample bottle was reextracted with 10 mL acetonitrile for 3 to 4 h at room temperature after rinsing 2 times with deionized water, and each PTFE septum from individual sample serum bottles, if saved, was re-extracted with 10 mL acetonitrile 4 to 5 times at 50C (each extraction lasted 4 to 7 days) and l4C activity was measured for mass balance calculation. Also, some of the acetonitrile solution was pooled from each sample for LC/ARC analysis to confirm whether extended extraction at 50C caused any degradation of ^C-labeled 8-2 TBA. As mentioned above, some of the original septa (piercing damaged) were not saved for the re-extraction. Fluoride ion was measured by fluoride-selective electrode after mixing the sample.with equal volume (5 mL) of Thermo Orion TISABII buffer (15) using 5 to 500.'ug NaF L"' as standard calibration. Fluorinated acids in the spiked samples were quantified by iLC/MS/MS in negative electrospray ionization mode by a Waters 2795 HPLC/Micromass Quattro Micro system. The 10-50 ug L'1 calibration standards were made in sample matrix blank (from Treatment 4) and the injection volume was 10 uL. The HPLC column used is a reverse phase Zorbax Rx-Cg (150 >; 2.1 mm, 5 um particle size. Mobile phase was 0.15% acetic acid/acetonitrile in a gradient with a flow rate of 0.4 mL min'1. Multiple reaction monitoring (313 > 269 for CF3(CF2)4COOH, 413 > 369 for CF3(CF2)6COOH, 457 > 393 for CF3(CF2)6CF=CHCOOH, and 477 > 393 for CF3(CF2)6CF2CH2COOH) was used for compound quantification. Analysis of "C-labeled 8-2 TBA and ^C-labeled Transformation Products by LC/ARC. Original acetonitrile-extracted samples from Treatment 1 and 2 in sealed glass GC vials were diluted 1:1 with 2 mM ammonium acetate in water. The diluted samples were analyzed by LC/ARC to quantify '^-labeled 8-2 TBA and ^-labeled transformation products by integrating radioactivity of apparent Subject to Copyright. Do not reference. Not for further reproduction (baseline-resolved) chromatograptlie peaks during the separation. The LC/ARC system (AIM Research Company, Newark, USA) utilizes advanced stop flow counting technologies to accurately detect and quantify radioisotope with a sensitivity of several CPM (counts per minute) above background level (24). The initial analysis (LC/ARC Method 1) used a Zorbax Rx-Cig column (4.6 nun x 150 mm, 5 urn particle size) and mobile phase was 2 mM ammonium acetate/acetonitrile in a gradient with a flow rate of 1.0 mL min"1. The injection volume was 0.5 mL, the fraction size was 10 sec, and me ^C counting time was 120 sec for each fraction. The initial analysis (LC/ARC Method 1) was unable to separate CF3(CT2)614CF2<%COOHfrom CF3(CF2)614CF=CHCOOHto give individual transformation product peaks. A second chromatographic method (LC/ARC Method 2) was used, with 0.15% acetic acid/acetonitnie as mobile phase, to farther separate and quantify me two ^C-labeled transformation products. Identification of "C-Iabeled Transformation Products by LC/MS Analysis. A HPLC (Alliance HT, Model 2795, Waters)/Q-TOF (Micromass) system in negative', electrospray ;ionization mode was used to identify the transformation products observed with the LC/ARC system. :iThe chromatographic conditions of the LC/MS analysis were kept essentially the. .same as for LC/ARC analysis in order to match the retention times of the observed signals in the LC/MS and LC/ARCanalysis as closely as possible. The LC/MS analysis was done off-line from the LC/ARC system. A Zorbax Rx-Cig (2.1 mm x 150 mm, 5 u,m particle size) column was used with a gradient mobile phase ( 0,15% acetic aeid/acetonitrile) and a flow rate of 0.25 mL min'1; the lower flow rate was used to compensate for column diameter change from 4.6 nun to 2.1 mm and was necessary for proper electrospray operation. Other MS parameters are as follows: capillary voltage of 2.5 kV, cone voltage of 10 V, source temperature of 120C, and desolvation temperature of 350C. The transformation product identification was based on matching the retention time, observed molecular ion, daughter ion spectrum, and accurate mass measurement of the observed transformation product with that of respective standards. The retention time regions of the LC/MS ehromatograms where the LC/ARC Subject to Copyright. Do not reference. Not for further reproduction peaks were observed were carefully examined to identify transformation products that could