Supporting Information. Fluorinated Compounds in U.S. Fast Food Packaging

Transcription

1 Supporting Information Fluorinated Compounds in U.S. Fast Food Packaging Laurel A. Schaider,,* Simona A. Balan,,1 Arlene Blum,, David Q. Andrews, Mark J. Strynar, # Margaret E. Dickinson, David M. Lunderberg, Johnsie R. Lang, # and Graham F. Peaslee Silent Spring Institute, Newton, MA 02460, USA California Department of Toxic Substances Control, Sacramento, CA 95814, USA Green Science Policy Institute, Berkeley, CA 94709, USA Department of Chemistry, University of California at Berkeley, Berkeley, CA 94720, USA Environmental Working Group, Washington, DC 20009, USA # National Exposure Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711, USA Chemistry Department, Hope College, Holland, MI 49423, USA Oak Ridge Institute for Science and Education, Oak Ridge, TN 37831, USA Department of Physics, University of Notre Dame, Notre Dame, IN 46556, USA *Corresponding author. Phone: schaider@silentspring.org Content of Supporting Information: 25 pages (including cover sheet) 7 Tables 4 Figures 1 Formerly at the Green Science Policy Institute, Berkeley, CA 94709, USA 1

2 PIGE Total Fluorine Analyses Analytical methods Approximately 2-cm 2 pieces of each sample were cut and manually adhered to a stainless steel target ladder with a 1-cm diameter hole over which the thin target was taped. Nearly all food packaging samples were clean samples that had not come into contact with food. If any food residues were apparent on the packaging, we selected an area of the packaging that did not have food residues for the analysis. An inorganic standard target (NaF with 412 ppm F) was analyzed at the beginning and end of each batch of samples to normalize among batches. A blank target frame was measured for each batch to confirm a clean background. Each sample was irradiated with approximately 10 nanoamperes of 3.4 megaelectron-volts (MeV) protons for 180 seconds. The beam was extracted ex vacuo through a thin (8 µm) Kaptan foil and impinged on each target in air, through the center hole of each target frame. The characteristic gamma-rays emitted from the de-excitation of 19 F at 110 kev and 197 kev (Figure S1) had background-subtracted integrations summed for each sample irradiation. In order to transform the integrated gamma-ray counts in the 110 and 197 kev peaks per microcoulomb of beam on target (counts/µc) into a unit proportional to the quantity of fluorine applied per area of surface (nmol F/cm 2 ) the PIGE signal was calibrated with both thick inorganic standards and papers soaked in PFOA. These calibration methods are described in detail in Ritter et al. 1 A typical calibration curve on paper was constructed between 0 and 6000 counts/µc, yielding linear areal concentrations of fluorine between 0 and ~1000 nmol F/cm 2. Two additional 515 nmol F/cm 2 paper standards were also made using PFBA and PFHxA and yielded a similar signal compared to a PFOA standard of the equivalent concentration of F. This PIGE method for total fluorine detection has a limit of detection (LOD) of approximately 16 nmol F/cm 2 and a limit of quantification (LOQ) of ~50 nmol F/cm 2. The ion beam typically penetrates microns into a solid material. The emitted gamma-rays could be produced at any depths penetrated by the beam and exit the material without measurable attenuation because of their high energy. For wrapper samples, it is likely that the beam can completely penetrate the samples. We assessed this by analyzing stacks of up to 5 sheets of the same sample. With two sheets, the signal was always doubled. With three or four sheets, the signal typically increased by a factor of around 2.6, although for some samples 2

3 we observed a 3- or 4-fold increase in signal. These results indicate that the beam can penetrate at least two layers of wrappers, therefore it was not necessary to analyze both the front and back of each sample. Paperboard samples are much thicker, so it is likely that the beam is not able to penetrate through the entire sample. For thick paperboard samples and cups, we ran samples on both sides and selected the higher concentration in order to capture total fluorine signal associated with PFASs added as a surface treatment. QA/QC results Laboratory replicates. The stability of the PIGE signal for fluorine on treated papers was previously analyzed in Ritter et al., 1 in which a single popcorn bag sample was repeatedly measured to confirm that the signal does not degrade rapidly with time (and irradiation). In order to evaluate the precision of the total fluorine measurements in this survey, 5% of samples (22 of 407) were analyzed repeatedly (2-5 times), for a total of 22 samples and 45 replicate analyses (Table S2). These laboratory replicate results show high reproducibility of fluorine detection. For seven samples (total of 27 measurements), all measurements were below the LOD. For 13 samples (total of 32 measurements), all measurements were above the LOD, with some values falling between the LOD and LOQ. In these samples, concentrations ranged from ~18 to 363 nmol/cm 2 with an average relative standard deviation (RSD) among measurements for the same sample of 7% (range: 2-15%). Two samples (total of 8 measurements) had a mixture of measurements below the LOD and above the LOD but well below the LOQ (~18 23 nmol/cm 2 ), suggesting the presence of fluorine in these two samples at concentrations close to the LOD. Field replicates. Ten percent of samples (42 of 407) were collected with one or two field replicates (i.e., an identical sample collected at the same location and time). Of these, 21 samples (13 of which were collected in duplicate and 8 in triplicate, for a total of 50 analyses) were consistently below the LOD. Another 20 samples (19 of which were collected in duplicate and one of which was collected in triplicate, for a total of 41 analyses) were consistently above the LOD (Table S3). Among these 20 samples, the average relative percent difference (for N=2) or relative standard deviation (for N=3) was 17%, using estimated concentrations for measurements below the LOQ. Only one sample showed a discrepancy in fluorine detection between duplicates, with one replicate below the LOD and the other below the LOQ, with an estimated concentration (22 nmol F/cm 2 ) just above to the LOD. Overall, the results of these quality 3

