Article pubs.acs.org/est
Observation of a Novel PFOS-Precursor, the Perfluorooctane Sulfonamido Ethanol-Based Phosphate (SAmPAP) Diester, in Marine Sediments Jonathan P. Benskin,*,†,‡ Michael G. Ikonomou,‡ Frank A. P. C. Gobas,§ Million B. Woudneh,† and John R. Cosgrove† †
AXYS Analytical Services Ltd., 2045 Mills Road West, Sidney BC, Canada, V8L 5X2 Institute of Ocean Sciences, Fisheries and Oceans Canada (DFO), 9860 West Saanich Road, Sidney BC, Canada V8L 4B2 § School of Resource and Environmental Management, Faculty of Applied Sciences, Simon Fraser University, Burnaby, British Columbia, Canada V5A 1S6 ‡
S Supporting Information *
ABSTRACT: The environmental occurrence of perfluorooctane sulfonate (PFOS) can arise from its direct use as well as from transformation of precursors ((N-alkyl substituted) perfluorooctane sulfonamides; FOSAMs). Perfluorooctane sulfonamidoethanol-based phosphate (SAmPAP) esters are among numerous potential PFOS-precursors which have not been previously detected in the environment and for which little is known about their stability. Based on their high production volume during the 1970s−2002 and widespread use in food contact paper and packaging, SAmPAP esters may be potentially significant sources of PFOS. Here we report for the first time on the environmental occurrence of SAmPAP diester in marine sediments from an urbanized marine harbor in Vancouver, Canada. SAmPAP diester concentrations in sediment (40−200 pg/g dry weight) were similar to those of PFOS (71−180 pg/g). A significant (p < 0.05) correlation was observed between SAmPAP diester and N-ethyl perfluorooctane sulfonamido acetate (an anticipated degradation product of SAmPAP diester). ∑PFOS-precursor (FOSAM) concentrations in sediment (120−1100 pg/g) were 1.6−24 times greater than those of PFOS in sediment. Although SAmPAP diester was not detected in water, PFOS was observed at concentrations up to 710 pg/L. Among the per- and polyfluoroalkyl substances monitored in the present work, mean log-transformed sediment/water distribution coefficients ranged from 2.3 to 4.3 and increased with number of CF2 units and N-alkyl substitution (in the case of FOSAMs). Overall, these results highlight the importance of FOSAMs as potentially significant sources of PFOS, in particular for urban marine environments.
■
INTRODUCTION Among the numerous per- and polyfluoroalkyl substances (collectively “PFASs”) observed in the global environment, perfluorooctane sulfonate (PFOS, C8F17SO3−) is typically the most common. Due to concerns regarding its persistence in humans1,2 and the environment,3,4 as well as the adverse health outcomes linked to this compound,5 PFOS was voluntarily phased out by its primary manufacturer in 2001/2002, and in 2009 was added to the list of Persistent Organic Pollutants (POPs) regulated by the International Stockholm Convention. However, use exemptions listed under Annex B of the Convention have resulted in ongoing manufacture and use of PFOS in developing countries.6 Although a rapid decline in PFOS concentrations has been observed in some humans and wildlife over the past decade,1,7,8 samples from other parts of © 2012 American Chemical Society
the world indicate a continued increase or no change in levels following the 2002 phase-out.2,9−12 The potential sources of PFOS are numerous and not well characterized. 13 In addition to emissions from direct manufacture and use, the environmental occurrence of PFOS can arise from both abiotic14,15 and biological16,17 degradation of precursors (i.e., (N-alkyl substituted) perfluorooctane sulfonamides; FOSAMs). These substances are structurally and functionally diverse, and only a fraction of them are routinely monitored.13 In addition to their direct use as commercial products, FOSAMs can arise as unintentional Received: Revised: Accepted: Published: 6505
February 29, 2012 May 12, 2012 May 16, 2012 May 16, 2012 dx.doi.org/10.1021/es300823m | Environ. Sci. Technol. 2012, 46, 6505−6514
Environmental Science & Technology
Article
Figure 1. Structures of SAmPAP mono-, di-, and tri-esters (i−iii, respectively), FOSA, FOSAA, EtFOSAA, MeFOSAA (iv−vii, respectively), and PFOS (viii) examined in the present study. Other FOSAMs are possible.
