Document 8584YKK2ggR3q02mNa6eKj2RB

ARW6-or?s I BEST COPY AVAILABLE iI Laboratory Composite Report Analytical Reports o f Data for Fluorochemical Analysis in Human Sera LIMSNo. 1623 Testing Laboratory 3M Environmental Technology & Safety Services 3M Environmental Laboratory Fluorine Analytical Chemistry Team (FACT) 2-3E-09 935 Bush Avenue, St. Paul, MN 55106 Laboratory Contact Kris Hansen, Ph D. Bldg. 2-3E-09 P.O. Box 3331 St. Paul, MN 55133-3331 Phone: (651)778-6018 FAX: (651)778-6176 Requesters Larry Zobel, JeffMandel, Geary Olsen 3M Medical Department Bldg. 220 St. Paul, MN Page 1 004345 3MENVIRONMENTAL LABORATORY BEST copy AVAILABLE 2 Sum m ary This composite report includes ten technical reports that were issued to the 3M Medical Department btween March 2,1998 and April 24,1998. Each report documents the results of analysis of human sera samples collected from various sources. In most cases, each technical report includes information about the date of delivery o f the data to the requester. Interpretation of the results is the responsibility of the 3M Medical Departm ent Although rigorous quality control measures and 3M Environmental Laboratory Standard Operating Procedures and methods were followed when possible, the acquisition o f this data was not necessarily collected according to applicable Good Laboratory Practices. Data was collected to meet immediate needs to help characterize the potential for human health concerns; it was not always possible to follow laboratory procedures that were in place at the time. Data was technically reviewed by 3M Environmental Lab technical staff but was not reviewed by the Quality Assurance Unit in place at 3M Appendix A includes the analytical methods that were followed in the collection ofthe data supporting these technical reports. ETS-8-4,1 Extraction of Potassium Perfluorooctanesulfonate or other Fluorochemical Compounds from Serum for Analysis Using HPLC-Electrosprav/Mass Spectrometry with the following exception: ethyl acetate was added to the aqueous extract instead ofMTBE. ETS-8-5.1 Analysis of Potassium Perfluorooctanesulfonate or other Fluorochemical Compounds in Serum Extracts Using HPLC-Electrosnrav/Mass Spectrometry. Data included in the Phase n and Phase HI reports was collected using ESMS, while data for subsequent phases were collected using ESMSMS. Details can be found in each technical report, included in Appendices B through K. Chromatographic data for each technical report is archived at the 3M Environmental Lab. h -- -- ____________ Kris Hansen, PhD ., Principal Analytical Investigator 3 Appendices Appendix A: Analytical Methods AppendixB: Phase 2 Appendix C: Phase 3 Appendix D: Phase 4 Appendix E: Phase 5 Appendix F: Phase 6 Appendix G: Phase 7 Appendix H: Phase 8 Appendix I: Phase 9, part 1 Appendix J: Phase 9, part 2 Appendix K: Summary ofPOAA levels for Phases 2-9 Page 2 lap D ate '/iv*'fr 004346 3M Enviro nm ental L abo rato ry M ethod Ex tra c tio n of P o tassium P erfluo ro o ctanesulfo nate o r O th er Fluorochem ical com pounds from Serum for A n a ly sis U sing H P L C - Electrospray/M a ss Spectrom etry Method Number: ETS-8-4.1 V Author: Lisa Clemen, Glenn Langenburg Adoption Date: 03/01/99 Revision Date: Approved By: Laboratory Manager Date Group Leader Date Technical Reviewer Date 1.0 Scope and Application 1.1 Scope: This method is for the extraction of potassium perfluorooctanesulfonate (PFOS) or other fluorochemical compounds from serum. 1.2 Applicable compounds: Fluorochemical surfactants or other fluorinated compounds. 1.3 Matrices: Rabbit, rat, bovine, monkey, and human serum or other fluids as designated in the validation report. 004347 Word 6/95 ETS-8-4.1 Extraction of PFOS from Serum Page I of 14 2.0 Su m m a r y o f M eth o d 2.1 This method describes the procedure for extracting potassium perfluorooctanesulfonate (PFOS) or other fluorochemical surfactants from serum, or other fluids, using an ion pairing reagent and methyl-ferf-butyl ether (MtBE). In this method, seven fluorochemicals were extracted: PFOS, PFOSA, PFOSAA, EtFOSE-OH, PFOSEA, M556, and surrogate standard (see 3.0 Definitions). An ion pairing reagent is added to the sample and the analyte ion pair is partitioned into MtBE. The MtBE extract is removed and put onto a nitrogen evaporator until dry. Each extract is reconstituted in 1.0 mL of methanol, then filtered through a 3 cc plastic syringe attached to a 0.2 pm nylon filter into glass autovials. 