4 assurance samples show good agreement on determining the presence or absence of F in the food packaging samples. In order to evaluate variations in F concentrations for similar products within a small geographical area, we collected identical pastry wrappers from 10 locations of the same fast food chain on the same day in the same city. When analyzed independently, these samples yielded an average of 465 ± 63 nmol F/cm 2, with an RSD of 14%. To evaluate the presence of fluorinated chemicals in white paper not treated for grease or water resistance, we analyzed 10 samples of white copier paper. All of these samples contained total F < LOD. PFAS Analyses Sample selection Twenty samples were selected for analysis of individual PFAS compounds: eleven wrapper samples and three paperboard samples with total F above 200 nmol F/cm 2 according to PIGE analysis, plus three wrapper samples and three paperboard samples that did not have detectable total F, for comparison. In selecting the 20 samples, we aimed to achieve an even distribution among the five main geographical regions and to include no more than two samples from each restaurant chain. Five samples (three wrappers and two paperboard samples with detectable total F) were analyzed in duplicate to assess the reproducibility of the results. Analytical methods Each sample was cut into an approximately 10 cm x 10 cm piece. The dimensions of each piece and its mass were recorded. Each piece was placed into a B&D Falcon 15 ml polypropylene centrifuge tube along with 20 ml of methanol containing 1 ng 13 C 4 -PFOS/mL. Samples were placed in a sonicator for 30 minutes and then centrifuged to pelletize the paper. 5 ml of supernatent was passed through a 3 cc Supelclean ENVI-Carb 250 mg solid phase extraction (SPE) cartridge (Supelco, Bellefonte, PA) that was twice pre-treated with 5 ml of methanol. The ENVI-Carb cartridge removes dyes and other extractable compounds in paper that are severe matrix interferents. Polyfluorinated compounds are not retained on the ENVI-Carb cartridges while chemicals of a planar nature are retained. The resulting eluates were evaporated to ~0.5 ml and prepared for analysis by mixing the methanol-based extracts with 2 mm ammonium acetate at a 60:40 ratio. These extracts were 4

5 analyzed for 89 PFASs with known structures and for unidentified polyfluorinated compounds using non-targeted analyses via liquid chromatography time of flight mass spectrometry (LC/TOF/MS) in negative mode on an Agilent 1100 series HPLC interfaced with a 6210 series Accurate-Mass LC-TOF system (Agilent Technologies, Palo Alto, CA). Five samples (25%) were analyzed in duplicate. We did not attempt to quantify concentrations of individual analytes, but instead used peak area as a semi-quantitative indication of abundance. Peak areas were normalized to a constant 100 cm 2 area to account for variations in the size of samples. Extracts were subjected to suspect screening based on accurate mass and spectral isotope clusters for 89 PFASs with chemical structures that have been previously characterized following the methods described in Rager et al. 2 In addition, all samples were screened for additional, unknown polyfluorinated compounds (series of compounds whose molecular weight differs by , the mass of a CF 2 group, and/or has a negative mass defect or homologous series). QA/QC results Solvent blank and method blank analysis was included in data analysis. Only compounds found to be present in the paper sample extracts that were not present in the blanks were considered. In one laboratory blank, only PFNA was detected, which was not detected in any samples. For the unknown polyfluorinated compounds, four blank samples were run. Peak areas of samples were blank-corrected by subtracting three times the average area for peaks of the same molecular weight detected in blank samples. As a means of sample normalization, 13 C 4 - PFOS was added to each sample prior to extraction. Non-targeted screening was accomplished to detect the spiked 13 C 4 -PFOS in each sample. The average scoring of the matching to the 13 C 4 - PFOS in each sample was 94 ± 10% (in four samples scoring match was <80%). Matching scores for suspect screening compounds were 66% to 99.8% compared to an internal personal compound database list (PCDL) for 89 per- and polyfluorinated compounds used for suspect screening matching. Eight perfluorinated compounds could be confirmed with authentic standards based on retention time (RT) and accurate mass matches in analytical batches that bracketed our samples (Table S5). Additional chemical matches were based solely on software matching of detected peaks with theoretical spectral features (monoisotopic mass, isotope spacing, and isotope intensity). 5

