BiodegradationYields Poly- and
M A R Y J O Y C E
E L I Z A B E T H
S C O T T
Department of Chemistry, University of Toronto,
80 Saint George Street, Toronto, Ontario, Canada M5S 3H6,
and Department of Chemical Engineering and Applied
Chemistry, University of Toronto, 200 College Street,
Toronto, Ontario, Canada M5S 3E5
A . D I N G L A S A N ,²
A . E D W A R D S ,³
A . M A B U R Y *, ²
Y U N Y E ,²
A N D
The widespread detection of environmentally persistent
(PFOA) and its longer chained homologues (C9>C15) in
biota has instigated a need to identify potential sources.
It has recently been suggested that fluorinated telomer
alcohols (FTOHs) are probable precursor compounds that
mayundergotransformationreactions inthe environment
leading to the formation of these potentially toxic and
bioaccumulative PFCAs. This study examined the aerobic
CH2CH2OH) using a mixed microbial system. The initial
measured half-life of the 8:2 FTOH was ∼0.2 days mg-1of
initial biomass protein. The degradation of the telomer
with an electron capture detector (GC/ECD). Volatile
metabolites were identified using gas chromatography/
mass spectrometry (GC/MS), and nonvolatile metabolites
were identifiedandquantifiedusing liquidchromatography/
tandemmass spectrometry (LC/MS/MS). Telomer acids
(CF3(CF2)7CH2COOH; CF3(CF2)6CFCHCOOH) and PFOA were
identified as metabolites during the degradation, the
unsaturatedtelomer acidbeing the predominant metabolite
measured. The overall mechanisminvolves the oxidation
of the 8:2FTOHtothe telomeracidvia the transienttelomer
aldehyde. The telomer acid via a ?-oxidation mechanism
ultimately producing the highly stable PFOA. Telomer
alcohols were demonstrated to be potential sources of
PFCAs as a consequence of biotic degradation. Biological
transformation may be a major degradation pathway for
fluorinated telomer alcohols in aquatic systems.
The extensive use of perfluorinated organic compounds, in
both commercial and industrial applications, has recently
prompted research into the disposition, fate, persistence,
Their widespread application is attributed to the unique
properties that the perfluoroalkyl chain imparts upon the
compound. Many of these compounds have been found to
be highly stable in the environment due to the strength of
the carbon-fluorine bond (1).
Extensive biological monitoring studies in recent years
have revealed widespread global distribution of perfluori-
nated acids such as perfluoroalkane sulfonate, perfluorooc-
tane sulfonate (PFOS), and perfluorinated carboxylic acids
(PFCAs) of which perfluorooctanoic acid (PFOA) and per-
fluorodecanoic acid (PFDA) are examples (2-5). Some long
chained homologues of PFCAs were first reported in fish
samples collected from a creek after a large spill of aqueous
film formingfoam (AFFF) (6);subsequentmonitoringoffish
from this and nearby creeks suggests that the PFCAs, other
than PFOA, did not arise from the spill (7). More recently,
biota samples collected from the Canadian arctic (8) were
shown to contain the full suite of PFCAs (C9-C15). PFOA
has also been detected in trace concentrations from human
serum samples worldwide (9). Long chain perfluorinated
acids have been found to be environmentally persistent,
bioaccumulative (10, 11), and potentially toxic (12, 13).
Perfluorinated acids are stronger acids as compared to
their hydrocarbon counterparts and the correspondingly
lower pKa(i.e., PFOA is 2.80) (14) results in the dominance
of the anionic form with little propensity to escape via
volatilization.Toexplain theoccurrenceofPFCAsin remote
neutral compounds might serveas atmospheric precursors.
These would undergo environmental decomposition either
biotically, or abiotically, to the more persistent acids (15).