not be matched with available standards. "C Radioactivity Measurement By Liquid Scintillation Counting. Throughout the study, 5 mL aliquots of liquid scintillation cocktail (Packard BioScience Ultima Gold XR) were mixed with processed samples for ^C radioactivity counting by a liquid scintillation counter (Beckman LS 5000TD). The counting time was normally 10 min to reduce counting error to within 2% except for samples with only background or close to a background level of ^C radioactivity. Typical counting efficiency was >95% and the final radioactivity was automatically calculated and reported as DPM based on: (CPM/counting efficiency) = DPM. Results and Discussion The measured starting concentration of the parent material [CF3(CF2)614CF2CH2CH20Hplus nonlabeled CF3(CF2)6CF2CH2CH2QH3 was 337 13 ug L'1 for the test vessels and 306 9 ug L-1 for , : s ; ? abiotic control vessels, which is about 2.3-fold above 8-2 TBA solubility limit (20). The purpose of L su"n ' using the concentration above the solubility of the test chemical is to increase the sensitivi1y,ifor':'< 181M|"' detecting the ^C-labeled transformation products. Based on the amount of ^C counts ayailableito -each ; ,. '-, test vessel after the sample extraction, the LOQ for ^C-labeled transformation products was approx. 0.5% of the initial "C counts applied. The recoveries of the starting material and fluorinated acid standards that had been spiked (dosed) into the test medium were acceptable. The recoveries for spiked test medium averaged 92 7% for "C-labeled 8-2 TBA, 100 20% for CF3(CF2)6CF2CH2COOH, 81 21% for CF3(CF2)6CF=CHCOOH, 104 14% for CF3(CF2)6COOH, and 100 20% for CF3(CF2)4COOH (n = 16 for day 0,7,14, and 28 samples). The reference compound, ^C-labeled aniline-HCl with a starting concentration of approx. 1.2 mg L'1, degraded over time, with 45% of total ^C counts remaining at day 7 compared with that of day 0,41% at day 14, and 29% at day 28 (n = 2 samples), indicating that more than 70% of the ^C-labeled aniline-HCl had been mineralized. C02 evolution in excess of 50% of theoretical is interpreted as complete or near-complete biodegradation of a test chemical because a substantial portion of ^C 10 Subject to Copyright. Do not reference. Not for further reproduction applied was incorporated into the structural components of the microbes (25). This suggests that the sludge inoculums used in this study were metabolically active during the test Formation of Biotransformation Products. Under the test conditions with live microbial inoculums, the starting material (parent) was readily transformed to various transformation products (Figure 1). The LC/ARC Method 1 (Figure 1A, left column) was initially used to separate and quantify me "C-labeled transformation products. As shown in Figure 1A, at day 28, one major peak with a retention time (tn) of 21.6 min emerged in addition to the parent (peak 5, tn =; 33.6 min) and two additional smaller peaks (peaks 1 and 4) with retention times of approx. 20 and 23 min are clearly visible. For theabiotic control, only the parentwas observed at day 28. These results demonstrate that the formation of transformation products was due to microbial activity. An LC/MS analysis using analogous chromatographic conditions revealed that the major peak (tn = 21.6 in Figure 1A) contained two '^-labeled transformation products. A second LC/ARC method. Method 2, was developed that . allowed efficient separation and quantification of all observed transformation products (Figure 1B, right <ii- w column). -' '.'. y ft"K:' '. Identification of Biotransformation Products. The identification of biotransfonnation; products was conducted using negative ion electrospray ionization on an LC/Q-TOF MS system and applying chromatographic conditions matching those used on the LC-ARC system. Figure 2 shows the daughter ion spectra of major transformation products (LC/ARC peaks 1,2, 3, and 4 of figure 1B). The identity of transformation products (metabolites) 1, 2, 3 was elucidated by matching the retention times and daughter ion spectra of the respective authentic standards [CP3(CF2)6COOH, CF3(CF2)6CF!