6 Background on Danish Ministry for Environment and Food s guideline for food packaging In 2015, the Danish Ministry for Environment and Food issued a guideline 3,4 for the total organic fluorine content in paper and board food contact materials of 0.35 µg fluorine/dm 2. This guideline was based on an evaluation of existing toxicological and epidemiological data and took into account pragmatic considerations such as the ease, availability, and cost of analytical techniques and existing limit values in European legislation. A single limit value set in the material was also considered to be simpler to enforce since the analytical result corresponds directly to the limit value and it is easier to communicate with suppliers. Following a review of available data, PFOA was selected as the basis for the guideline, since there is limited toxicological data for PFASs other than PFOA and PFOS, and perfluorocarboxylates and their precursors are among the families of PFASs that have been commonly detected in food packaging. PFOA was selected with the recognition that some PFASs may have greater toxicity and some may have lower toxicity. Based on a toxicological evaluation of existing studies, a recommended limit value for food packaging was set at 0.35 µg fluorine/dm 2, corresponding to 0.5 µg PFOA equivalents/dm 2, which under a set of assumptions would correspond to a tolerable daily intake from food of 5 µg PFOA equivalents/kg food. To convert from a concentration of PFOA in food to a concentration of organofluorine in food contact materials, several additional assumptions were made. One important consideration is the extent of PFAS migration from food packaging into food products. Migration can vary widely depending on the type of food, temperature, food moisture, and properties of each individual PFAS compound. In calculating the guideline, a migration efficiency of 100% was used, which is an overestimate in most cases since 10% or lower are more typical migration efficiencies. 5 The guideline also assumes 100% absorption of PFASs into the body, which also may be in overestimate in many cases. For instance, the uptake of 8:2/8:2 dipaps in rats and in humans is less than 10%. 6 The limit of detection for the total organofluorine methods, including a method blank value, * was estimated to be 0.35 µg fluorine / g paper, or 0.35 ppm. Using an approximate density of 1 g/dm 2, this is also equivalent to 0.35 µg fluorine / dm 2 paper. * Method blank value was calculated based on the water used throughout the analysis, since there was no blank paper material available at the time of the initial validation. 6

7 Converting from PFOA (molecular weight g/mol) to organofluorine (15 moles of F per mole PFOA, atomic weight of F g/mol), this is approximately equivalent to: 0.5 µg PFOA / dm 2 paper Using the conventional European Food Safety Authority (EFSA) assumption that a person eats 1 kg of food/day, and that 1 kg of food comes into contact with 10 dm 2 of paper,7 and assuming 100% migration of PFASs from packaging into food this is equivalent to: 5 µg PFOA equivalents / kg food Although this value was above German and U.S. toxicological guidelines for PFOA, it represents a sum value for all PFASs, and is 18 times lower than the 90 µg PFOA/kg food limit (based on EFSA s 2008 tolerable daily intake of 1.5 µg PFOA/kg body weight/day 8 for a 60 kg person). It was also noted that the assumptions of 100% migration, 100% uptake, and equivalent toxicity for all organofluorine compounds compared to PFOA would deliver sufficient safety margins by an estimated factor of The Danish guideline value for organofluorine in food contact materials is well below the LOD of the PIGE total fluorine method (approximately 60 µg/g in wrappers and 14 µg/g in paperboard). Thus, the Danish guideline is much more sensitive and may include both PFASs that are intentionally added to packaging and PFASs that are present in samples unintentionally, for instance through recycled content. Concentrations of dipaps of 600 9,000 µg/g in food packaging (~360 5,400 µg F/g paper) were attributed to intentional additions to paper pulp, and dipap concentrations of µg/g (~ µg F/g paper) were hypothesized to be associated with surface coatings. 3,5 Lower concentrations of dipaps that may come from background impurities may be above the Danish guideline but are not expected to be detected by the PIGE method. The average surface area to volume for paper and board is dm 2 /kg; however for the types of food typically packaged in PFAS coated packaging (fast food, cakes, muffins, popcorn bags, flour, müsli), an average of 10 dm 2 /kg was assumed. According to the recommendation, when calculating the actual µg PFOA/kg for a specific sample, the specific surface area to volume should be used. 7

8 Table S1. Per- and polyfluorinated compounds permissible in paper and paperboard food contact materials according to the U.S. Food and Drug Administration. 9 Under Limits/Specifications, Table 1 refers to certain categories of foods and Table 2 refers to certain conditions of use; both tables are provided on the U.S. FDA website. 10 FCN or FCS number or CFR section Name CAS Registration No. Company Intended Use Effective date Limits/Specifications propenoic acid, 2-methyl-, 2- hydroxyethyl ester, polymer with 2-propenoic acid and 3,3,4,4,5,5,6,6,7,7,8,8,8- tridecafluorooctyl 2-methyl-2- propenoate, sodium salt Asahi Glass Co., Ltd. AGC Chemicals Americas, Inc. As an oil, water and grease proofing agent in paper and paperboard, except for use in contact with infant formula and breast milk. 21-Sept-16 For use at a maximum level of 1.2% by weight of the finished paper, in contact with all food types under Conditions of Use B through H, as described in Table 2. The FCS is not for use in contact with infant formula and breast milk. Such uses were not included as part of the intended use of the substance in the FCN Copolymer of 2- (dimethylamino) ethyl methacrylate with 3,3,4,4,5,5,6,6,7,7,8,8,8- tridecafluorooctyl methacrylate, N-oxide, acetate Archroma Management GmbH As a grease resistant treatment employed either prior to or after the sheet forming operation for paper and paperboard intended to contact food, except for use in contact with infant formula and breast milk. 31-Dec-14 The FCS may be used at a level not to exceed 0.26 mg/in 2 paper on a polymer solids basis. The finished paper may be used in contact with all food types under Conditions of Use B through H as described in Tables 1 and 2. The FCS is not for use in contact with infant formula and breast milk. Such uses were not included as part of the intended use of the substance in the FCN (1360) 2-Propenoic acid, 2-methyl-, 2- (dimethylamino)ethyl ester, polymer with 1-ethenyl-2- pyrrolidinone and 3,3,4,4,5,5,6,6,7,7,8,8,8- tridecafluorooctyl 2-propenoate, acetate Daikin America, Inc. The FCS will be added at the size press or wet end to impart grease and oil resistance to paper and paperboard, except for use in contact with infant formula and breast milk. 4-Sep-14 The FCS will be used as an additive in paper and paperboard at levels not exceeding 1 percent of the dry fiber. The FCS is intended for use in paper that contacts all food types under Condition of Use J (including microwave susceptor applications). The FCS is not for use in contact with infant formula and breast milk. Such uses were not included as part of the intended use of the substance in the FCN Butanedioic acid, 2-methylene-, polymer with 2-hydroxyethyl, 2- methyl-2-propenoate, 2-methyl- 2-propenoic acid and 3,3,4,4,5,5,6,6,7,7,8,8,8- tridecafluorooctyl 2-methyl-2- propenoate, sodium salt Asahi Glass Co., Ltd. As an oil, grease, and waterresistant treatment for paper and paperboard. 21-Sep-12 For use at a level not to exceed 1.2 percent by weight of the finished paper. The finished paper may be used in contact with all types of foods under Conditions of Use B through H as described in Tables 1 and Hexane, 1,6-diisocyanato-, homopolymer, α-[1-[[[3-[[3 (dimethylamino)propyl]amino]p ropyl]amino]carbonyl]-1,2,2,2- tetrafluoroethyl]-ω-(1,1,2, 2,3,3,3-heptafluoropropoxy) poly[oxy[trifluoro(trifluorometh yl)-1,2-ethanediyl]]-blocked Archroma U.S., Inc. As a grease resistant treatment for paper and paperboard employed either prior to or after the sheet forming operation. 24-Aug-11 The FCS may be used at a level not to exceed 0.6 percent by weight (on a polymer basis) of dry paper and paperboard. The finished product is intended for use in contact with all food types, except foods containing greater than 15 percent alcohol, under Conditions of Use B through H as described in Tables 1 and 2. 8