Fluorotelomer alcohols (FTOHs) are polyfluorinated
compounds typically characterized by even numbered per-
to a hydroxyl group. FTOHs are typically used as precursor
compounds in the production of fluorinated polymers used
as those of PFOS-based products (14). They are also used in
themanufactureof a widerangeof products such as paints,
adhesives, waxes, polishes, metals, electronics, and caulks
(14). During the years 2000-2002, an estimated 5 × 106kg
year-1of these compounds was produced worldwide, 40%
of which was in North America (15). Their name is derived
from the telomerization process from which they are
produced. FTOHs are given nomenclature based upon the
numberofperfluorinated carbonsin relation tothenumber
of hydrogenated carbons they possess (i.e., 8:2FTOH; Table
1). Measured vapor pressures of FTOHs range from 140 to
990 Pa (16). The calculated dimensionless Henry's law
pressure reveals the propensity of these compounds to
partition into air. This is supported by a recent air sampling
campaign in which FTOHs were detected at tropospheric
concentrations typically ranging from 17 to 135 pg m-3(17,
18) with urban locations apparently having higher concen-
trationsthan rural areas.A study by Ellisetal.showsthatthe
atmospheric lifetime of short chain FTOHs as determined
by its reaction with OH radicals is approximately 20 days
(15). These results demonstrate that fluorotelomer alcohols
arewidely disseminated in thetroposphereand arecapable
of long-range atmospheric transport. Sources of these
compoundsarecurrently unknown, although it islikely that
they may be released from the decomposition of polymeric
or nonpolymeric materials that incorporate FTOHs or from
themselves that failed to be covalently linked to polymers
during production (15). If polyfluorinated polymers are
indeed a source for these compounds, then a potential fate
*Corresponding author e-mail:
²Department of Chemistry.
³Department of Chemical Engineering and Applied Chemistry.
Environ. Sci. Technol. 2004, 38, 2857-2864
10.1021/es0350177 CCC: $27.50
Published on Web 04/16/2004
2004 American Chemical SocietyVOL. 38, NO. 10, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY92857
of these materials is to end up in an aqueous environment
such as carpet or upholstery cleaning. This type of an
environment would subject these polymers to potential
microbial degradation, possibly releasing FTOHs to the
aqueous systems where they too are subjected to biodeg-
An earlier study by Hagen et al. (19) showed compelling
evidence for FTOH biotransformation. They identified 2H,-
2H-perfluorodecanoic acid (8:2FTCA, Table1) and PFOA as
metabolites in rats given a singledoseof 8:2FTOH using19F
NMR and a gas chromatograph equipped with a microwave
plasma detector (GC/MPD) (19). They have also suggested
the unsaturated form of the acid (8:2 FTUCA, Table 1) as
another metabolite using retention time matching of a
synthesized standard by gas chromatography using an
electron capture detector (GC/ECD). A ?-oxidation mech-
Thisrat metabolism study suggeststhat FTOHssubjected to
acids. It has also been reported that a mixture of fluoro-
chemical telomeralcoholswerebiodegraded when exposed
perfluorinated acids after a 16 day incubation period and
proposes that ?- and R-oxidation mechanisms may be
involved in the degradation pathway. Transient species
detected by this study using LC/MS/MS included the
C9, C10, and C12).
alcohols could undergo biodegradation under aerobic con-
ditions using 8:2 FTOH (Table 1) as a model telomer in a
hypothesized that fluorotelomeralcoholscan beoxidized to
the corresponding aldehyde (Table 1), subsequently to the
related acids, and ultimately leading to the production of
et al. in rats (19) and that of the screening study by Lange
(20). To test this hypothesis, we developed a method to
measure 8:2 FTOH via headspace using solid-phase mi-
croextraction (SPME) coupled with GC/ECD. Methodswere
developed to identify volatile metabolites using gas chro-
matography/mass spectrometry (GC/MS) and liquid chro-
nonvolatile metabolites. Unlike the use of gas chromatog-
raphy for the analysis of PFCAs and telomer acids, applying
the technique of LC/MS/MS involves little sampling prepa-
ration and eliminates the need for derivatization. It also
provides enhanced confidence in chemical identification
from MS/MS spectra.
Experim ental Procedures
Media and Chemicals. The enrichment culture used in the
experiments was routinely grown in a defined mineral
medium which contained the following constituents added
to distilled and deionized water to make one liter: 65 mL of
phosphate buffer (27.2 g of KH2PO4and 38.4 g of K2HPO4
L-1), 10 mL of salt solution (53.5 g of NH4Cl, 7.0 g of CaCl2‚6
H2O, 2.0gofFeCl2‚4H2O L-1), 2mL oftracemineral solution
(0.3 g of H3BO3,0.1 g of ZnCl2, 0.1 g of Na2MoO4‚2 H2O, 0.75
g of NiCl2‚6 H2O, 1.0 g of MnCl2‚4 H2O, 0.1 g of CuCl2‚2 H2O,
1.5 g of CoCl2‚6 H2O, 0.02 g of Na2SeO3, 0.1 g of Al2(SO4)3‚18
H2O, and 1 mL of concentrated H2SO4L-1), 2 mL of MgCl2‚6
H2O solution (48.8 g L-1), and 10 mg of yeast extract. The
mixture was autoclaved for 20 min at a temperature of 120
°C andpressureof18psi.ThepH wassubsequentlyadjusted
to approximately 7 by the addition of 1N HCl.
