Abiotic Hydrolysis of Fluorotelomer-Based Polymers as a Source of

Nov 3, 2015 - Emissions of Per- and Polyfluoroalkyl Substances in a Textile Manufacturing Plant in China and Their Relevance for Workers' Exposure. Fr...
0 downloads 9 Views 2MB Size
Download PDF
Article pubs.acs.org/est

Abiotic Hydrolysis of Fluorotelomer-Based Polymers as a Source of Perfluorocarboxylates at the Global Scale John W. Washington*,† and Thomas M. Jenkins‡ †

USEPA, National Exposure Research Laboratory, 960 College Station Road, Athens, Georgia 30605-2700, United States USEPA, Senior Environmental Employment Program, Athens, Georgia 30605-2700, United States

‡

S Supporting Information *

ABSTRACT: Fluorotelomer-based polymers (FTPs) are the main product of the fluorotelomer industry. For nearly 10 years, whether FTPs degrade to form perfluorooctanoate (PFOA) and perfluorocarboxylate (PFCA) homologues has been vigorously contested. Here we show that circum-neutral abiotic hydrolysis of a commercial FTP proceeds with half-life estimates of 55−89 years and that base-mediated hydrolysis overtakes neutral hydrolysis at about pH = 10, with a half-life of ∼0.7 years at pH ∼ 12. Considered in light of the large production volume of FTPs and the poor efficacy of conventional treatments for recovery of PFCAs from waste streams, these results suggest that FTPs manufactured to date potentially could increase PFCAs 4- to 8-fold over current oceanic loads, largely depending on the integrity of disposal units to contain PFCAs upon hydrolytic generation from FTPs.

■

modeling approach was too simplistic.11 Russell et al. commented on our paper, indicating that their extraction had excellent recoveries of spiked standards and that our experiment had too few time points.12 They went on to reaffirm their “conclusion that the half-life of a commercial fluorotelomerbased acrylate polymer is between 1200 and 1700 years and that this source represents a small potential contributor to levels of PFOA in the environment.”12 As recently as 2013, a literature review of perfluorinated compounds by a former industry employee fell squarely with the long half-life contingent by questioning the practicality of testing degradability of highmolecular-weight polymers at all, presuming likely half-lives of a thousand years or longer.13 In late 2014, two new studies bolstered the short half-life evidence. Rankin et al. reported upon the biodegradation of an acrylate FTP they synthesized in-house, with half-life estimates of 8 to 111 years.14 And following extensive method-development efforts,15 we reported the results of a biodegradation experiment with two commercial acrylate FTPs yielding half-life estimates of 33 to 112 years.16 An unexpected finding for one of our controls, consisting of an FTP mounted on cotton in water, was that the commercial FTP evidently degraded by abiotic hydrolysis at rates falling within the range of the “biologically mediated” microcosms. In an effort to elucidate this unexpected result for the hydrolysis controls, we conducted hydrolysis experiments at

INTRODUCTION In the early 2000s, perfluorinated compounds such as perfluorooctanesulfonate (PFOS) and perfluorooctanoate (PFOA) attracted intense concern with recognition of their combined traits of toxicity1−4 and widespread environmental occurrence.5,6 Soon thereafter, attention turned to the possible role of fluorotelomer-based polymers (FTPs; Figure 1A) as a source of PFOA and homologous perfluorocarboxylic acids (PFCAs) because of their high production volumes. From 2006 to 2010, for example, PFOA precursor production for FTP synthesis is estimated to have been roughly 3-fold that for summed perfluorooctanoate production of all other major polymeric lines,7 and production of PFCA homologues of C10 and longer almost entirely went to FT production.7 However, researchers soon discovered that resolving whether FTPs degrade to form PFCAs and related compounds presented unique challenges because the very trait that FTPs were designed to impart, a disinclination to interact with all other materials, confounds characterization of FTP stability under environmental conditions. In 2008, Russell et al. published the first effort to determine the biodegradability of acrylate FTPs, reporting that no degradation was detectable up to half-lives of >1000 years.8 Roughly a year later, we published a paper disputing this conclusion, arguing that Russell’s extraction of analytes from the FTP microcosms was ineffective, among numerous other concerns,9,10 and modeled experimental results with an industry-synthesized test acrylate FTP to generate a half-life estimate for commercial FTPs of 10 to 17 years. Predictably, these antithetical conclusions met with considerable controversy. A news brief fronting our paper quoted an industry-funded researcher as stating that our © 2015 American Chemical Society