9 propenoic acid, 2-methyl-, 2- hydroxyethyl ester polymer with 1-ethyenyl-2-pyrrolidinone, 2- propenoic acid and 3,3,4,4,5,5,6,6,7,7,8,8,8- tridecafluorooctyl 2-propenoate sodium salt Daikin America, Inc. The FCS will be added at the size press or prior to sheet formation to impart grease and oil resistance to paper and paperboard. 16-Feb-11 The FCS will be used as a polymer additive in paper and paperboard at levels not exceeding 1 percent of the dry fiber. The FCS is intended for use in paper that contacts all food types under Conditions of Use A-H and J (including microwave susceptor applications) as described in Table (885) 2-propenoic acid, 2-methyl-, polymer with 2- (diethylamino)ethyl 2-methyl-2- propenoate, 2-propenoic acid and 3,3,4,4,5,5,6,6,7,7,8,8,8- tridecafluorooctyl 2-methyl-2- propenoate, acetate The Chemours Company FC, LLC As an oil and grease resistant treatment for paper and paperboard employed either prior to or after the sheet forming operation. 12-Jan-11 The FCS may be used at a level not to exceed 0,8 percent by weight of dry paper when applied prior to the sheet forming operation, or 0.42 percent by weight of dry paper and paperboard when applied after sheet formation. The finished product is intended for use in contact with all food types under Conditions of Use B through H and Condition of Use J (microwave susceptor applications only) as described in Tables 1 and (416 & 195) Diphosphoric acid, polymers with ethoxylated reduced Me esters of reduced polymerized oxidized tetrafluoroethylene. This substance is also known as: phosphate esters of ethoxylated perfluoroether, prepared by reaction of ethoxylated perfluoroether diol (CAS Reg. No ) with phosphorous pentoxide (CAS Reg. No ) or pyrophosphoric acid (CAS Reg. No ) Solvay Specialty Polymers USA, LLC As a water and oil repellant in the manufacture of foodcontact paper and paperboard. 11-May-10 For use at levels not to exceed 1.0% by weight of finished dry paper or paperboard, in contact with all food types, in microwave susceptor applications. Also for use at levels not to exceed 1.5% by weight of finished dry paper or paperboard in contact with all food types, under Conditions of Use B through H, and J, as described in Table Hexane, 1,6-diisocyanato-, homopolymer, 3,3,4,4,5,5,6,6,7,7,8,8,8- tridecafluoro-1-octanol-blocked The Chemours Company FC, LLC As an oil and grease resistant treatment for paper and paperboard employed either prior to the sheet forming operation or at the size press. 03-Apr-10 The FCS may be used at a level not to exceed 0.18% by weight of dry paper and paperboard intended for use in contact with all food types under Conditions of Use B through H, as described in Tables 1 and propenoic acid, 2-methyl-, polymer with 2-hydroxyethyl 2- methyl-2-propenoate, α-(1-oxo- 2-propen-1-yl)-ω- hydroxypoly(oxy-1,2- ethanediyl) and 3,3,4,4,5,5,6,6,7,7,8,8,8- tridecafluorooctyl 2-propenoate, sodium salt Daikin America, Inc. As an oil and grease resistant treatment for paper and paperboard employed at the size press or prior to sheet formation. 30-Dec-09 The FCS may be used at a level not to exceed 0.8% by weight of dry paper and paperboard intended for use in contact with all food types under Condition of Use A through H and J (including microwave susceptor applications) as described in Tables 1 and 2. 9