The 8:2 FTOH (97%) was purchased from Oakwood
Research Chemicals (West Columbia, SC). The 8:2 FTOH
telomer aldehyde (8:2 FTAL) was synthesized as described
by Napoli et al. (21), and the 8:2 FTOH telomer acid (8:2
FTCA) and the8:2FTOH unsaturated acid (8:2FTUCA) were
synthesized as described by Achilefu et al. (22). Character-
ization of these synthesized standards was done using13C,
19F, and1H NMR alongwith high-resolution electron impact
mass spectrometry, negative chemical ionization, and posi-
FTAL, 8:2FTCA, and 8:2FTUCA was>95%. PFOA (96%) and
mercuric chloride were purchased from Aldrich Chemical
Co. (Milwaukee, WI).
using theBradford method, using a microassay kit (Bio-Rad
as a standard.
Growth Conditions and Culture Preparation for Deg-
site and had been enriched on 1,2-dichloroethane and
subsequently maintained using ethanol as the sole carbon
source (23). This mixed culture was chosen because it was
and therefore may also be active on fluorinated alcohols.
Cellswereharvestedby centrifugation, washedwith defined
mineral medium, and resuspended in 2% of the total liquid
volume used in the experiments. Degradation experiments
were performed in triplicate using 1 L glass vessels (Pyrex)
filled with 950 mL of defined mineral medium and sealed
with mininert caps. The 8:2 FTOH was added to the culture
vessels by adding 14 µL of a concentrated stock solution (50
µg µL-1) made up in ethanol to attain a target aqueous
concentration of 50 µg L-1. The vessels were allowed to
similarly except that 500mg ofmercuric chloridewasadded
to inhibit microbial activity. All cultures were stored in the
dark at room temperatureon a shaker at 95rpm to allow for
continued mixing and to enhance mass transfer of oxygen
from the headspace to the liquid phase. For degradation
sludge obtained from Ashbridges Bay Treatment Plant
(Toronto, ON) was used as the inoculum and prepared as
described previously without acclimation to ethanol.
GC/ECD and GC/MS Analysis of 8:2 FTOH and Volatile
Transformation Products.Thedegradation ofthe8:2FTOH
was monitored using solid-phase microextraction (SPME).
A 30 µm fiber with poly(dimethylsiloxane) (PDMS) coating
(Supelco, Bellefonte, PA) was exposed to the headspace of
TABLE1. Acronym , Structure, andM olecular Weight of PerfluorinatedCom pounds of Interest
compoundacronym structure molecular wt. (amu)
8:2 fluorotelomer alcohol
8:2 fluorotelomer aldehyde
8:2 fluorotelomer acid
8:2 fluorotelomer unsaturated acid
allylic 8:2 fluorotelomer alcohol
allylic 8:2 FTOH
28589ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 10, 2004
the sealed culture vessels and was allowed to equilibrate for
5min. Headspaceanalysiswasused to addressthevolatility
pressure of the 8:2 FTOH at 212 Pa (16), and the water
solubility wasmeasured to be148µg L-1(24) from which we
calculated a Henry's law constant value of 270. Aqueous
concentrations of the 8:2 FTOH were then determined, to
ensure that the system was below the water solubility, from
the following relationship:
where Mtot is the total mass of the compound, CL is the
aqueous concentration, H is the Henry's law constant, and
Vgistheheadspacevolume; typical aqueousconcentrations
were ∼50 µg L-1.
Analysis was done using a Hewlett-Packard 5890 Series
II gas chromatograph equipped with an electron capture
detector(AgilentTechnologies, Wilmington, DE) and a30m
× 0.5 mm × 250 µm DB-35 column (Phenomenex, Missis-
sauga, ON). The injector temperature was 250 °C, and the
wasinitially held at45°C for2min followed by a10°C min-1
ramp to 95 °C and held for 5 min and a final ramp of 30 °C
min-1to 250 °C. The carrier gas was hydrogen at a pressure
were used for calibration. Standards used had aqueous
concentrations ranging from 2 to 55 µg L-1, and response
was linear with r2typically >0.99.