synthetic byproducts during the manufacture of commercial polymers and phosphate surfactants (1−2% of final product).18 These latter substances contain C8F17SO2N- units bound to a polymeric backbone or phosphate moiety via ester linkages which, if hydrolyzed, could lead to the formation of PFOS. In fact, pre-2002 production volume estimates19,20 indicate that FOSAM-based commercial polymers and surfactants may be among the largest potential historical reservoirs of PFOS of all FOSAM or PFOS-containing commercial substances, but to date these substances have never been detected in the environment, and there are limited data available on their stability.21 FOSAM-based phosphate surfactants were first introduced in 1974 by the 3M Co. for use in food contact paper and packaging.22 Formulations typically consisted of 10% mono-, 85% di-, and 5% tri-substituted phosphate esters of N-ethyl perfluorooctane sulfonamido ethanol ((C8F17SO2N(CH2CH3)CH2CH2−)nPO4, n = 1−3; collectively termed “SAmPAPs”,23 Figure 1, structures i, ii, iii, respectively).18 In 1997, sales of FC807 (a commercial SAmPAP formulation) represented the highest quantity of PFOS-equivalents sold by 3M out of all PFOS-precursor (i.e., FOSAM) or PFOS-containing commercial substances.20 The 3M Co. ceased production of these materials along with other perfluorooctane sulfonyl fluoridebased products in 200224 but since then a resurgence in their production has occurred in Asia.6,13 Howard and Muir25 recently included SAmPAP mono- and diesters among their predictions of 610 commercially relevant, persistent, and bioaccumulative organics (Table S1, SI) but to date these substances remain unquantified in any environmental samples. In fact, only two studies (both examining human sera), have attempted to monitor for SAmPAP diesters,23,26 and only one of these reported detection, albeit below quantification limits. In contrast, polyfluoroalkyl phosphoric diesters (diPAPs; e.g. C8F17CH2CH2O)2PO2−), which have replaced SAmPAPs in the food packaging and paper industry, have been detected at elevated concentrations in human sera23,27 and environmental samples, 27,28 and have been shown to biodegrade to fluorotelomer alcohols and perfluoroalkyl carboxylates (PFCAs) in wastewater treatment plant sludge and in rats.29−31
Recent studies of marine sediments from Baltimore Harbor (MD),32 the San Francisco Bay area,32 and Tokyo Bay (Japan)33 have reported potential SAmPAP transformation products (e.g., N-ethyl perfluorooctane sulfonamidoacetate; EtFOSAA)33 at concentrations similar to or exceeding the concentration of PFOS. Due to their extensive historical production and their anticipated partitioning to sediment (sediment/water partition coefficients tend to increase with number of CF2 units),34,35 it is reasonable to hypothesize that the occurrence and biodegradation of SAmPAPs in sediment may contribute to the elevated concentration of other FOSAMs and PFOS in urban marine sediments. In this work, we examined water and sediment samples from an urban marine inlet in Vancouver, BC, Canada for the presence of SAmPAP diester and other FOSAMs; as well as perfluoalkyl sulfonates (PFSAs) and PFCAs. Spatial distribution, sediment−water partition coefficients, and potential sources are discussed. To our knowledge this is the first study which has quantified SAmPAP diester in any environmental samples and provides baseline information on the occurrence and disposition of this historical, high-production volume chemical.