10 888 (827) 2-propenoic acid, 2- hydroxyethyl ester, polymer with α-(1-oxo-2-propen-1-yl)-ω- hydroxypoly(oxy-1,2- ethanediyl), α-(1-oxo-2-propen- 1-yl)-ω-[(1-oxo-2-propen-1- yl)oxy]poly(oxy-1,2-ethanediyl) and 3,3,4,4,5,5,6,6,7,7,8,8,8- tridecafluorooctyl 2-propenoate Daikin America, Inc. As an oil and grease resistant treatment for paper and paperboard employed at the size press 18-Jun-09 The FCS may be used at a level not to exceed 0.4% by weight of dry paper and paperboard intended for use in contact with all food types under Condition of Use J (microwave susceptor applications only), Tables 1 and Propenoic acid, 3,3,4,4,5,5,6,6,7,7,8,8,8- tridecafluorooctyl ester, polymer with α-(1-oxo-2-propen-1-yl)-ω- hydroxypoly(oxy-1,2- ethanediyl). Daikin America, Inc. For use as an oil or grease resistant treatment for paper and paperboard intended for food-contact use. 31-Jul-08 The polymer additive is intended to be used in paper and paperboard at levels not exceeding 0.2% of the finished food-contact paper. 783 (746 & 542) 2-propen-1-ol, reaction products with 1,l,1,2,2,3,3,4,4,5,5,6,6- tridecafluoro-6-iodohexane, dehydroiodinated, reaction products with epichlorohydrin and triethylenetetramine as manufactured in accordance with the description in Hercules Incorporated As an oil/grease resistant sizing agent employed either prior to the sheet forming operation and/or at the size press in the manufacture of paper and paperboard. 06-Mar-08 For use at levels not to exceed 0.75% by weight of dry paper and paperboard for food-contact under Condition of Use J (microwave heat-susceptor packaging applications only), as described in Table (599) Copolymer of perfluorohexylethyl methacrylate, 2-N,Ndiethylaminoethyl methacrylate, 2-hydroxyethyl methacrylate, and 2,2'-ethylenedioxydiethyl dimethacrylate, acetic acid salt Asahi Glass Company, Ltd. (Manufacturer) AGC Chemicals Americas, Incorporated As an oil, grease, and water resistant treatment for paper and paperboard employed either prior to the sheet forming operation or at the size press. 05-Aug-06 The FCS may be used at a level not to exceed 1.2% by weight of dry paper and paperboard intended for use in microwave susceptor applications in contact with all food types. 604 (599) Copolymer of perfluorohexylethyl methacrylate, 2-N,Ndiethylaminoethyl methacrylate, 2-hydroxyethyl methacrylate, and 2,2'-ethylenedioxydiethyl dimethacrylate, malic acid salt Asahi Glass Company, Ltd. (Manufacturer) AGC Chemicals Americas, Incorporated As an oil, grease, and water resistant treatment for paper and paperboard employed either prior to the sheet forming operation or at the size press. 05-Aug-06 The FCS may be used at a level not to exceed 1.2% by weight of dry paper and paperboard intended for use in contact with all food types under Conditions of Use B through H as defined in Tables 1 and (398) Perfluoropolyether dicarboxylic acid (CAS Reg. No ), ammonium salt Solvay Specialty Polymers Italy S.p.A. As an oil and water repellent employed in the manufacture of food-contact paper and paperboard either prior to the sheet-forming operation or at the size press. 19-Nov-05 When applied at the size press the total use level of the FCS is not to exceed 0.5% by weight of the finished dry paper and paperboard to be used in contact with aqueous foods and 1% by weight of the finished dry paper and paperboard to be used in contact with all other food types. 10

11 518 (487, 314) 2-propen-1-ol, reaction products with pentafluoroiodoethanetetrafluoroethylene telomer, dehydroiodinated, reaction products with epichlorohydrin and triethylenetetramine (CAS Reg. No ) Hercules Inc. As an oil/grease resistant sizing agent at levels up to 0.5 percent by weight employed either prior to the sheet forming operation or at the size press for paper and paperboard intended for use in microwave heat-susceptor packaging. 20-Oct-05 The FCS may be used in contact with all food types, as described in Table Fluorinated polyurethane anionic resin prepared by reacting perfluoropolyether diol (CAS Reg. No ), isophorone diisocyanate (CAS Reg. No ), 2,2- dimethylolpropionic acid (CAS Reg. No ), and triethylamine (CAS Reg. No ) Solvay Specialty Polymers Italy S.p.A. As a water and oil repellent in the manufacture of paper and paperboard. 23-Mar-02 The FCS will be used in the manufacture of paper and paperboard, at a level not to exceed 1.5% by weight of the finished dry paper and paperboard. Paper containing the FCS may contact all food types under Conditions of Use B through H, as described in Tables 1 and 2 respectively Chromium (Cr III) complex of N-ethyl - N -heptadecylfluorooctane sulfonyl glycine containing up to 20 percent by weight of the chromium (Cr III) complex of heptadecylfluorooctane sulfonic acid may be safely used as a component of paper for packaging dry food when used in accordance with the following prescribed conditions The chromium (Cr III) complex of N-ethyl - N - heptadecylfluoro-octane sulfonyl glycine containing up to 20% by weight of the chromium (Cr III) complex of heptadecylfluoro-octane sulfonic acid may be safely used as a component of paper for packaging dry food when used in accordance with the following prescribed conditions. (a) The food additive is used as a component of paper in an amount not to exceed 0.5% by weight of the paper. (b)(1) The food-contact surface of the paper is overcoated with a polymeric or resinous coating at least 1 3 -mil in thickness, that meets the provision of ; or (2) The treated paper forms one or more plies of a paper in a multiwall bag and is separated from the food by at least one ply of packaging films or greaseresistant papers which serves as a functional barrier between the food additive and the food. Such packaging films or grease-resistant papers conform with appropriate food additive regulations. (c) The labeling of the food additive shall contain adequate directions for its use to insure compliance with the requirements of paragraphs (a) and (b) of this section. 11