A Hewlett-Packard 6890 gas chromatograph coupled to
DE) was used under full scan positive chemical ionization
mode to identify the volatile metabolites observed in the
degradation. The carrier gas was helium, and methane was
used as the ionizing gas at a flow rate of 1 mL min-1. The
sourcetemperaturewas250°C, and theelectron energy was
at 100 eV. Gas chromatographic separation was performed
using a DB-Wax column (30 m × 0.25 mm × 250 µm) (J&W
Scientific, Folsom, CA). Theinitial oven temperaturewas 45
°C for 5 min and ramped at 15 °C min-1to 210 °C. Pulsed
1µL splitlessinjectionswereperformedatan initialpressure
of 25 psi and 220 °C, returning to 10 psi at 1.2 min, and
followed by an injector purge.
LC/MS/MS Analysis of NonvolatileMetabolites. Priorto
vessel and centrifuged to remove biomass. One mL of
methanol and 1 mL of supernatant were then transferred to
and identification of target nonvolatile metabolites were
performed using a Waters 717 autosampler along with an
C18 column (5 µm, 4.6 × 250 mm) at a flow rate of 400 µL
min-1. Gradient elution was not applied because of con-
tamination problems for PFOA with the available gradient
pump. Isocratic elution proved to bean adequateand faster
alternative for the analysis using a mobile phase comprised
of 70% Optima grade methanol and 30% 18MΩ deionized
water. Samples were injected at a volume of 20 µL, and the
HPLC column eluate entered the mass spectrometer ion
than 10 min.
Acquisition of the mass spectra was carried out using a
Micromass Quattro micro Triple Quadrupole Mass Spec-
trometer (Micromass; Manchester, UK) operated under
negative electrospray ionization mode. A standard with a
the equipped syringe pump at a flow rate of 30 µL min-1for
positioning of the ion sprayer and tuning of the mass
spectrometer. The capillary voltage was 2.9 kV, while the
cone voltage was set at 15 V. Specific operating parameters
are listed in Table 2. The source block and desolvation
time was 0.5 s. The nebulizer and desolvation gas flow rates
were 20 and 260 L h-1, respectively. For tandem mass
spectrometric analysis, argon was used as the collision gas
(2.86 × 10-3mbar), and collision energies (Table 2) were
varied to optimize for the sensitivity of each compound.
Quantification was achieved under multiple reaction
monitoring (MRM) mode and by calibrating the primary
matrix-matched calibration curves were generated daily by
using freshly prepared standards. Standard concentrations
ranged from 100 to 1000 µg L-1, and r2was typically >0.99.
Cone voltages and collision energies were optimized by
standards made up in methanol for each individual com-
pound to ensure the sensitivity of MS/MS analysis. Late in
the study, when 8:2 FTUCA concentrations were declining,
suppression of the PFOA signal, was revealed. Standard
additions were performed for PFOA at the final time point
of theexperiment (day 81) so that an accuratemassbalance
was determined; no suppression was observed for the other
analytes of interest during the investigation. Along with the
hypothesized metabolites, other prospective products such
as perfluorononanoic acid (PFNA) and perfluoroheptanoic
acid (PFHpA) were also monitored although neither were
detected in any samples.
QA/QC ofLC/MS/MS Analysis.Toguaranteedataquality,
a reagent blank (methanol) was injected after each time-
point sample group (four samples/group) to reveal any
problems of carryover. A typical chromatogram is shown in
for 8:2 FTCA and PFOA to confirm their identity and only
one for 8:2 FTUCA since no further fragmentation was
observedforthiscompound.Each ion wasmonitoredunder
its own optimal condition listed in Table 2. The ratio of the
ion was calculated and compared between samples and
the repeatability of the product-ion ratios obtained in the
injections of both the standards and the samples (Figure 2);
the difference between samples and standards in this study
TABLE2. O ptim izedM S/M S Conditions for M etabolite Confirm ation
compound parent ion(m/z)cone vol. (V) daughter ion(m/z)collisionenergy (eV)
8:2 FTCA 47715 393
8:2 FTUCA45715393 15
Mtot) CL(VL+ H × Vg) (1)
VOL. 38, NO. 10, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 92859
Despite the rising number of published studies looking at
the detection of perfluorinated acids in the environment,
only a small numberhaveexamined theirpotential sources.