<;HCOOH, and CF3(CF2)6CP2CH2COOH] with those of the "C-labeled transformation products. The peaks 1, 2, and 3 were identified as CP3(CP2)614COOHC, F3(CF2)614CF2CH2COOHan, d CF3(CF2)614CF;<;HCOOH, respectively. The daughter ion spectra of identified transformation products are presented in Figure 2A-C. The positive identification was based on that me daughter ion spectrum of a metabolite from the sample extract is essentially identical to that of a corresponding standard. The match of the spectrum of an unknown metabolite and a standard is evaluated by the &., 11 Subject to Copyright. Do not reference. Not for further reproduction comparison of observed masses of fragment ions in the daughter ion spectrum and their relative intensities in the spectrum. Taking into account that observed metabolites contain one ^C atom and therefore, some of the resulting fragment ions would be shifted two mass units. Full agreement was observed between the ions in the daughter ion spectrum of metabolites and non-labeled standards as well as their relative intensities. Further conformation of the observed transformation products structures was obtained by accurate mass measurements using LC/Q-TOF MS (Table 1). The measured masses agreed well with the calculated masses for the proposed elemental composition of transformation products, especially for (he two transformation products (LC/ARC peaks 2 and 3 of Figure IB) observed at higher concentration. The accurate mass measurement for CF3(CF2)6COOH exhibited highest error, outside of the typically accepted level of 5 pptn. The Signal for this transformation product was very weak, which in turn affected the accuracy of the mass measurement. The above three transformation products had also been reported in blood plasma of male rats after a single oral dose (11), in mixed bacterial cultures (15) and in enrichment culture dosed with 8-2 TBA (19), and are also predicted by the CATABOL software (5). This suggests that common biodegradation pathways for 8-2 TBA may be shared between animals and microorganisms. Accurate mass measurement for me ion m/z 443 (LC/ARC peak 4 of Figure IB) pointed to an elemental composition of the deprotonated molecular ion containing four hydrogens (Table 1). Further evidence of the presence of four hydrogens in the molecule was obtained from the daughter ion spectrum of ion m/z 443 (Figure 2D). The ions observed in the daughter ion spectrum of m/z 443 can be rationalized as follows: ro/z 339 represents loss ofCOa (molecular weight = 44) and of3HF(3 x 20 = 60) from me deprotonated molecular ion (m/z 443) and m/z 319 represents additional loss ofHF from ion m/z 339. The fragmentation pattern of m/z 443 closely resembles the fragmentation patterns observed for standards CF3(CF2)6CF2CH2COOH and CF3(CF2)6CF=CHCOOH that would predominantly form ions resulting from losses of CC>2 and HF from the deprotonated molecular ion. However, it is not possible from the daughter ion spectrum of m/z 443 alone to assign the placement of 12 Subject to Copyright. Do not reference. Not for further reproduction the four hydrogens in the carbon chain of the molecule. The structure for peak 4 was postulated as: CF3(CF2)614CH2CH2COOH2H, , 2H, 3H, 3H-perfluorodecanoic acid, a transformation product that has not been reported in the literature previously. A standard for the postulated structure is not commercially available. Further support of the proposed structure was obtained from studies of a commercially available analog: 2H, 2H, 3H, 3Hnonanoic acid (CF3(CF2)5CH2CH2COOH). Detailed comparison of ^F-NMR spectra, LC/MS daughter ion spectra, and electron impact (El) GC7MS spectra of methyl and trimethyl sityl (TMS) derivatives of the transformation product (peak 4) derived from biodegradation studies conducted with non-labeled 82 TBA and the analog 2H, 2H, 3H, 3H-nonanoic acid clearly supports the proposed structure for the unknown peak 4 (unpublished results). The quantities of the unknown compound (peak 4) generated in this study were not sufficient to obtain the above data. However, a very good match was obtained for the retention time and the daughter ion spectrum of the metabolite obtained from a non-labeled 8-2 TBA biodegradation study and .the peak 4 data from this study. However,'matching all the spectral data obtained for me peak 4 metabolite with that of an authentic standard (CT^C^f^CWsGOOH) is still required for a final structural confirmation of this new transformation product.: n. . Among the unidentified "C-labeled transformation products, no base-line resolved LC/ARC peak corresponding to ^C-labeled perfluorononanoic acid (CF3(CF2)614CF2COOHw) as