12 ** Ammonium bis(n-ethyl-2- perfluoroalkylsulfonamido ethyl) phosphates, containing not more than 15% ammonium mono (N-ethyl-2- perfluoroalkylsulfonamido ethyl) phosphates, where the alkyl group is more than 95% C8 and the salts have a fluorine content of 50.2% to 52.8% as determined on a solids basis ** Perfluoroalkyl acrylate copolymer (CAS Reg. No ) containing 35 to 40 weight percent fluorine, produced by the copolymerization of ethanaminium, N,N,N-trimethyl- 2-[(2-methyl-1-oxo-2-propenyl)- oxy]-, chloride; 2-propenoic acid, 2-methyl-, oxiranylmethyl ester; 2-propenoic acid, 2- ethoxyethyl ester; and 2- propenoic acid, 2[[(heptadeca fluoro- octyl)sulfonyl] methyl amino]ethyl ester Undecafluorocyclohexanemetha nol ester mixture of dihydrogen phosphate, compound with 2,2 iminodiethanol (1:1); hydrogen phosphate, compound with 2,2 - iminodiethanol (1:1); and P,P - dihydrogen pyrophosphate, compound with 2,2 - iminodiethanol (1:2); where the ester mixture has a fluorine content of 48.3 pct to 53.1 pct as determined on a solids basis For use only as an oil and water repellant at a level not to exceed 0.17 pound (0.09 pound of fluorine) per 1,000 square feet of treated paper or paperboard of a sheet basis weight of 100 pounds or less per 3,000 square feet of paper or paperboard, and at a level not to exceed 0.5 pound (0.26 pound of fluorine) per 1,000 square feet of treated paper or paperboard having a sheet basis weight greater than 100 lb. per 3,000 square feet as determined by analysis for total fluorine in the treated paper or paperboard without correction for any fluorine that might be present in the untreated paper or paperboard, when such paper or paperboard is used as follows: 1. In contact, under conditions of use C, D, E, F, G, or H described in table 2 of paragraph (c) of this section, with nonalcoholic food. 2. In contact with bakery products of Type VII, VIII, and IX described in table I of paragraph (c) of this section under good manufacturing practices of commercial and institutional baking. For use only as an oil and water repellent at a level not to exceed 0.5 percent by weight of the finished paper and paperboard in contact with nonalcoholic foods under conditions of use C, D, E, F, G, or H described in table 2 of paragraph (c) of this section. For use only as an oil repellent at a level not to exceed lb (0.046 lb of fluorine) per 1,000 ft 2 of treated paper or paperboard, as determined by analysis for total fluorine in the treated paper or paperboard without correction for any fluorine which might be present in the untreated paper or paperboard, when such paper or paperboard is used in contact with food only of the types identified in paragraph (c) of this section, table 1, under Types IVA, V, VIIA, VIII, and IX, and under the conditions of use B through G described in table 2 of paragraph (c) of this section. 12

13 * Diethanolamine salts of monoand bis (1 H, 1 H, 2 H, 2 H perfluoroalkyl) phosphates where the alkyl group is evennumbered in the range C8-C18 and the salts have a fluorine content of 52.4 percent to 54.4 percent as determined on a solids basis * Pentanoic acid, 4,4-bis [(gamma-omega-perfluoro-c8-20-alkyl)thio] derivatives, compounds with diethanolamine (CAS Reg. No ) * Perfluoroalkyl substituted phosphate ester acids, ammonium salts formed by the reaction of 2,2-bis[([gamma], [omega]-perfluoro C4-20 alkylthio) methyl]-1,3- propanediol, polyphosphoric acid and ammonium hydroxide. *On January 4 th, 2016 FDA amended the food additive regulations to remove these three specific substances. **On November 22, 2016, the U.S. FDA amended the food additive regulations to remove these two specific substances. 13

14 Table S2. Results for field replicates (collected at the same location and time) with at least one detection of total F above the LOD according to PIGE analyses. For 9 food contact papers, 7 cups, and 5 paperboard samples collected in either duplicate or triplicate, all replicates were below the LOD. %RSD = percent relative standard deviation. Results in nmol F/cm 2. sample type Replicate 1 Replicate 2 Replicate 3 %RSD paper < paper % paperboard % paper % paper % paper % paper % paper % paper % paper % paper % paper % paper % paper % paper % paper % paper % paper % paper % paper % paper % 14

15 Table S3. Results of multiple PIGE analyses (N = 2 5) on the same sample of food contact paper or paperboard. %RSD = percent relative standard deviation. Results in nmol F/cm 2. sample type Run 1 Run 2 Run 3 Run 4 Run 5 %RSD paperboard <LOD <LOD paper <LOD <LOD paper <LOD <LOD <LOD paperboard <LOD <LOD <LOD <LOD <LOD paper <LOD <LOD <LOD <LOD <LOD paper <LOD <LOD <LOD <LOD <LOD paper <LOD <LOD <LOD <LOD <LOD paper <LOD <LOD 18.4 paperboard <LOD <LOD <LOD paperboard % paperboard % paperboard % paper % paper % paper % paper % paper % paper % paper % paper % paper % paperboard % 15