Received: Revised: Accepted: Published: 14129

July 31, 2015 October 28, 2015 November 3, 2015 November 3, 2015 DOI: 10.1021/acs.est.5b03686 Environ. Sci. Technol. 2015, 49, 14129−14135

Article

Environmental Science & Technology

in the SI. Sources of other chemicals used in this study are reported in the SI. The composition of water solutions, buffered to pH 5, 6, 7, 8, 9, 10, 11, and 12, is summarized in Table S2. Microcosm Design and Storage. All treatments, controls, and blanks were prepared in Nalgene 16 mL polypropylene copolymer centrifuge tubes with caps. For treatments, an FTPbearing cotton tuft was added to the tube, then ∼4 g of deionized buffered water (SI) was added, and the tube was capped. Process blank microcosms were constructed identically to treatment microcosms, except that no FTP was added. Control microcosms were constructed identically to treatment microcosms, except that no FTP was added, but known amounts of 8:2 FTOH were added by injection from a syringe into a submerged cotton tuft. All treatments, controls, and blanks were kept in a dark incubator maintained at 25 °C until they were extracted. Extraction. Upon the basis of extensive method development,15 and considering the relatively simple microcosm matrix of this experiment (i.e., commercial FTP, cotton, buffered water), extractions were performed with two serial MTBE extractions of each microcosm as previously described.15 At each of the eight pHs, five replicate treatment microcosms were extracted at 0, 7, 14, 21, 28, 35, 49, 63, and 77 days. At the same time that treatment microcosms were extracted, one treatment microcosm was dedicated to pH analysis, and a control and process blank were extracted for GC/MS analysis as well. Analytical. Analyses were performed on an Agilent 6890N gas chromatograph equipped with an Agilent 5975 mass spectrometer (GC/MS) running in positive chemical-ionization (PCI) mode. Quantitation was performed with authentic standards for each analyte, 8:2 FTOH and 10:2 FTOH, using a mass-labeled internal standard, 13C2-6:2 nFTOH. Instrument analytical details are described in our earlier papers.15,16 We found that in order to avoid contamination of the GC/MS (inlet, column, or ionizing source), samples needed to be diluted so that the highest anticipated concentrations fell at no more than 100 ng/mL, roughly half of our maximum calibration standard. Sample concentrations higher than this range contaminated the inlet (necessitating replacement of the gooseneck inlet and gold seal), the front of the column (requiring trimming of the column), or plated out on the ionizing source to diminish the signal (requiring cleaning of the source). Analytical Sequence. Once all samples are extracted, the sample set is a three-dimensional array: (replicates 1 through 5) × (sample rounds 1 through 9) × (pHs 5 through 12). We chose to analyze each replicate completely, starting at sample round 1 and pH 5, proceeding through subsequent sampling rounds of rep 1 at pH 5, then chronologically analyzing rep 1, pH 6, and so on until the round 9 of rep 1 pH 12 was complete. Then we started replicate 2 (round 1, pH 5) and proceeded in the same order. This blocking analytical sequence is preferred for kinetic studies because the rate constant is determined from the slope of the line defined by Ln(FTP moiety remaining) versus time. By blocking, or segregating replicates from each other during analysis, each replicate justifiably can be used singularly to define a slope, and thereby estimate the rate constant. By following this practice, each rate constant is represented by as many experimental estimates as there are replicates, five in the case of our experiment. In addition to logically justifying grouping analytical results by replicate for the reason for generating independent estimates of rate constants, this analytical sequence

Figure 1. General structure of an acrylate FTP (A). The inferred hydrolysis mechanism, nucleophilic acyl substitution21 (B). Nucleophilic attack is by OH− at high pHs (BAC2) and water at lower pHs (WAC2).