visible, suggesting that a-oxidation (26) of CF3(CF2)614CF2CH2COOHt,he major transformation product, did not occur. This is consistent with the result ofDinglasan et al. (19), in which CF3(CF2)6CF2COOH and CF3(CF2)4CF2COOH were not detected in the experiment system. This conveys the notion that potential biodegradation of fluorotelomers in the environment may not lead to the formation of significant amount of odd-numbered perfluorinated carboxylic acids such as perfluorononanoic acid due to microbial a-oxidation. Mass Balance. Compared with day 0, the total ^C radioactivity in the acetonitrile extract (from the first extraction) decreased continuously over 28 days in both the biodegradation test vessels and abiotic controls (Table 2). It was initially thought that the decrease was due to both adsorption to the 13 Subject to Copyright. Do not reference. Not for further reproduction test vessel surface and volatilization of the parent. However, after analyzing the ^C counts recovered from the glass surface and PTFE septa, it was shown that the decrease was actually caused by the strong adsorption of the parent to the PTFE septa of the sealed vessels. The ^C recovered from the septa was 8-2 TBA. After taking into account the UC counts recovered from the septa, the mass balance of the parent plus all transformation products was nearly 100% (Table 2). This observation demonstrates that me potential volatilization of 8-2 TBA from the experimental systems was minimized at the presence of a strongly adsorbing matrix such as PTFE; otherwise, it is impossible to achieve a near 100% mass balance. The adsorption of ^C-labeled 8-2 TBA to the glass was minimal or at least was reversible as it can be easily recovered by acetonitrile at room temperature (Table 2). In this regard, although some of the original septa (days 14 and 28 for the biodegradation test vessels and day 28 for the abiotic control) were not saved after the first acetonitrile extraction for subsequent re-extraction and l4C counting, it is reasonable to assume that most of me ^C counts not recovered from the first extraction also reside in these septa. Also, available septum ^C counts data (Table 2) indicate that the degree of septum adsorption of 8-2 TBA increased over time. The adsorption to the PTFE septa was so steengdhat it took about 4 weeks at 50 C to recover most of the "C counts from the septa using acetonitrile. The parent, CF3(CF2)614CF2CH2CH20Ha,ccounted for virtually all "C counts recovered from the septa as revealed by LC/ARC analysis, indicating 8-2 TBA was remarkably stable abioticaUy in aqueous solution under moderately high temperature. Based on the "C counts of individual transformation products versus the total ^C applied at day 0, CF3(CF2)614CF2CH2COOHis clearly the most abundant transformation product accounting for about 27% of the mass balance at day 28 (Table 2). The CF3(CF2)614CF:;=CHCOOwHas the second major transformation product accounting for about 6.0% of the mass balance at day 28, In contrast, CF3(CF2)6CF=CHCOOH was found to be the predominant metabolite in mixed bacterial cultures (15) and in microbial enrichment culture (19), in which a much higher bacterial density (microbial loading) in the test media and more favorable conditions may accelerate the conversion ofCFa(CF2)6CF2CH2COOH to CF3(CF2)6CF=CHCOOH, in comparison with the very low bacterial loading and organic carbon supply used in this study. The newly identified 14 Subject to Copyright. Do not reference. Not for further reproduction transformation product, CF3(CF2)614CH2CH2COOHaccounted for about 2.3%, and CF3(CF2)614COOH accounted for about 2.1% of the mass balance at day 28. The parent still contribute about 57% of the mass balance at day 28, about 41% of which resulted from adsorption to the septa. It appears mat the strong adsorption of me parent to the PTFE septa during the test reduced its bioavailability for microbial biodegradation. Defluorination of "C-labeled 8-2 TBA during the Biotransformation. Under me test conditions, the increase offluoride ion due to defluorination of the parent was not discernible from the background level offluoride present in the test vessels. The level of calculated fluoride increase (0.9 ug L'1 at day 7, 1.4 ug L"1 at day 14, and 2.0 ug L"1 at day 28 compared with day 0) based on the biotransformation products formed is low compared with that of the background