2.2 These sample extracts are analyzed following method ETS-8-5.1 or other appropriate method. 3.0 Definitions_______ |________________________________________________________ 3.1 PFOS: perfluorooctanesulfonate (anion of potassium salt) CgFnSCV 3.2 PFOSA: perfluorooctane sulfonylamide C8F17SO2NH2 3.3 PFOSAA: perfluorooctane sulfonylamido (ethyl)acetate CgF17S02N(CH2CH3)CH2C02` 3.4 EtFOSE-OH: 2(N-ethylperfluorooctane sulfonamido)-ethyl alcohol C8Fi7S02N(CH2CH3)CH2CH20H 3.5 PFOSEA: perfluorooctane sulfonyl ethylamide CgFi7S02N(CH2CH3)H 3.6 M556: C8FnS02N(H)(CH2C00H) .............. 3.7 Surrogate standard: 1H-1H-2H-2H perfluorooctane sulfonic acid 4.0 Warnings and Cautions________________________________ _ _________________... 4.1 Health and safety warnings 4.1.1 Use universal precautions, especially laboratory coats, goggles, and gloves when handling animal tissue, which may contain pathogens. 5.0 Interferences_____________________________________________________________ 5.1 There are no interferences known at this time. 6.0 Equipment_________________________________________________________________ 6.1 The following equipment is used while performing this method. Equivalent equipment is acceptable. 6.1.1 Vortex mixer, VWR, Vortex Genie 2 6.1.2 Centrifuge, Mistral 1000 or IEC 6.1.3 Shaker, Eberbach or VWR 004348 6.1.4 Nitrogen evaporator, Organomation ETS-8-4.1 Extraction of PFOS from Serum Page 2 of 14 6.1.5 Balance ( 0.100 g) 7.0 Supplies and Materials_____________________________ _________________ _ 7.1 Gloves 7.2 Eppendorf or disposable pipettes 7.3 Nalgene bottles, capable of holding 250 mL and 1 L 7.4 Volumetric flasks, glass, type A 7.5 I-CHEM vials, glass, 40 mL glass 7.6 Centrifuge tubes, polypropylene, 15 mL 7.7 Labels 7.8 Oxford Dispenser-O.O to 10.0 mL 7.9 Syringes, capable of measuring 5 pL to 50 pL 7.10 Graduated pipettes 7.11 Syringes, disposable plastic, 3 cc 7.12 Syringe filters, nylon, 0.2 pm, 25 mm 7.13 Timer 7.14 Crimp cap autovials and caps 7.15 Crimpers Note: Prior to using glassware and bottles, rinse 3 times with methanol and 3 times with Milli-QTMwater. Rinse syringes a minimum of 9 times with methanol, 3 rinses from 3 separate vials. 8.0 Reagents and Standards 8.1 Type I reagent grade water, Milli-QTM or equivalent; all water used in this method should be Milli-QTM water and may be provided by a Milli-Q TOC PlusTM system 8.2 Sodium hydroxide (NaOH), J.T Baker or equivalent 8.3 Tetrabutylammonium hydrogen sulfate(TBA), Kodak or equivalent 8.4 Sodium carbonate (Na2C03), J.T. Baker or equivalent 8.5 Sodium bicarbonate (NaHCOs), J.T. Baker or equivalent 8.6 Methyl-T-Butyl Ether, Omnisolv, glass distilled or HPLC grade 8.7 Methanol, Omnisolv, glass distilled or HPLC grade 8.8 Serum or blood, frozen from supplier 8.9 Fluorochemical standards 8.9.1 PFOS (3M Specialty Chemical Division), molecular weight = 538 8.9.2 PFOSA (3M Specialty Chemical Division), molecular weight = 499 004349 8.9.3 PFOSAA (3M Specialty Chemical Division), molecular weight = 585 ET S-W .l Extraction of PFOS from Serum Page 3 of 14 I 8.9.4 EtFOSE-OH (3M Specialty Chemical Division), molecular weight = 570 8.9.5 PFOSEA (3M Specialty Chemical Division), molecular weight = 527 8.9.6 M556 (3M Specialty Chemical Division), molecular weight = 557 8.9.7 Surrogate standard: 4-H, perfluorooctane sulfonic acid (1-H,1-H, 2-H, 2-H CsFi3S0 3H) molecular weight = 428 8.9.8 Other fluorochemicals, as appropriate 8.10 Reagent preparation NOTE: When preparing larger volumes than listed in reagent, standard, or surrogate preparation, adjust accordingly. 8.10.1 ION sodium hydroxide (NaOH): Weigh approximately 200 g NaOH. Pour into a 1000 mL beaker containing 500 mL Milli-Q water, mix until all solids are dissolved. Store in a 1 L Nalgene bottle. 8.10.2 1 N sodium hydroxide (NaOH): Dilute ION NaOH 1:10. Measure 10 mL of 10 N NaOH solution into a 100 mL volumetric flask and dilute to volume using MilliQTM water. Store in a 125 mL Nalgene bottle. 8.10.3 0.5 M tetrabutylammonium hydrogen sulfate (TBA): Weigh approximately 169 g of TBA into a 1 L volumetric containing 500 mL Milli-QTM water. Adjust to pH 10 using approximately 44 to 54 mL of 10 N NaOH (While adding the last mL of NaOH, add slowly because the pH changes abruptly). Dilute to volume with MilliQTM water. Store in a 1 L Nalgene bottle. 