16 Table S4. Total fluorine concentrations based on PIGE analysis and total peak areas of PFASs with known structures and unknown polyfluorinated compounds identified in LC/TOF/MS analyses of methanol extracts of 20 fast food packaging samples. Five samples were analyzed in duplicate (identified as dup ). -- = no peaks detected. Total F Sample ID (nmol F/cm 2 ) Food contact papers (e.g., wrappers) sum peak areas for knowns sum peak areas for unknowns Count of knowns PFASs P1 <LOD 10,962 66,699 1 P2 <LOD 37,015 5,707 2 P3 <LOD 157,613 9,918 1 P , P ,484 7,139,713 2 P , ,155 7 P ,037 35,846 1 P8 dup ,649 25,205 1 P ,404 19,030 1 P ,758 0 P6 dup ,514 74,608 1 P ,223, ,681 9 P14 dup ,580, ,516 9 P ,058, ,239,617 8 P ,378 0 P , ,093 2 P ,693 78,263 1 Paperboard B1 <LOD 19, ,618 1 B2 <LOD 122, ,603 1 B3 <LOD 158, ,233 2 B ,510, ,411,469 5 B6 dup 278 2,849, ,427,896 5 B , ,866 7 B ,169, ,052 3 B5 dup 412 2,228, ,

17 Table S5. PFASs identified in LC/TOF/MS analyses of methanol extracts of 20 fast food packaging samples. Compound Perfluorocarboxylic acids (PFCAs) Formula Confirmed with authentic standard # of times detected (out of 20) PFBA (perfluorobutanoic acid) C4 H F7 O2 Yes 2 PFPeA (perfluoropentanoic acid) C5 H F9 O2 Yes 5 PFHxA (perfluorohexanoic acid) C6 H F11 O2 Yes 7 PFHpA (perfluoroheptanoic acid) C7 H F13 O2 Yes 1 PFOA (perfluorooctanoic acid) C8 H F15 O2 Yes 6 PFDA (perfluorodecanoic acid) C10 H F19 O2 Yes 1 PFUnA (perfluoroundecanoic acid) C11 H F21 O2 1 PFDoA(perfluorododecanoic acid) C12 H F23 O2 1 PFTriA (perfluorotridecanoic acid) C13 H F25 O2 1 Perfluorosulfonic acids (PFSAs) PFBS (perfluorobutanesulfonic acid) C4 H F9 S O3 Yes 7 PFHxS (perfluorohexanesulfonic acid) C6 H F11 S O3 Yes 1 Fluorotelomer sulfonates (FTSs) 4:2 FTS C6 H5 F9 S O3 2 6:2 FTS C8 H5 F13 S O3 2 10:2 FTS C10 H5 F17 S O3 2 Polyfluoroalkyl phosphate diesters (dipaps) 6:2-8:2-diPAP C18 H9 F30 O4 P 2 8:2-8:2-diPAP C20 H9 F34 O4 P 2 Polyfluorinated ethers (PF ethers) C5 polyfluoro ether C5 H F9 O5 1 C6 polyfluoro ether C6 H F11 O6 3 17

18 Table S5 (continued) Other compounds Compound Formula Confirmed with authentic standard # of times detected (out of 20) 5:3 FTCA (fluorotelomer carboxylic acid) C8 H5 F11 O2 2 PFHxPA (perfluorohexylphosphonic acid) C6 H2 F13 O3 P 2 6:6 FTMAP (fluorotelomer mercaptoalkyl C19 H13 F26 O4 P phosphate diester congener) S2 1 ADONA (ammonium 4,8-dioxa-3Hperfluorononanoate) C7 H2 F12 O4 2 FHUEA (2H-perfluoro-2-octenoic acid) C8 H2 F12 O2 6 n-me-fbse (N-methylperfluorobutane sulfonamidoethanol) C7 H8 F9 N O3 S 1 Nafion CA derivative C8 H2 F14 O4 2 PFOSulfinate C8 H F17 O2 S 3 GenX (perfluoro-2-propoxypropanoic acid) C6 H F11 O3 2 18

19 Table S6. Comparison of percentage of food contact paper and paperboard samples with detectable F (>16 nmol F/cm 2 ) across five regions of the U.S. (N=323), along with maximum concentrations. This table does not include other categories of packaging (e.g. paper cups) where samples were not collected in all five regions, nor samples that were collected outside these five regions (84 samples not included). State N Food contact paper detection frequency maximum total F (nmol F/cm 2 ) N Food contact paperboard detection frequency maximum total F (nmol F/cm 2 ) California 38 55% % 26 DC Metro 41 54% % 280 Massachusetts 59 46% % 32 Michigan 79 41% % 412 Washington State 26 38% %

20 Table S7. Odds ratios (with 95% confidence intervals) for comparisons among (a) types of packaging and (b) states based on binomial logistic regressions. For example, an odds ratio of 2 indicates that samples in that category have two times higher odds of containing fluorine than the reference group. For cups, non-contact papers, and miscellaneous samples, the detection frequencies were 0%, so odds ratios could not be estimated. = <0.1, * = p<0.05, ** = p<0.01, *** = p< (a) Comparisons among types of packaging, adjusted for region Reference group sandwich/burger Mexican dessert/bread paperboard Mexican 1.99 ( ) dessert/bread 2.07 ( )* 1.04 ( ) paperboard 0.38 ( )** 0.19 ( )*** 0.18 ( )*** -- other beverage 0.29 ( ) 0.14 ( )** 0.14 ( )** 0.76 ( ) (b) Comparisons among states, adjusted for type of packaging Reference group California DC Metro Mass. Michigan DC Metro 1.65 ( ) Mass ( ) 0.49 ( ) Michigan 0.8 ( ) 0.49 ( ) 1 ( ) -- WA state 1.02 ( ) 0.62 ( ) 1.27 ( ) 1.27 ( ) 20