Ellis et al. have identified thermolysis of fluoropolymers as
an abiotic mechanism that can potentially lead to the
et al. have been the first to observe the production of PFOA
from the 8:2 telomer alcohol in a biotic system (19), and
more recently, the production of perfluorocarboxylic acids
were observed from a telomer alcohol biodegradation
screeningstudy by Lange(20).Thestudy presented herenot
only provides further evidence that telomer alcohols are a
potential source of PFCAs through biotransformation reac-
tions, but it also presents a plausible biotic mechanism in
a microbial system.
In this laboratory study, an initial mass of 750 µg (1.5
µmol) of8:2FTOH wasaddedtovessels(1L) inoculatedwith
microorganisms. As seen in Figure 3, the 8:2 FTOH spiked
was85% degraded asofday 7and wasbelow detection limit
levels (2 µg/L) by day 16. Triplicate vessels showed similar
of initial biomass protein followed by a second half-life of
FIGURE1. (a)Typical GC/ECDchromatogramshowing the8:2FTOHandtheimpurityallylic 8:2FTOH.(b)Typical LC/MS/MS chromatogram
of the acid metabolites.
FIGURE 2. Confirmationofmetabolites observedfromthe degrada-
tion of 8:2 FTOH. Comparison of transition parent-daughter m/z
ratios in samples and standards. Base peak set at 100%.
28609ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 10, 2004
∼0.8 days mg-1. The complex kinetics observed in the
degradation of 8:2 FTOH maybe attributed to a couple of
probable explanations. First, there may have been a change
in the activity of specific microbial populations comprising
the observed change in the rate of the degradation after day
3. Second, more than one mechanism of degradation may
be operative, resulting in varying rates of reaction involved;
this study, however, could only provide evidence for one
Thedegradation of the8:2 FTOH was presumably dueto
microbial activitysincethesterilecontrol showedlittletono
transformation during the experimental period as shown in
Figure 3 (inset). The small decrease in the concentration
from thevessel during sampling. Thesterilecontrol wasnot
routinely sampled for 8:2 FTOH past day 16 of the study.
was driven by microbial activity in active bottles was the
observed and quantified acidsin theactivebottlesand their
absence in the sterile control (Figure 3). The degradation of
the 8:2 FTOH occurred concurrently with the production of
8:2 FTUCA. This transformation step occurred via the
formation of the telomer aldehyde, 8:2 FTAL. This volatile
data (Figure 4a). The synthesized standard of the 8:2 FTAL
showed a distinctive double peak for its molecular ion m/z
463 (M + 1), the presence of which was also confirmed in
the samples. This observation may be due to the existence
of two different conformations of the compound in the gas
phase. Studies are ongoing to further investigate this
observation. The 8:2 FTOH aldehyde appeared to be a
possible with our current method. The 8:2 FTCA and 8:2
FTUCA were more stable in the system.
Upon the depletion of the 8:2 FTOH in the system, the
in the production of 8:2 FTUCA. There are potentially two
an -HF, or biotically, where perhaps an acyl-coA dehydro-
genase type of enzyme oxidizes the CR-C? bond. This
reaction proceeds via the removal of the R-proton, followed
by hydridetransferofthe?-proton presumably toacofactor
in our laboratory indicate that the abiotic elimination of HF
from FTCAs are slower (half-lives > 1 week) than observed
in these biological systems, although it is likely that both
pathways were involved as the experiment progressed.
The production of the unsaturated acid can also be
attributed to thedegradation oftheallylic 8:2FTOH present
as an impurity in the 8:2 FTOH alcohol (Figure 1a). Mass
spectral data from previous studies within the group have
present since the purity of the 8:2 FTOH used in this study
wasof97%(27),andwecan assumethattheallylic 8:2FTOH
comprised at most 3% of the total mass of FTOH initially.
This impurity would likely be metabolized in analogous
fashion as the saturated alcohol, presumably forming the
of the 8:2 FTUCA early in the experiment (day 2-5) at
approximately the same time as the detection of 8:2 FTCA
may be a consequence of the oxidation of the allylic form of
the 8:2 FTOH.