25 °C on one commercial FTP17 consisting of the following: (1) eight pHs buffered over the range 5−12; (2) nine sample rounds over 77 days; (3) five replicate microcosms for each pH and sample round; and (4) two analytes that are products of FTP hydrolysis, 8:2 fluorotelomer alcohol (8:2 FTOH) and 10:2 FTOH. We then applied the results of this experiment to a simple model to explore the implications of decades-scale abiotic FTP hydrolysis.

■

EXPERIMENTAL SECTION Chemicals. A commercial acrylate FTP, manufactured by DuPont, was tested for hydrolytic degradability. This is one of the DuPont FTPs we reported upon in our recent report on FTP degradation.16 The FTP tested here contained ∼50% C8 telomers and ∼30% C10 telomers (Supporting Information (SI); Table S1). Information on the FTP preparation for the experiment, including mounting it on cotton tufts,16 is provided 14130

DOI: 10.1021/acs.est.5b03686 Environ. Sci. Technol. 2015, 49, 14129−14135

Article

Environmental Science & Technology minimizes potential artifactual effects from changing instrument conditions during the sequence run, say from plating sample out on the ionizing source to decrease instrument sensitivity (this example is illustrated in the Data-Quality Inspection section of the SI). Data Quality Inspection. Process blanks, controls, and check standards were run every day of analysis alongside the treatments. Process blanks uniformly returned low to nondetectable values of 8:2 FTOH and 10:2 FTOH, consistent with intent. Check standards uniformly fell within 10% of the nominal value, consistent with the QA criterion of ±30%. Controls decreased in [8:2 FTOH] through time (Table S4). This was not unexpected because FTOH is volatile, so it could potentially have been lost by volatilization. Considering that we found hydrolytic degradation of the FTP, inferred from increases in 8:2 FTOH and 10:2 FTOH over the course of the

experiment, the evident losses in these controls were not fatal to the experiment. Instead, if treatments suffered FTOH losses similar to controls, then this effect would result in underestimation of hydrolysis rates. Analytical outliers were identified and excluded using the coefficient of variation (COV) among the five replicates. The first sampling round in which analytes were detected (Day 1 for pHs 5−11, Day 7 for pH 12) were found to be more variable than subsequent rounds, so these data were all included. In subsequent rounds, when the COV exceeded 0.3, the most extreme data point was excluded. This screening filter omitted seven out of a possible 720 data points (8 pHs × 9 sample rounds × 5 replicates × 2 analytes) and the resulting COVs all fell below 0.3 following these exclusions. Of these 720 possible data points, those data missing as a consequence of mishandling during laboratory operations (6 samples, 11 data points) or expunged according to the COV quality criterion are summarized in Table S5. Additional data-quality details are described in the SI.

■

RESULTS AND DISCUSSION Both monitored hydrolysis products, 8:2 FTOH and 10:2 FTOH, were observed to increase over the course of the experiment at each of the eight buffered pHs (Figure 2 and

Figure 2. [8:2 FTOH] vs time at pH 6, 8, 10, and 12 (with remaining experimental pHs not shown for clarity). Y axis is depicted in log scale to accommodate wide range of concentrations at pH 12. Error bars depicting 1 standard deviation among five microcosms are shown but do not exceed the span of many data points.

Figure 4. Log kobs for 8:2 FTOH (A) and 10:2 FTOH (B) vs pH: Each kobs data point is determined from the least-squares slope of a nine-sampling round effort as shown in Figure 3. Dashed lines are modeled kH2O (horizontal) and kOH−[OH−] (sloped). Heavy curve is composite kobs. For both 8:2 FTOH and 10:2 FTOH, hydrolysis was evidently slightly slower than other circum-neutral pHs; pH 5 alone was buffered with citrate (Table S2), and we suspect this uniquely employed buffer caused this small artifact (SI discussion).