■
EXPERIMENTAL METHODS Standards and Reagents. A full list of reagents is provided in the SI. Perfluorohexanoic acid (PFHxA), perfluoroheptanoic acid (PFHpA), perfluorononanoic acid (PFNA), perfluorodecanoic acid (PFDA), perfluoroundecanoic acid (PFUnDA), perfluorododecanoic acid (PFDoDA), perfluorotetradecanoic acid (PFTeDA), perfluorobutanesulfonate (PFBS), perfluorohexanesulfonate (PFHxS), perfluorooctanoate (PFOA), and PFOS were all ≥97% purity and were purchased from SigmaAldrich (Milwaukee, WI). Perfluorooctane sulfonamide (FOSA) was purchased from SynQuest Laboratories (Alachua, FL). Perfluorodecanesulfonate (PFDS), perfluorooctane sulfonamido acetate (FOSAA), N-methyl perfluorooctane sulfonamidoacetate (MeFOSAA), EtFOSAA, and all isotopically labeled internal standards (see Table S2, SI) were purchased from Wellington Laboratories (Guelph, ON, Canada). A solution of FC-807 commercial product containing SAmPAP 6506
dx.doi.org/10.1021/es300823m | Environ. Sci. Technol. 2012, 46, 6505−6514
Environmental Science & Technology
Article
Figure 2. Map of False Creek Harbor (Vancouver, BC, Canada) showing water (letters/green circles) and sediment (numbers/red circles) sampling locations. Orange circles represent stormwater outfalls, yellow circle indicates a permitted effluent outfall, and purple circles denote combined sewer overflows. Dotted lines denote local marinas while parallel lines represent bridges. Tidal flow was east to west during sampling.
Extraction, Treatment, and Analysis of Water and Sediment Samples. All seawater samples (1 L) and blanks were spiked with isotopically labeled internal standards (Table S2, SI) and extracted using a previously developed solid-phase extraction (SPE) method,36 details of which are provided in the SI. Samples were not filtered prior to extraction based on previous reports of chain-length dependent sorption of PFASs to glass fiber or nylon filters (45−60% sorption for PFDS, PFTriDA, and PFTeDA using Nylon and 25−30% sorption for PFUnDA, PFDoDA, PFTrDA, PFTeDA, and PFDS using glass fiber).37 In-house filter sorption tests have demonstrated that diPAPs are also prone to filter sorption (up to 60%, depending on chain length and filter material). Since the bias introduced by filtration is expected to be far greater than that potentially introduced by quantifying suspended solid PFAS concentrations as part of the aqueous phase, whole water was extracted. Sediments were dried in a desiccator, homogenized, then 4 g was transferred to a 50-mL polypropylene (PP) centrifuge tube and spiked with isotopically labeled internal standards (Table S2, SI). Sediment samples were extracted in triplicate using a method modified from Powley et al.38 (see SI for details). Analysis of extracts was accomplished by a previously developed high-performance liquid chromatography tandem mass spectrometry (HPLC-MS/MS) method39 which utilized a Dionex HPLC coupled to a API 5000Q triple quadrupole mass spectrometer (Applied Biosystems/Sciex, Concord, ON, Canada). Mass spectral data were collected under negative ion, multiple reaction monitoring (MRM) mode. Ions used for quantification and confirmation are stated in Table S2 (SI). Details of HPLC gradient conditions and optimized instrument parameters can be found in the SI.
diester at a concentration of 30% (w/v) in isopropanol/water was provided by Timothy Begley (U.S. Food and Drug Administration). Sample Collection. Samples were collected using a 14-ft skiff on August 3, 2011 from False Creek, a relatively small (4.0 × 0.3 km) and shallow (mean depth of 8 m) marine embayment of Burrard Inlet in the city of Vancouver, BC (population ∼612,000; ∼2.2 million in greater Vancouver). A total of 18 effluent outfalls exist in False Creek, including 1 permitted effluent, 2 sewer overflows, and 15 stormwater outfalls (Figure 2). In addition, the harbor contains 7 marinas, including houseboat mooring, and a mix of industrial, commercial, and residential buildings. Tides were ebbing (east to west flow) throughout the sampling trip (high tide 3.9 m at 08:48 hrs, low tide 1.5 m at 14:58 hrs). The sample area was divided into 7 study regions: east basin, inner harboreast, inner harbor-central, inner harbor-west, outer harbor, and entrance (Figure 2). Within each region, between 1 and 4 individual sediment samples were collected using a petit ponar and spoon (the latter rinsed with MeOH and HPLC-grade water between each use) and then transferred to high-density polyethylene (HDPE) bottles (top 0.5−1.0 cm layer, later split into triplicate). Subsurface water samples (1 L, n = 3 samples/ region;