8.10.3.1 TBA requires a check prior to each use to ensure pH - 10. Adjust as needed using 1 N NaOH solution. 8.10.4 0.25 M sodium carbonate/sodium bicarbonate buffer (Na2C0 3/NaHC0 3): Weigh approximately 26.5 g of sodium carbonate (Na2C0 3) and 21.0-g of sodium bicarbonate (NaHC03) into a 1 L volumetric flask and bring to volume with Milli-. QTM water. Store in a 1 L Nalgene bottle. 8.11 Standards preparation 8.11.1 Prepare PFOS standards for the standard curve. 8.11.2 Prepare other fluorochemical standards, as appropriate. Multicomponent fluorochemical standards are acceptable (for example, one working standard solution containing 1.00 ppm PFOS, 1.02 ppm PFOSA, 0.987 ppm PFOSAA, and 1.10 ppm EtFOSE-OH.) 8.11.3 Weigh approximately 100 mg of PFOS into a 100 mL volumetric flask and record the actual weight. 8.11.4 Bring to volume with methanol for a stock standard of approximately 1000 ppm (ng/mL). 8.11.5 Dilute the stock solution with methanol for a working standard 1 solution of approximately 50 ppm. 8.11.6 Dilute working standard 1 with methanol for a working standard 2 solution of approx. 5.0 ppm. 004350 i ETS-8-4.1 Page 4 of 14 Extraction of PFOS from Serum 8.11.7 Dilute working standard 1 with methanol for a working standard 3 solution of approx. 0.50 ppm. 8.12 Surrogate stock standard preparation 8.12.1 Weigh approximately 50-60 mg of surrogate standard 1-H,1-H, 2-H, 2-H, C8F13SO3H into a 50 mL volumetric flask and record the actual weight. 8.12.2 Bring to volume with methanol for a surrogate stock of approximately 1000-1200 ppm. 8.12.3 Prepare a surrogate working standard. Transfer approximately 1 mL of surrogate stock to a 10 mL volumetric flask and bring to volume with methanol for a working standard of 100 ppm. Record the actual volume transferred. 9.0 Sample Handling________________________________________ ;_________________ 9.1 All samples are received frozen and must be kept frozen until the extraction is performed. 9.2 Allow samples to thaw to room temperature prior to extraction. 10.0 Quality Control________________________________________________________ 10.1 Solvent Blanks, Method blanks and matrix blanks 10.1.1 An aliquot of 1.0 mL methanol is used as a solvent blank. 10.1.2 Extract two 1.0 mL aliquots of Milli-QTM water following this procedure and use as method blanks. 10.1.3 Extract two 1.0 mL aliquots of the serum following this procedure and use as matrix blanks. See 11.1.4, 10.2 M atrix spikes 10.2.1 Prepare and analyze matrix spike and matrix spike duplicate samples to determine the accuracy of the extraction. 10.2.2 Prepare each spike using a sample chosen by the analyst, usually the control matrix received with each sample set. 10.2.3 Expected concentrations will fall in the mid-range of the initial calibration curve. Additional spikes may be included and may fall in the low-range of the initial calibration curve. 10.2.4 Prepare one matrix spike and matrix spike duplicate per 40 samples, with a minimum of 2 matrix spikes per batch. 10.3 Continuing calibration checks 10.3.1 Prepare continuing calibration check samples to ensure the accuracy of the initial calibration curve. 10.3.2 Prepare, at a minimum, one continuing check per group o f 10 samples. For example, if a sample set = 34, four checks are prepared and extracted. 10.3.3 Prepare each continuing calibration check from the same matrix used to prepare the initial curve. 0^ 0~ 4 3~ 5r-1* ETS-8-4.1 Extraction of PFOS from Serum Page 5 of 14 10.3.4 The expected concentrations will fall within the mid-range of the initial calibration curve. Additional spikes may be included that fall in the low-range of the initial calibration curve. This is necessary if the analyst must quantitate using only the low end of the calibration curve (for example, 5 ppb - 100 ppb, rather than 5 ppb - 1000 ppb). 11.0 Calibration and Standardization_______________________________________ __ 11.1 Prepare matrix calibration standards 11.1.1 Transfer 1 mL of serum to a 15 mL centrifuge tube. 11.1.2 If most sample volumes are less than 1.0 mL, extract standards with matrix volumes equal to the sample volumes. Do not extract less than 0.50 mL of matrix. Record each sample volume on the extraction sheet. 11.1.3 While preparing a total of twenty aliquots in 15 mL centrifuge tubes, mix or shake between aliquots. 