Figure 3. Ln[8:2 FTOH equiv.] vs time. Depicted values are the mean measured at each time for five microcosms. Standard deviation is shown only for pH 12 for clarity, error generally is less for all other pH values. Note the Y axis of 2A spans a narrower range than 2B. 14131

DOI: 10.1021/acs.est.5b03686 Environ. Sci. Technol. 2015, 49, 14129−14135

Article

Environmental Science & Technology

Figure 5. Conceptual Potential Energy (increases upward) vs Reaction Coordinate (time progresses left to right) of FTP degradation. As a possible explanation of why nFTOHs apparently are so stable in our FTP experiment, we propose that, as telomers cleave from the polymer backbone, the activation-energy barrier and reaction-product potential energy are lower for the telomers to remain closely co-associated with the polymer, as molecular-clip clathrate-type structures, than for the hydrophobic telomers to disperse into the bulk water phase. After some critical degree of polymer degradation, the thermodynamic potential of the polymer-clathrate superstructure will likely exceed that for dispersion of the largely unbound telomers, resulting in a critical dispersion point in the degradation reaction.

Tables S6−S13). Hydrolysis reactions are generally first order in the hydrolyzing species and, when the pH is held constant, follow the following form: dC = −kobsC dt

a bimodal rate equation wherein neutral H2O hydrolysis is predominant at low and moderate pHs, and base-mediated hydrolysis controls reaction rate at high pHs: d[FTP] = −k H2O[FTP] − k OH −[FTP][OH−]m dt

(1)

(2)

where kH2O is the rate constant for water hydrolysis of the FTP, kOH− is the rate constant for base-mediated hydrolysis, and m is the order of the reaction rate in [OH−]. Given that this reaction takes place at the ester linkage (Figure 1A), yields alcohol products (Figure 2) and is OH− mediated at high pHs (Figure 4), the reaction mechanism evidently is nucleophilic acyl substitution21 (Figure 1B); base catalyzed (BAC2) at high pH and mediated by water at intermediate pHs (WAC2). The variables in eq 2 can be evaluated from experimental data by equating eqs 1 and 2:

where C is the concentration of the hydrolyzing species, kobs is a (pseudo)first-order rate constant, and t is time.18 In this case, C is the FTP. Upon the basis of data in the original patent16,17 (Table S1) we took the original concentration of the 8:2 and 10:2 moieties in the FTP to be 884 and 436 μmol/g FTP, respectively. Assigning these values as the time-zero concentration in our experiment, we calculated FTP moieties remaining at any given time by subtracting from these values the extracted μmoles of FTOH per g FTP. Plotting the natural log of FTP moieties remaining vs time (Figure 3), the resulting leastsquares linear slope solves for the value of kobs.18 When log kobs is plotted against pH in buffered experiments, a nonzero slope indicates that the reaction was pseudo-firstorder.19,20 Examining our data (Figure 4), kobs varies little for pHs 5, 6, 7, and 8, indicating that the rate of FTP hydrolysis is not a strong function of pH over this range. At higher pHs, log kobs increases roughly linearly with increasing pH for pHs 10, 11, and 12, with pH 9 falling close to the inflection point. A positive relationship between rate constant and pH indicates that the reaction rate is a function of [OH−]. Considering the near-horizontal and pH-dependent domains together suggests

kobs = k H2O + k OH −[OH−]m

(3)

At low and moderate pHs, kobs ≈ kH2O (Figure 4), so we calculated kH2O, as the geometric mean value of kobs values for pHs 5−6 (those pHs furthest from the inflection with base mediation), to be 3.43 × 10−5 d−1 for 8:2 FTOH. At high pHs where kH2O ≪ kOH−[OH−]m, kobs equates directly to the composite function of kOH−[OH−]m, which includes two unknowns, kOH− and m. Noting that log(kOH−[OH−]m) = mlog[OH−] + logkOH−, both unknowns can be identified simultaneously by regressing logkobs against pOH for pHs 11−12 (those pHs 14132