PFOA was detected in the system at very low concentra-
tions beginning at day 16. A sample chromatogram is
presented in Figure4ofall nonvolatilemetabolitesdetected
in samples and their absence in the blank and in the sterile
control. It appears that the highest concentrations of PFOA
occurredatthepeak concentration ofthe8:2FTUCA.Byday
degradation of the 8:2 FTUCA in the system may lead to an
studies are looking at degradation products when only the
8:2 FTUCA is spiked in a microbial system. Hagen et al. (19)
in their earlier study identified 8:2 FTUCA in their system
but were unable to definitively show that this was the initial
Lange et al. (18) reported the detection of 6-7% of PFOA at
the conclusion of their biodegradation study of telomer
FIGURE3. Typical transformationkineticsandmassbalanceofmetabolitesobservedindegradationexperiments;degradationof8:2FTOH
inactive microcosm(vessel B). Loss of8:2FTOH, productionof8:2FTCA, 8:2FTUCA, PFOA, andoverall mass balance. Noobservable loss
of 8:2 FTOH in sterile control (n ) 1) (inset). PFOA values were obtained using standard additions at day 81.
VOL. 38, NO. 10, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 92861
alcohol mixtures and have also identified the unsaturated
acid as an intermediate in the degradation process.
An assessment of the mass balance between the parent
compound and the nonvolatile metabolites resulted in
approximately 55% of products accounted for at the conclu-
sion of the study (Figure 3), with 8:2 FTUCA being the most
abundant metabolite along with PFOA. During the time
interval,day 1-5,therewasalsoanoticeablelossin thetotal
produced early on in the pathway was unaccounted. This
apparentlossin massmay beduetoourinability toquantify
the 8:2 FTAL, along with other observed but unidentified
ofthedegradation in experimentalbottlesandtheirabsence
in thesterilecontrol. By day 81, theobserved 45% lossofthe
products may be due to a number of reasons. As previously
alluded to, other volatile metabolites observed in the
degradation that were left unidentified may account for
partial loss in measured products, as well as that volatile
metabolites may have been lost during routine sampling. It
is also possible the unaccounted mass could arise from the
aldehyde) being covalently bound by biological macromol-
ecules such as extrapolymeric substances (EPS) produced
extensively by most bacteria leading to its perceived loss.
The unsaturated metabolites are presumably quite electro-
philic and hence susceptible to attack by endogenous
FIGURE 4. (a) GC/MS (PCI) extractedchromatograms of synthesized8:2FTAL standardand8:2FTAL detectedinsamples inoculatedwith
sludge obtained froma sewage treatment plant. Distinctive double peak for m/z 463(M + 1) in standard was also detected in samples.
(b) LC/MS/MS chromatogramof sample taken froma blank, sterile control, and an active bottle. 8:2 FTCA, 8:2 FTUCA, and PFOA were
detected in the active sample and were absent in the blank and sterile control.
28629ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 10, 2004
nucleophilespresent in biological systems. Theobservation
that the unsaturated acid was the dominant metabolite
produced from telomer alcohol biodegradation may be of
significance since it may very well be a toxic metabolite for
Ion chromatography was used in an attempt to quantify
inorganic fluorine (F-) in the system that may account for
asmall partoftheobserved lossin mass.Ifthetelomeracids
and PFOA were the only primary metabolites of telomer
alcohol degradation as observed in this study, then it is
suggested that these compounds fail to undergo extensive
defluorination; hence, expected fluoride concentrations
would be low. The high amounts of chloride present in the
matrix (mineral medium) prevented the detection limits for
fluoride to be any lower than 1 mg L-1. The use of a fluoride
specific electrode was also considered, although a similar
ofdetection forfluoridewastoohigh forthetelomeralcohol
concentration used in this study; the initial spike of 750 µg
of8:2FTOH would haveproduced approximately 60µg/L of
fluoride (1:2 molar ratio) from the hypothesized pathway
acid. The concentration used in this study for dosing the
FTOH was chosen to be well below its saturated water
fluoride. It should be noted that if the FTOH underwent
(>1 mg/L) would have been observable.