21 19 F 110 kev PFAS-treated paper Non-treated paper 19 F 197 kev Figure S1. Typical PIGE spectra of a PFAS-treated paper from a microwave popcorn bag (red) compared with an untreated bond paper (black). Two characteristic 19 F peaks are identified, and a small peak is visible between them, arising from beam interactions with lead shielding upstream. The total fluorine measurement of this sample yielded a value of ~800 counts/µc (or ~130 nmol F/cm 2 ). 21

22 Replicate 1 (Counts/μC) Figure S2. Reproducibility of PIGE results in field duplicates (two independent samples collected at the same location and time). 22

23 Peak area (Thousands) 1, Total F < 16 nmolf/cm 2 Total F > 200 nmolf/cm 2 PFCAs PFSAs FTSs dipaps PF ethers Other Approx. LOD for PIGE * * * * P1 P2 P3 B1 B2 B3 P4 P5 P6 P7 P8 P9 P10P11P12P13 P14 B4 B5 B6 Figure S3. Total peak areas for six classes of known PFAS compounds based on LC/TOF/MS analyses of 14 paper wrapper (P) and 6 paperboard (B) samples. Samples numbered 1-3 (green background) had total fluorine concentrations below the detection limit according to PIGE. Samples numbered 4 and above (orange background) had total fluorine >200 nmol F/cm 2. Individual chemicals detected in each category are listed in Table S5. Five samples (P6, P8, P14, B5, B6) were analyzed in duplicate. Total peak areas for four samples were above the maximum value on the y-axis (denoted with ). The horizontal line shows the maximum total peak area (sum of known and unknown PFASs) among six samples with total F <LOD for PIGE (P1-P3, B1-B3). 23

24 1,000 Total F < 16 nmolf/cm 2 Total F > 200 nmolf/cm 2 * * * * * Peak area (Thousands) Knowns Unknowns Approx. LOD for PIGE P1 P2 P3 B1 B2 B3 P4 P5 P6 P7 P8 P9 P10P11P12P13 P14 B4 B5 B6 Figure S4. Total peak areas for known and unknown PFASs based on from LC/TOF/MS analyses of 14 paper wrapper (P) and 6 paperboard (B) samples. Samples numbered 1-3 (green background) had total fluorine concentrations below the detection limit according to PIGE and samples numbered 4 and above (orange background) had total fluorine >200 nmol F/dm 2. The names of 27 PFASs with known structures are listed in Table S4. The unknown polyfluorinated compounds were classified on the basis of series of compounds that differed in mass by (-CF 2 ). Five samples (P6, P8, P14, B5, B6) were analyzed in duplicate. Total peak areas for four samples were above the maximum value on the y-axis (denoted with ). The horizontal line shows the maximum total peak area (sum of known and unknown PFASs) among six samples with total F <LOD for PIGE (P1-P3, B1-B3). 24

25 References 1. Ritter, E. E.; Dickinson, M. E.; Harron, J. P.; Lunderberg, D. M.; DeYoung, P. A.; Roebel, A. E.; Field, J. A.; Peaslee, G. F., Novel total fluorine detection method for papers and textiles in commerce. To be submitted for publication, Environ. Sci. Technol. Lett. 2. Rager, J. E.; Strynar, M. J.; Liang, S.; McMahen, R. L.; Richard, A. M.; Grulke, C. M.; Wambaugh, J. F.; Isaacs, K. K.; Judson, R.; Williams, A. J.; Sobus, J. R., Linking high resolution mass spectrometry data with exposure and toxicity forecasts to advance highthroughput environmental monitoring. Environ. Int. 2016, 88, Technical University of Denmark. Proposal for Maximum Levels for Total Organic Fluorine in Paper and Cardboard Food Packaging (in Danish); Kongens Lyngby, Denmark, Danish Environmental Protection Agency. Short-chain Polyfluoroalkyl Substances (PFAS): A literature review of information on human health effects and environmental fate and effect aspects of short-chain PFAS; Environmental Project No. 1707; Danish Ministry of the Environment: Copenhagen, Trier, X. Polyfluorinated surfactants in food packaging of paper and board in Department of Basic Sciences and Environment. Copenhagen: Copenhagen, D'Eon, J. C.; Mabury, S. A., Exploring indirect sources of human exposure to perfluoroalkyl carboxylates (PFCAs): Evaluating uptake, elimination, and biotransformation of polyfluoroalkyl phosphate esters (PAPs) in the rat. Environ. Health Perspect. 2011, 119 (3), VKM (Norwegian Scientific Committee for Food Safety). Evaluation of the EU exposure model for migration from food contact materials (FCM): Opinion of the Panel on Food Additives, Flavourings, Processing Aids, Materials in Contact with Food and Cosmetics of the Norwegian Scientific Committee for Food Safety 06/406-5 final; Oslo, Norway, EFSA (European Food Safety Authority), Perfluorooctane sulfonate (PFOS), perfluorooctanoic acid (PFOA) and their salts: Scientific opinion of the panel on contaminants in the food chain. The EFSA Journal 2008, 653, Inventory of Effective Food Contact Substance (FCS) Notifications. United States Food and Drug Administration: Silver Spring, MD ( (Accessed 9 July 2016). 10. Food Types & Conditions of Use for Food Contact Substances. United States Food and Drug Administration: Silver Spring, MD ( onsofuse/default.htm) (Accessed 15 Nov 2016). 25