DOI: 10.1021/acs.est.5b03686 Environ. Sci. Technol. 2015, 49, 14129−14135

Article

Environmental Science & Technology

FTPs in landfills to current PFCA loads. In combination with the FTP hydrolysis rates we report here, Wang et al. reported estimates of the data necessary to perform this comparison including estimates of PFCA loads in the oceans and annual FTP manufacturing rates.7 Modeling the FTP manufacturing rate as zeroorder and FTP hydrolysis as first-order, the mass of commercial FTP at any time (t) is given by (see SI Model Derivation):

furthest from the inflection with water mediation). Regression of log-transformed values, as opposed to untransformed, also: (i) addresses the heteroscedasticity common in data sets spanning orders of magnitude, and (ii) rightly assumes that the kOH−[OH−]m component of kobs as having zero value at [OH−] = 0. Using this approach, kOH− = 2.68 × 10−2 d−1 mol−1, and the reaction is m = 0.597 order in [OH−] for 8:2 FTOH. Values of kobs modeled from the data and eq 3 are plotted in Figure 4, and the kinetic constants modeled from these values are reported in Tables S14 and S15. The order of the basemediated reaction, m ≈ 0.6 for both 8:2 and 10:2 FTOH, is noteworthy because the bimolecular BAC2 reaction usually is first order in [OH−].22 The lower order here suggests hindrance related to the FTP solid phase. Considering that OH− only catalyzes the reaction and is not consumed as a reactant (Figure 1B), diffusional constraints seem unlikely, leaving steric hindrance as the likely cause. A noteworthy feature of the data is the persistence of 8:2 FTOH and 10:2 FTOH in our experimental treatments despite their volatility.23−26 In our recent biodegradation study,16 we suggested that this persistence is sustained by the alcohol products of the FTP-degradation reaction occupying their original position in the FTP as clathrate guest compounds, a consequence of thermodynamic stabilization imparted to the system by minimizing the hydrophobic-aqueous interface. In this current experiment, the diametric contrast, we observed of FTOH increases in all of our treatments (Figure 2, Tables S6−S13) versus the dramatic decreases we observed for 8:2 FTOH in our controls (which did not contain FTP; Table S4) strongly supports this hypothesis. To the extent that clathrates are formed during FTP hydrolysis, the evolution of FTOHs from FTPs might occur sporadically as critical dispersion points are achieved during FTP degradation (Figure 5). The 8:2 FTOH results indicate that the commercial acrylate FTP we tested at 25 °C has an abiotic hydrolytic half-life of roughly 55 years at low pHs where water hydrolysis controls reaction rate, decreasing to ∼0.7 years at pH ∼ 12 (Figure 4A). Hydrolytic modeling of 10:2 FTOH data suggest that the FTP half-life at low pH is as high as 89 years (Figure 4B; Table S14), but the 8:2 FTOH results are more reliable because 8:2 FTOH moieties are present at nearly twice concentration of 10:2 FTOH in the original FTP (Table S1) and because laboratory handling and analysis of 8:2 FTOH is less subject to laboratory artifacts than 10:2 FTOH as well.

[FTP] =

Φ − (Φ − kobs[FTP]0 )e−kobsΔt kobs

(4)

where Φ is commercial FTP production rates (metric tons/yr), and [FTP]0 is FTP mass at the beginning of the period of Δt. PFCA yields from 6:2 FTOH have been reported to be as low as ∼6% in 28-day experiments mediated by wood-rotting fungus.35 For 8:2 FTOH incubated in soils for roughly 200 days, however, Wang et al. reported a mean PFOA yield to be ∼25% (10% to 40% range).36 Using this 25% yield, Figure 6 shows a potential future yield of PFCAs from FTPs

■

IMPLICATION OF ABIOTIC FTP HYDROLYSIS Numerous studies have shown that landfills are a source of polyfluorinated chemicals to the atmosphere27 and water,28−30 and that conventional water treatment is partially to largely ineffective for removing these compounds from wastestreamflow.31−33 Prevedouros et al. have argued that perfluorinated compounds released to the environment ultimately make their way to the oceans and remain in the marine ecosystems until the slow geologic process of sediment burial sequesters them.34 Considered in light of these FTP-hydrolysis findings, FTP-bearing refuse disposed in landfills constitutes a potential long-term source of PFCAs to the environment, including oceans. The general magnitude of possible environmental PFCA loads from FTP hydrolysis can be roughly assessed by assuming that most FTP products ultimately will be landfilled where they are likely to be exposed to high-humidity and saturated conditions, and comparing potential future PFCA releases from