A proposed biodegradation scheme (Figure 5) is based
upon results of this laboratory study and built on earlier
results presented by Hagen et al. (19). Under this proposed
pathway, 8:2 FTOH can be oxidized by an alcohol dehydro-
genase enzyme, fairly common in bacteria (28-30), to form
the 8:2 FTAL. Subsequent oxidation of the terminal carbon
dehydrogenase type of enzyme. Murphy et al. reported the
isolation of an aldehyde dehydrogenase enzyme capable of
converting fluoroacetaldehyde to fluoroacetate in Strepto-
myces cattleya (31). Although this study was looking at the
capable of mediating such metabolic transformations. This
in the conversion of ethanol to acetic acid in the absence of
molecular oxygen (32). Dehydrogenation reactions also
require the coenzyme nicotinamide adenine dinucleotide
(NAD+), which servesashydrogen carriers. However, in the
presence of molecular oxygen, typical oxidation reactions
by mixed function oxidases (MFO) or monooxygenase type
of enzymes such as cytochrome P450, also widespread in
microorganisms, animals, and humans (32). Despite per-
forming the biodegradation experiments under aerobic
conditions, the oxygen concentration was not measured;
8:2 FTCA leading to its unsaturated form and ultimately to
proposed by Hagen et al. (19). Several critical enzymes are
possibly involved in such a mechanism. We suggest that
enzymessuch asacyl-CoA synthasesand crotonasesmay be
required. The oxidation step from the unsaturated acid to
PFOA is thermodynamically costly and hence is expected to
thatPFOA wasfirstdetectedin thesystem in thelatterphase
of the experiment (day 16). This suggests that oxidation of
using the activated sludge from a sewage treatment plant
showed that a similar pathway was operable with all the
identified degradation products indicated in Figure 5 ob-
served (data not shown). Theprevious observations suggest
that other telomer alcohols may degrade analogously pro-
ducing their corresponding even perfluorocarboxylic acid
FIGURE 5. Proposed biodegradation pathway and products of 8:2 FTOH based upon laboratory experiments. Structures in brackets are
proposed transitional intermediates and were not determined in this study.
VOL. 38, NO. 10, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 92863
under the proposed mechanism (i.e., 10:2 FTOH may
biodegrade producing the perfluorodecanoic acid, and the
6:2 FTOH would form the perfluorohexanoic acid).
The current study showed that telomer alcohols readily
biodegrade, producing telomer acids and perfluorinated
acids, with the unsaturated telomer acid being the predomi-
nant metabolite; we are currently investigating whether the
Microbial transformation reactions such as demonstrated
by these experiments have strong implications for other
as surrogates for metabolic reactions of higher organisms.
and other carboxylic fluorinated acids detected widely in
biota and as previously demonstated by Hagen et al. (19).
There are likely several pathways under which telomer
although it appears that the ?-oxidation pathway described
previously is a principal fate for these compounds. There
existsthepotential forR-oxidation oftheFTOHsto yield the
odd numbered FTCAs recently detected in biota (8); our
investigation, however, indicated no evidence for this
pathway being operable under these microbial conditions.
Further studies are underway to determine the identity of
other volatile metabolites observed in these experiments.
Hagen et al. (19) were also unable to identify a major
metabolite in their experiments. The identity of these
unknown volatile metabolites may provide further clues to
degradation of telomer alcohols may very well be a primary
TheauthorsacknowledgeDr.David Ellisforhisinput tothis
paper along with Dan Mathers, of the ANALEST facility at
sampling. The Natural Science and Engineering Research
Council of Canada, through a strategic grant, generously
(1) Key, B. D.; Howell, R. D.; Criddle, C. S. Environ. Sci. Technol.
1997, 31, 2445-2454.
(2) Hansen, K. J.; Clemen, L. A.; Ellefson, M. E.; Johnson, H. O.
Environ. Sci. Technol. 2001, 35, 766-770.
(3) Kannan, K.; Koistinen, J.; Beckmen, K.; Evans, T.; Gorzelany, J.
F.; Hansen, K. J.; Jones, P. D.; Helle, E.; Nyman, M.; Giesy, J. P.
Environ. Sci. Technol. 2001, 35, 1593-1598.
(4) Giesy, J. P.; Kannan, K. Environ. Sci. Technol. 2001, 35 1339-
(5) Taniyasu, S.; Kannan, K.; Horii, Y.; Hanari, N.; Yamashita, N.
Environ. Sci. Technol. 2003, 37, 2634-2639.