level (day 0 level) of fluoride in the test vessels (13.7 1.0 ug L'1 for n =- 4 samples), and, thus, made the increase indistinguishable from the background. This observation was consistent with the result that ..'MI- . CFatC^s^FaCHaCOOH is the most abundant, transformation product, whose formation does not-;. ma', K),,- involve defluorination. On the other hand, a significant increase of fluoride ion concentration was observed when CF3(CF2)6CF==CHCOOH became a major transformation product, which involved '^v^SS^ .L- ;cvs>w;. defluorination of the non-labeled 8-2 TBA (15). Biotransformation Pathways of UC labeled 8-2 TBA. Several pathways may be available to convert the parent, CF3(CF2)614CF2CH2CH20Hto, various observed transformation products (Figure 3). The first pathway for CF3(CF2)614CF2CH2CH20trHansformation is conversion to CF3(CF2)614COOH most likely via HF elimination (19) and monooxygenase-mediated reactions. First, the parent can be oxidized to the major transformation product, CF3(CF2)614CF2CH2COO(HI), via alcohol and aldehyde dehydrogenase reactions. No ^C-labeled fluoroaldehyde (CF3(CF2)614CF2CH2CHwOa)s detected with a LOO of 0.5% of total mass, indicating that the fluoroaldehyde is either unstable or may be quickly oxidized by an aldehyde dehydrogenase before day 7 sampling. The fluoroaldehyde was detected as a transient intermediate (19) and its level cannot be quantified, indicating that the observed level may be low. After forming CF3(CF2)614CF2CH2COOHth,is acid may be further converted in multiple steps or Subject to Copyright. Do not reference. Not for further reproduction via HP elimination to form CF3(CF2)6l4CF==CHCOOH(II) and then potentially to CF3(CF2)6KCOOH perhaps catalyzed by monooxygenase-mediated reactions, as also predicted by CATABOL (5). Both I and II are the principal stable transformation products and do not rapidly degrade under the test conditions. A second transformation pathway may involve the B-oxidation of the newly identified transformation product, CF3(CF2)614CH2CH2COOH(III). First, the ^C-labeled 8-2 TBA parent may be converted in multiple enzymatic reactions to form CF^(CF2)614CH2CH2COOHa,lthough the mechanisms of these reactions are currently unknown. Incubation of 2 mg L'1 of CF3(CF2)6CF2CH2COOH in mixed bacterial cultures resulted in the formation of only trace amount of CF3(CF2)6CH2CH2COOH (N Wang and B Szostek, unpublished results) , suggesting that some other transformation products may also contribute the formation of CP3(CP2)614CH2CH2COOvHia multiple reactions. Because CF3(CF2)614CH2CH2COOHis a likely fatty acid analog, it can be further oxidized ' via B-oxidation reactions. However, direct experimental evidence is still needed to further confirm this ".' hypothesis. . , . Although long-chain fluorinated carboxylic acids can induce the proliferation ofperoxisomal B- oxidation in rats and mice (27), it is not clear whether such acids can be directly used as substrates for the B-oxidation reactions (28, 29). It has been proposed that 8-2 saturated acid and 8-2 unsaturated acid can be used as substrates for B-oxidation to form PFOA and the B-oxidation pathway may be a principal fate of fluorotelomers such as 8-2 Telomer B Alcohol (19). However, based on current knowledge regarding the enzymology of this pathway, a direct B-oxidation of 8-2 saturated acid and 8-2 unsaturated acid cannot occur. The B-oxidation pathway for fatty acids is well understood and is illustrated by Nelson and Cox (29). As Nelson and Cox have stated, in one pass through me B-oxidation sequence, one molecule of acetyl-CoA, two pairs of electrons, and four protons (from 2 and 3 carbons) are removed from the long-chain acyl-CoA, shortening it by two carbon atoms. Because 8-2 saturated acid contains no H atoms in its B-carbon (3 carbon) and 8-2 unsaturated acid contains only one H atom, B- oxidation of these two acids cannot occur because the proton deficiency prevents the proton/electron 16 Subject to Copyright. Do not reference. Not for further reproduction shuffling that is essential for tfae completion of the reactions. Also, since the B-carbon of 8-2 Saturated acid is already highly oxidized with two fluorine atoms attached, it is unlikely that any direct oxidation of this carbon can occur under aerobic conditions. On the other hand, the B-oxidation may occur via the newly discovered metabolite, 2H, 2H, 3H, 