Figure 6. Tons of PFCAs (PFOA in A; PFDA in B) from FTPs starting in 2015 vs time (modeled with eq 4) using FTP production rates and estimated oceanic loads reported in Wang et al. (estimated load ranges: PFOA = 3500−7000 t; PFDA = 890−1800 t).7 Depicted half-lives bracket uncertainty of hydrolysis rates from data generated herein. These efforts suggest that hydrolysis of commercial FTPs already produced can potentially release more PFCAs to the environment than present loads if landfilled FTPs are not contained more effectively than present practice.

produced through 2015. This model predicts a quadrupling of global oceanic PFOA load from landfilled FTPs over the next two centuries and an 8-fold increase for perfluorodecanoic acid (PFDA). This modeling approach does not address the timing of when hydrolysis begins (e.g., during product use or upon disposal) or other possible rate-limiting processes 14133

DOI: 10.1021/acs.est.5b03686 Environ. Sci. Technol. 2015, 49, 14129−14135

Article

Environmental Science & Technology

(5) Giesy, J. P.; Kannan, K. Global distribution of perfluorooctane sulfonate in wildlife. Environ. Sci. Technol. 2001, 35 (7), 1339−1342. (6) Hansen, K. J.; Johnson, H. O.; Eldridge, J. S.; Butenhoff, J. L.; Dick, L. A. Quantitative characterization of trace levels of PFOS and PFOA in the Tennessee River. Environ. Sci. Technol. 2002, 36 (8), 1681−1685. (7) Wang, Z.; Cousins, I. T.; Scheringer, M.; Buck, R. C.; Hungerbuehler, K. Global emission inventories for C-4-C-14 perfluoroalkyl carboxylic acid (PFCA) homologues from 1951 to 2030, Part I: production and emissions from quantifiable sources. Environ. Int. 2014, 70, 62−75. (8) Russell, M. H.; Berti, W. R.; Szostek, B.; Buck, R. C. Investigation of the biodegradation potential of a fluoroacrylate polymer product in aerobic soils. Environ. Sci. Technol. 2008, 42 (3), 800−807. (9) Washington, J. W.; Ellington, J. J.; Jenkins, T. M.; Evans, J. J.; Yoo, H.; Hafner, S. C. Degradability of an Acrylate-Linked, Fluorotelomer Polymer in Soil. Environ. Sci. Technol. 2009, 43 (17), 6617−6623. (10) Washington, J. W.; Ellington, J. J.; Jenkins, T. M.; Yoo, H. Response to Comments on ″Degradability of an Acrylate-Linked, Fluorotelomer Polymer in Soil. Environ. Sci. Technol. 2010, 44 (2), 849−850. (11) Renner, R. Perfluoropolymer degrades in decades, study estimates. Environ. Sci. Technol. 2009, 43 (17), 6445. (12) Russell, M. H.; Wang, N.; Berti, W. R.; Szostek, B.; Buck, R. C. Comment on ″Degradability of an acrylate-linked, fluorotelomer polymer in soil. Environ. Sci. Technol. 2010, 44 (2), 848−848. (13) Liu, J.; Avendano, S. M. Microbial degradation of polyfluoroalkyl chemicals in the environment: A review. Environ. Int. 2013, 61, 98−114. (14) Rankin, K.; Lee, H.; Tseng, P. J.; Mabury, S. A. Investigating the biodegradability of a fluorotelomer-based acrylate polymer in a soilplant microcosm by indirect and direct analysis. Environ. Sci. Technol. 2014, 48, 12783−12790. (15) Washington, J. W.; Naile, J. E.; Jenkins, T. M.; Lynch, D. G. Characterizing Fluorotelomer and Polyfluoroalkyl Substances in New and Aged Fluorotelomer-Based Polymers for Degradation Studies with GC/MS and LC/MS/MS. Environ. Sci. Technol. 2014, 48 (10), 5762− 5769. (16) Washington, J. W.; Jenkins, T. M.; Rankin, K.; Naile, J. E. Decades-scale degradation of commercial, side-chain, fluorotelomerbased polymers in soils & water. Environ. Sci. Technol. 2015, 49 (2), 915−923. (17) Fitzgerald, J. J. Oil, Water and Solvent Resistant Paper by Treatment with Fluorochemical CoPolymers. U.S. Patent #5,674,961, 1997. (18) Washington, J. W. Hydrolysis rates of dissolved volatile organic compounds: principles, temperature effects and literature review. Groundwater 1995, 33 (3), 415−424. (19) Hoffman, M. R. Kinetics and mechanism of oxidation of hydrogen sulfide by hydrogen peroxide in acidic solution. Environ. Sci. Technol. 1977, 11 (1), 61−66. (20) Lasaga, A. C.; Kirkpatrick, R. J., Kinetics of Geochemical Processes. In Reviews in Mineralogy; Mineralogical Society of America: Chantilly, VA, 1981; Vol. 8. (21) Larson, R. A.; Weber, E. J. Reaction Mechanisms in Environmental Organic Chemistry. CRC Press: Boca Raton, FL, 1994. (22) Robinson, B. A.; Tester, J. W. Kinetics of alkaline hydrolysis of organic esters and amides in neutrally-buffered solution. Int. J. Chem. Kinet. 1990, 22 (5), 431−448. (23) Cobranchi, D. P.; Botelho, M.; Buxton, L. W.; Buck, R. C.; Kaiser, M. A. Vapor pressure determinations of 8−2 fluorotelomer alcohol and 1-H perfluorooctane by capillary gas chromatography Relative retention time versus headspace methods. J. Chromatogr. A 2006, 1108 (2), 248−251. (24) Krusic, P. J.; Marchione, A. A.; Davidson, F.; Kaiser, M. A.; Kao, C. P. C.; Richardson, R. E.; Botelho, M.; Waterland, R. L.; Buck, R. C. Vapor pressure and intramolecular hydrogen bonding in fluorotelomer alcohols. J. Phys. Chem. A 2005, 109 (28), 6232−6241.