S. A. Environ. Sci. Technol. 2002, 36, 545-551.
(7) Mabury, S. A.; Ellis, D. A.; Martin, J. W.; Stock, N. L.; Smithwick,
M.; Muir, D. C. G. Polyfluorination Results in Volatile and
Persistent Chemicalswith High Transport PotentialsPresented
atthe13thAnnualSETAC EuropeMeeting: Hamburg,Germany,
April 30, 2003.
(8) Martin, J. W.; Smithwick, M. M.; Braune, B. M.; Hoekstra, P. F.;
Muir, D. C. G.; Mabury, S. Environ. Sci. Technol. 2004, 38 (2),
J. H. Environ. Sci. Technol. 2003, 37, 881-891.
(10) Martin, J. W.; Mabury, S. A.; Solomon, K. R.; Muir, D. C. G.
Environ. Toxicol. Chem. 2001, 22, 189-195.
(11) Martin, J. W.; Mabury, S. A.; Solomon, K. R.; Muir, D. C. G.
Environ. Toxicol. Chem. 2001, 22, 196-204.
(12) Berthiaume, J.; Wallace, K. B. Toxicol. Lett. 2002, 129, 23-32.
(13) Upham, B. L.; Deocampo, N. D.; Wurl, B.; Trosko, J. E. Int. J.
Cancer 1998, 78, 491-495.
(14) Kissa, E. Fluorinated Surfactants, Synthesis, Properties, Applica-
tions; Marcel Dekker: New York, 1994; Vol. 50.
(15) Ellis, D. A.; Martin, J. W.; Mabury, S. A.; Hurley, M. D.; Sulbaek
Andersen, M. P.; Wallington, T. J. Environ. Sci. Technol. 2003,
(16) Stock, N.; Ellis, D. A.; Deleebeeck, L.; Muir, D. C. G.; Mabury S.
A. Environ. Sci. Technol. 2004, 38 (6), 1693-1699.
K. R.; Mabury, S. A. Anal. Chem. 2002, 74, 584-590.
(18) Stock, N.; Lau, F.; Martin, J. W.; Muir, D. C. G.; Mabury, S. A.
Environ. Sci Technol. 2004, 38 (4), 991-996.
(19) Hagen, D. F.; Belisle, J.; Johnson, J. D.; Venkateswarlu, P. Anal.
Biochem. 1981, 118, 336-343.
3M Environmental Laboratory, November 6, 2002. U.S. EPA
Public Docket AR-226-0555.
(21) Napoli, M.; Scipioni, A.; Legnaro, C. E.; Krotz, L. N. J. Fluorine
Chem. 1994, 66, 249-252.
1995, 70, 19-26.
(23) Cox, E. E.; Major, D.; Edwards, E. A. Natural Attenuation of
facility. In Remediation of Chlorinated and Racalcitrant Com-
Press: Columbus, OH, 2000; Vol. 1.
(24) Telomer Research Program. Telomer Research Program Updates
Presented to the USEPA OPPT, November 25, 2002; U.S.
Environmental Protection Agency public docket AR226-1141.
(25) Commission of the European Communities. Commission
Decision 2002/657/ EC implementing Council Directive 96/
23/EC concerning the performance of analytical methods and
theinterpretation ofresults.Off.j.Eur.Communities2002, Aug
(26) Ellis, D.A.;Mabury, S.A.;Martin, J.;Muir, D.C.G.Nature2001,
(27) Information provided by Oakwood Products Inc., West Co-
(28) Ludwig, B.; Akundi, A.; Kendall, K. Appl. Environ. Microbiol.
1995, 61, 3729-3733.
(29) Van Iersel, M. F. M.; Eppink, M. H. M.; VanBerkel, W. J. H.;
Rombouts, F. M.; Abee, T. Appl. Environ. Microbiol. 1997, 63,
(30) Malone, V. F.; Chastain, A. J.; Ohlsson, J. T.; Poneleit, L. S.;
Nemecek-Marshall, M.; Fall, R. Appl. Environ. Microbiol. 1999,
2001, 67, 4919-4921.
(32) Crosby, D.G.Environmental ToxicologyandChemistry;Oxford
University Press: New York, 1998.
Received for review September 15, 2003. Revised manuscript
received December 20, 2003. Accepted March 4, 2004.
28649ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 10, 2004