3H-perfluorodecanoic acid. The third possible transformation pathway may be related to the formation of unidentified transformation products. Since each of the unidentified transformation products in general contributed less than 1% of the mass balance, it is extremely difficult to identify their molecular structures. However, formation of these transformation products may implicate some new pathways not yet understood. Compared with perfluorinated acids, the ethanol spacer (-CHzCHzOH) of CF3(CF2)614CF2CH2CH20Hmakes this molecule more flexible than a perfluorinated acid and could enhance its molecular accessibility to an active site of a given enzyme. This, in turn, could lead to the formation of certain unique metabolites that would not normally occur when a folly fluorinated acid is being used as, a starting material for microbial biodegradation. ..In mixed;i;bacterial cultures, CF3(CF2)6CF2CH2CH20H can be converted to perfluorohexanoic acid [CF3(CF2)4COOH] with about 0.4% of the mass balance (15), implying a mechanism for defluorination from the perfluorinated portion of C-F bonds. When CF3(CF2)614CF2CH2CH20Hwas incubated with a high concentration (330mL sludge L'1 medium) of activated sludge from the same source as for this study, "COi release contributed to about 0.4% of the mass balance, indicating a decarboxylation mechanism from the fluorinated B- carbon. In microbial enrichment culture (19), all the metabolites (8-2 saturated and 8-2 unsatiirated acids plus PFOA) together accounted for 55% of total mass balance at day 81. The portion of the missing part (45%) may include the metabolites that resulted from the degradation of 8-2 unsaturated acid by some other unknown mechanisms. Also, the PFOA level reduced significantly at day 81 (3% of total mass) compared with at day 52 (approx. 6% of total mass) (19). Does this indicate that PFOA may have been farther transformed by the microbial culture? All above evidences so far raise the possibility that alternative pathways may be available for degradation of 8-2 TBA beyond PFOA. Therefore, multiple pathways and enzymes, such as B-oxidation of 2H, 2H, 3H, 3H-perfluorodecanoic acid, HF 17 Subject to Copyright. Do not reference. Not for further reproduction elimination and monooxygenase-catalyzed oxidation of 8-2 saturated and 8-2 unsaturated acids, and other unknown mechanisms, may determine the biological fate of 8-2 Telomer B Alcohol under the study conditions. This report indicates that additional work needs to be done to clarify the prevalence and availability ofthe pathways identified and their products in a range ofmicrobial cultures, conditions and time frames to truly define the ultimate, stable, microbial biodegradation products. Acknowledgements We thank Drs. Theodore H. Carski, Robert A. Hoke, Mary A. Kaiser, S. Mark Kennedy, and Watze deWolf for consultation and manuscript preparation and Kirn Brebner, Keith Prickett, Richard Rossi, and Scott Swain for quality assurance and technical assistance. This research was partially funded by Telomer Research Program, with member companies include Asahi Glass Co., Ltd. (Japan), Clariant GmbH (Germany), Daikin Industries, Ltd. (Japan), and DuPont (USA). 18 Subject to Copyright. Do not reference. Not for further reproduction TABLE 1 Accurate Mass Measurements of Observed Transformation Products determined mass 478.9803 458.9764 414.9771 443.0037 "C-labeled transformation product ion elemental composition CF3(CF2)614CF;lCH2COO(P- eak 2) '^"CHaOaFn CF3(CF2)614CF=CHCOO-(Peak3) CFaCC^^COCT (Peak 1) CF3(CF2)614CH2CH2COO-(Peak4) '^^CI^Fie ^"COiFis '^"CH^Fis calculated mass 478.9816 458.9753 414.9691 443.0004 error in ppm -2.6 +2.3 +19.3 +7.4 19 Subject to Copyright. Do not reference. Not for further reproduction TABLE 2 Mass Balance of ^C-labeled 8-2 TBA and Quantified Transformation Products Compound 2. Test vessels First Acetonitrile Extraction CF3(CF2)6l4CF2CH2COC(P^eak 2) CF3(CF2)6I4CF=CHCOO~(Peak 3) CF3(CF2)614COO-(Peak1) CP3(CF2)614CH2CH2COO(P- eak 4) Unidentified transformation products CF3(CF2)614CF2CH2CH20(HParent; Peak 5) Second Acetonitrile Extraction CF3(CF2)614CF2CH2CH20H(Adsorbed to septa) CF3(CF2)614CF2CH2CH20H(Adsorbed to glass) Sum (Total label) Sum ofCF3(CF2)614CF2CH2CH20(HParent) DayO ND" ND" ND" ND1' ND" 99 0.3 0.3 0.2 0.7 0.3 100 100 Day 7 Day 14 % of total initial mass a Day 28 20.7 2,4 3.0 1,1 0.8 0.2 0.8 0.1 3.9 0.8 47.4 4.9 