(e.g., variable-saturation disposal conditions, slow leaching from landfills, slower kinetics at temperatures less than the 25 °C maintained in this experiment, variable kinetics among FTPs), but rate-limiting uncertainties simply shift the timing of the load curves in Figure 4, not the ultimate asymptotic limit. Following the premise of Prevedouros et al. regarding the ultimate fate of environmental PFCAs,34 so long as disposal of FTPs is mostly in landfills with existing effluent-treatment technology and geologic sequestration of PFCAs in deep-ocean sediments is slow relative to landfill releases, marine PFCAs will continue to approach the asymptotic limit. As such, these newly reported hydrolysis rates, applied to the mass of commercial FTPs already produced, document the potential for presently landfilled reserves of commercial FTPs to increase global PFCA loads considerably over the coming decades, a matter worthy of additional research and perhaps action.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.5b03686. Table S1: Physical properties of tested commercial fluorotelomer-based polymer; Table S2: pH Buffer Composition Mounting FTP on Cotton Data Quality Inspection; Figure S1: Mean 13C2-6:2 FTOH peak area plotted as a function of analytical sequence; Table S3: Measured pH in microcosms over the course of the experiment; Table S4: FTOH in controls; Table S5: Missing and Expunged Data Supporting Data; Tables S6− S13 pH 5−12 Raw Data and kobs; Table S14: Summary of Experimental and Modeled Kinetic Variables; Table S15: Modeled Kinetic Constants; Figure S2: [8:2 FTOH]/ [10:2 FTOH] molar ratio vs time Equations S1−S8; Table S16: Values Generated by Application of Equation 4 as Depicted In Figure 6; and Supporting References(PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Phone (706) 355-8227; e-mail [email protected] (J.W.W.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Thanks to Scott Mabury, Keegan Rankin, and Eric Weber for helpful input. This research was funded by the USEPA Office of Research & Development. Views expressed in this paper do not necessarily represent the views or policies of the EPA. Mention of trade names or products does not convey EPA approval, endorsement or recommendation.

■

REFERENCES

(1) Ophaug, R. H.; Singer, L. Metabolic handling of perfluorooctanoic acid in rats. Exp. Biol. Med. 1980, 163 (1), 19−23. (2) Staples, R. E.; Burgess, B. A.; Kerns, W. D. The embryo fetal toxicity and teratogenic potential of ammonium perfluorooctanoate (APFO) in the rat. Toxicol. Sci. 1984, 4 (3), 429−440. (3) Kennedy, G. L. Dermal toxicity of ammonium perfluorooctanoate. Toxicol. Appl. Pharmacol. 1985, 81 (2), 348−355. (4) Langley, A. E.; Pilcher, G. D. Thyroid, bradycardic and hypothermic effects of perfluoro-n-decanoic acid in rats. J. Toxicol. Environ. Health 1985, 15 (3−4), 485−491. 14134

DOI: 10.1021/acs.est.5b03686 Environ. Sci. Technol. 2015, 49, 14129−14135

Article

Environmental Science & Technology (25) Lei, Y. D.; Wania, F.; Mathers, D.; Mabury, S. A. Determination of vapor pressures, octanol-air, and water-air partition coefficients for polyfluorinated sulfonamide, sulfonamidoethanols, and telomer alcohols. J. Chem. Eng. Data 2004, 49 (4), 1013−1022. (26) Stock, N. L.; Ellis, D. A.; Deleebeeck, L.; Muir, D. C. G.; Mabury, S. A. Vapor pressures of the fluorinated telomer alcohols – Limitations of estimation methods. Environ. Sci. Technol. 2004, 38, 1693−1699. (27) Ahrens, L.; Shoeib, M.; Harner, T.; Lee, S. C.; Guo, R.; Reiner, E. J. Wastewater Treatment Plant and Landfills as Sources of Polyfluoroalkyl Compounds to the Atmosphere. Environ. Sci. Technol. 2011, 45 (19), 8098−8105. (28) Benskin, J. P.; Li, B.; Ikonomou, M. G.; Grace, J. R.; Li, L. Y. Per- and Polyfluoroalkyl Substances in Landfill Leachate: Patterns, Time Trends, and Sources. Environ. Sci. Technol. 2012, 46 (21), 11532−11540. (29) Busch, J.; Ahrens, L.; Sturm, R.; Ebinghaus, R. Polyfluoroalkyl compounds in landfill leachates. Environ. Pollut. 2010, 158 (5), 1467− 1471. (30) Allred, B. M.; Lang, J. R.; Barlaz, M. A.; Field, J. A. Physical and biological release of poly- and perfluoroalkyl substances (PFASs) from municipal solid waste in anaerobic model landfill reactors. Environ. Sci. Technol. 2015, 49, 7648−7656. (31) Schultz, M. M.; Higgins, C. P.; Huset, C. A.; Luthy, R. G.; Barofsky, D. F.; Field, J. A. Fluorochemical mass flows in a municipal wastewater treatment facility. Environ. Sci. Technol. 2006, 40, 7350− 7357. (32) Sun, H.; Li, F.; Zhang, T.; Zhang, X.; He, N.; Song, Q.; Zhao, L.; Sun, L.; Sun, T. Perfluorinated compounds in surface waters and WWTPs in Shenyang, China: Mass flows and source analysis. Water Res. 2011, 45 (15), 4483−4490. (33) Murakami, M.; Shinohara, H.; Takada, H. Evaluation of wastewater and street runoff as sources of perfluorinated surfactants (PFSs). Chemosphere 2009, 74 (4), 487−493. (34) Prevedouros, K.; Cousins, I. T.; Buck, R. C.; Korzeniowski, S. H. Sources, fate and transport of perfluorocarboxylates. Environ. Sci. Technol. 2006, 40 (1), 32−44. (35) Tseng, N.; Wang, N.; Szostek, B.; Mahendra, S. Biotransformation of 6:2 Fluorotelomer Alcohol (6:2 FTOH) by a Wood-Rotting Fungus. Environ. Sci. Technol. 2014, 48 (7), 4012−4020. (36) Wang, N.; Szostek, B.; Buck, R. C.; Folsom, P. W.; Sulecki, L. M.; Gannon, J. T. 8−2 Fluorotelomer alcohol aerobic soil biodegradation: Pathways, metabolites, and metabolite yields. Chemosphere 2009, 75 (8), 1089−1096.

14135

DOI: 10.1021/acs.est.5b03686 Environ. Sci. Technol. 2015, 49, 14129−14135