24.8 1.7 4.6 0.7 1.1 0.1 1.6 0.1 6.0 0.4 27.0 1.8 26.5 3.6 5.9 0.2 2.1 0.4 2.3 0.2 6.9 0.6 16.0 2.6 23.5 3.1 0.8 0.3 101 72 34.30 0.6 0.1 100 c 62 40.8C 0,5 0.0 100 c 57 II. Abiotic control First Acetonitrile Extraction CF3(CF2)6l4CF2CH2CH20H Second Acetonitrile Extraction CF3(CF2)6l4CF2CH2CH20H(Adsorbed to septa) CF3(CF2)6l4CF2CH2CH20H(Adsorbed to glass) Sum (Total Label) 99 0.3 0.1 0.0 0.5 0.3 100 74.7 1.9 27.0 1.9 0.5 0.2 102 61.1 4.0 49.2 3.6 37.5 3.3 0.3 0.1 99 51.4' 0.4 0.1 100c a The mass balance was calculated based on the ^C counts for the parent or individual transformation products at each sampling time points versus the ^C counts applied initially at day 0. All values represent mean SD (standard deviation) for n == 4 samples unless otherwise stated. b Not detected. c Because the original septa after the first acetonitnie extraction were not saved for re-extraction (second extraction), no direct ^C counts data are available. However, available ^C counts data indicated a virtually 100% mass balance for day 7 test vessels and abiotic samples and day 14 abiotic samples. Thus a 100% mass balance was reasonably assumed for the rest of samples and1 C counts in these septa were calculated based on a 100% mass balance. 20 Subject to Copyright. Do not reference. Not for further reproduction Figure Captions FIGURE 1, LC/ARC chromatograms of the parent, CF3(CF2)614CF2CH2CH20H(peak 5) and "Clabeled transformation products (peak 1-4; peak 1: CF3(CF2)614COO~, peak 2: CF3(CF2)6I4CP2CH2COO-p, eak 3: CF3(CF2)614CF:-CHCOO-,and peak 4: CF3(CF2)614CH2CH2COCr) at different sampling time points using LC/ARC method 1 (A, left column) and method 2 (B, right column). FIGURE 2. Daughter ion spectra ofdeprotonated molecular ions: m/z 415 (A), 479 (B), 459 (C), and 443 (D) obtained for transformation products observed in a day 28 sample. FIGURE 3. Proposed biotransformation pathways of "C-labeled 8-2 Telomer B Alcohol. The solid arrows indicate proposed transformation steps. The dotted arrows indicate potential transformation steps which may or may not be occurring. 21 Subject to Copyright. Do not reference. Not for further reproduction A. LC/ARC Method 1 urns Sslftf E333MEI_13D , Kl 3,11JJH. C!al:ffiC(CPM) 5 DayO xl03SfiaeDaaEr-aD .\bfcghjJH.agTd:ffic(cffl> B. LC/ARC Method 2 mz^iTht&^SOm .VdcWf.WWffKVW Day 7 2 5 \ 4 3 /V^A^V^-h^-^A-f^^/"J xlQZan]foB-G3KA ,UAGmJ1.Cfamd:jnc(CMl KiraaroH-EBBCTjm .it28>iJi.aiini:RqCTH Day 28 -Abiotic Control Mn 001 5 tin nm Figure 1 22 Subject to Copyright. Do not reference. Not for further reproduction 100 A 169.01 394.98 c 36B99 V-W--------4S----------300------SOQ----------SO------m--te--SO----------T5O0 "' ^ (LC/ARC Peak 1 - 415 m/z; PareBt: Ca'^(CF^)(14COO-) (LC/ARC Peak 3 - inte :"!0 " 5W TM 459 ra/z; Parent: CF3(CF,)6"CF=CHCOO" B D SO ino 200 300 400 500 mil 600 TOO (LC/ARC PeaK - 479 mA; Parent: CF3(CF2)61'tCF2CH2COO- ) SO W rnlz. 200 3011 400 500 600 700 LC/ARC Peak 4-443 m/z; Parent: CF^CF^CHiCHiCOcr) Figure 2 23 Subject to Copyright. Do not reference. Not for further reproduction CF3(CF2)6 14CFzCHzCHiOH CF3(CF2)614,CFzCHzCOO-O) - ?- - ^ CFaCT^C^CHzCOO- (m) |B-oxidation CF3(CF2)6MCF=CHCOO~ (II) - ?--^ CF3(CF2)6MCOO- CF3(CF2)614CF2COO-(nootbserved) Figure 3 24 Subject to Copyright. Do not reference. Not for further reproduction Literature Cited (1) Moody, C. A., Field, J. A. Perfluorinated surfactants and the environmental implications of menuse in fire-fighting foams. Environ. Sci. Technol. 2000, 34,3864-3869. (2) 3 M Company. Fluorochemical use, distribution and release overview. US EPA Public Docket AR-226-0550; St. Paul, MN, 2003. (Not peer reviewed) (3) Hekster, F. M.; Remi, W. P. M.; de Voogt, P. Environmental and toxicity effects of perfluorinated substances. Rev. Environ. Contain, and Toxicol. 2003, 779,99-121. (4) Schullz, M. M.; Barofsky, D. P.; Field, J. A. Quantitative determination of fluorotelomer sulfonates in ground water by LC MS/MS. Environ. Eng. Sci. 2003,20,487-501. (5) Dinutrov, S.; Kamenska, J. D.; Walker, W. W.; Purdy, R.; Lewis, M.; Mekenyan, 0- Predicting the biodegradation products of perfluorinated chemicals using CATABOL. SAR and QSAR in Environ. Res. 2004, 15, 69-82. (6) Stock, N. L.; Lau, F. K.; Ellis, D. A.; Martin, J. W.; Muir, D. C. G.; Mabury, S. A. Polyfluorinated telomer alcohols and sulfonamides in the North American troposphere. Environ. Sci. Technol. 2004. 38,991-996. (7) Smart, B. E. In Organoflwrine Chemistry; Charateristics ofC-F Systems; Banks, R. E., Smart, B. E., Tatlow, J. C. Eds.; Plenum Publishing: NewYork, 1994. (8) Baker, B. E.; Rao, N. S. In Organofluorme Chemistry; Textile Finishes and Fluorosurfactants; Banks, R. E., Smart, B. E., Tatlow, J. C. Eds.; Plenum Publishing: NewYork, 1994. (9) Rissa, E. Fluorinated Surfactants and Repellents; Marcel Dekker: New York, 2001. (10) Taylor, C. K. In Design and Selection of Performance Surfactants; Fluorinated Surfactants in Practice; Karsa, D. R. Eds; CRC Press: New York, 1999. 25 Subject to Copyright. Do not reference. Not for further reproduction (11) Hagen, D. F.; Belisle, J.; Johson, J. D.; Venkateswariu, P. Characterization of fliiorinated metabolites by a gas chromatographic-helium microwave plasma detector-The biotransfonnation of lH,lH,2H,2H-perfluorodecanol to perfluorooctanoate. Awl. Biochem. 1981,118, 336-343. (12) Remde, A.; Debus, R. Biodegradability of fluorinated surfactants under aerobic and anaerobic conditions. Chemosphere. 1998. 32,1563-1574. (13) Key, B. D.; Howell, R. D.; Criddle, C. S. Defluormation oforganofluorine sulfur compounds by Pseudomonas Sp. strain D2. Environ. Sci. Technol. 1998.32,2283-2287. (14) Schroder, H. F- J. Chromatography A. Determination of fInQrinated surfactants and their metabolites in sewage sludge samples by liquid chromatography with mass spectrometry and tandem mass spectrometry after pressurized liquid extraction and separation on fluorine- modified reversed-phasesorbents. 2003. J020,131-151. ".;,. ,./ (15) E.I. du Pont de Nemours and Company. Accelerated biodegradation of 8-2 Telomer B Alcohol - A preliminary screening study. US EPA Public Docket AR226-1264; Wilmington, DE, 2003. (Not peer reviewed) (16) 3 M Company. Biodegradation screen study for telomer-type alcohols. US EPA Public Docket AR-226-1149; St. Paul, MN, 2003. (Not peer reviewed) (17) E.I. du Font de Nemours and Company. Telomer B Alcohol, oral gavage range-finding study in rats. US EPA Public Docket AR226-1339; Wilmington, DE, 2001. (Not peer reviewed) (18) 3 M Company. Exploratory 28-day oral toxicity study. US EPA Public Docket AR-226-0038; St. Paul, MN, 2000. (Not peer reviewed) 26 Subject to Copyright. Do not reference. Not for further reproduction (19) Dinglasan, M. J. A.; Ye, Y.; Edwards, E. A.; Mabury, S. A. Fluorotelomer' alcohol biodegradation yields poly- and perfluorinated acids. Environ. Sci. Technol. 2004.35, 2857- 2864, (20) Kaiser, M, A.; Cobranchi, D. P.; Kao, C.-P.; Krusic, P. J.; Marchione, A. A.; Richardson, R. E.; Buck, R. C. Physicochenucal properties of 8-2 fluorinated Telomer B Alcohol. J. Chem. Eng. Data. 2004, 49(4), 912-916. (21) Szostek, B.; Prickett K.B.; Maslanka, J. C.; Kennedy, S, M. Development of analytical methodology for determination of Telomer B Alcohols, telomer-derived anionic, nonionic fluorosurfactants, and polymers. SETAC 24th North American Annual Meeting Abstract 527. Austin, TX, 2003. (22) OECD. Guideline for Testing of Chemicals, Section 3: Ready Biodegradability, 301D,,.Closed Bottle Test. 1992. (23) Szostek, B,; Prickett, K. B. Determination of 8-2 Telomer B Alcohol in animal plasma'and- tissues by gas chromatography/mass spectrometry. J. Chromatography B, manuscript submitted. (24) Nassar, A.-E. F.; Bjorge, S. B. On-line liquid chromatography-accurate radioisotope counting coupled with a radioactivity detector and mass spectrometer for metabolite identification in drug discovery and development. Anal. Chem. 2003, 75,785-790. (25) Bartha, R.; Yabannavar A. V. In Ecological Assessment of Polymers; Biodegradation Testing of Polymers in Soil; Hamilton, J. D., Suteliffe, R. Eds.; Van Nostrand Reinhold: New York, 1997. (26) Casteels, M.; Foulon, V.; Mannaerts, G. P.; Van Veldhoven, P. P. Alpha-oxidation of3-methylsubstituted fatty acids and its thiamine dependence. Eur. J, Biochem. 2003, 270,1619-1627. 27 Subject to Copyright. Do not reference. Not for further reproduction (27) DePierre, J. W. In Organofluorines; Effects on Rodents of Perfluorofatty Acids; Neilson, A. H. Eds.; Springer-Verlag: New York, 2002. (28) Alexander, M. Biodegradation and Bioremediation; 2nd Edition; Academic Press: New York, 1999. (29) Nelson, D. L.; Cox, M. M. Lehninger Principles of Biochemistry; 3rd Edition; Worth Publishers: New York, 2000. 28 Subject to Copyright. Do not reference. Not for further reproduction