11.1.4 Two 1 mL aliquots, or other appropriate volume, serve as matrix blanks. Typically use the standard concentrations and spiking amounts listed in Table 1, at the end of this section, to spike, in duplicate, two standard curves, for a total of eighteen standards, two matrix blanks, and two method blanks. 11.1.5 Refer to validation report ETS-8-4.0 & ETS-8-5.0-V-1, which lists the working ranges and the Linear Calibration Range (LCR) for calibration curves. 11.1.6 Use Attachment D as an aid in calculating the concentrations of the working standards. See Section 13.0 to calculate actual concentrations of PFOS in calibration standards. 11.2 To each standard, blank, or continuing check, add appropriate amount of surrogate working standard for the concentration to fall within the calibration curve range 5 ppb 1000 ppb. 11.3 Extract spiked matrix standards following 12.6-12.16 of this method. Use these standards to establish each initial curve on the mass spectrometer. ETS-8-4.1 Extraction of PFOS from Serum 004352 Page 6 of 14 I Table 1 Approximate spiking amounts for standards and spikes Using 1.0 mL of matrix Working standard pL Approx, final cone, of (approx, cone.) analyte in matrix - - Blank 0.500 ppm 10 0.005 ppm 0.500 ppm 20 0.010 ppm 5.00 ppm 5 0.025 ppm 5.00 ppm 10 0.050 ppm 5.00 ppm 20 0.100 ppm 50.0 ppm 5 0.250 ppm 50.0 ppm 10 0.500 ppm 50.0 ppm 15 0.750 ppm 50.0 ppm 20 1.00 ppm 12.0 Procedure________________________________________________________________ 12.1 Obtain frozen samples and allow to thaw at room temperature or in a lukewarm waterbath. 12.2 Vortex mix for 15 seconds, then transfer 1.0 mL or other appropriate volume to a 15 mL polypropylene centrifuge tube. 12.3 Return unused samples to freezer after extraction amounts have been removed. 12.4 Record the initial volume on the extraction worksheet. 12.5 Label the tube with the study number, sample ED, date and analyst initials. See attached worksheet for documenting the remaining steps. 12.6 Spike all samples, including blanks and standards, ready for extraction with surrogate standard as described in 11.2. 12.7 Spike each matrix with the appropriate amount of standard as described in 11.1, or Table 1 in that section, for the calibration curve standards. Also prepare matrix spikes and continuing calibration standards. 12.8 Vortex mix the standard curve samples, matrix spike samples, and continuing calibration samples for 15 seconds. 12.9 Check to ensure the 0.5 M TBA reagent is at pH 10. If not, adjust accordingly. 12.10 To each sample, add 1 mL 0.5 M TBA and 2 mL of 0.25M sodium carbonate/sodium bicarbonate buffer. 12.11 Using an Oxford Dispenser, add 5 mL methyl-/ert-butyl ether. 12.12 Cap each sample and put on the shaker at a setting of 300 rpm, for 20 minutes. 12.13 Centrifuge for 20 to 25 minutes at a setting of 3500 rpm, or until layers are well separated. 004353 ETS-8-4.1 Extraction of PFOS from Serum Page 7 of 14 12.14 Label a fresh 15 mL centrifuge tube with the same information as in 12.5. 12.15 Remove 4.0 mL of the organic layer to this clean 15 mL centrifuge tube. 12.16 Put each sample on the analytical nitrogen evaporator until dry, approximately 1 to 2 hours. 12.17 Add 1.0 mL of methanol to each centrifuge tube using a graduated pipette. 12.18 Vortex mix for 30 seconds. 12.19 Attach a 0.2 pm nylon mesh filter to a 3 cc syringe and transfer the sample to this syringe. Filter into a 1.5 mL glass autovial or low-volume autovial when necessary. 12.20 Label the autovial with the study number, animal number and gender, sample timepoint, matrix, final solvent, extraction date, and analyst(s) performing the extraction. 12.21 Cap and store extracts at room temperature or at approximately 4 C until analysis. 12.22 Complete the extraction worksheet, attached to this document, and tape in the study notebook or include in study binder, as appropriate. 13.0 Data Analysis and Calculations___________________________________________ 13.1 Calculations 13.1.1 Calculate actual concentrations of PFOS, or other applicable fluorochemical, in calibration standards using the following equation: mL of standard x concentration of standard (ue /mLl________________ L= mL of standard + mL of surrogate standard + initial matrix volume (mL) Final Concentration (pg/mL) of PFOS in matrix 14.0 M e t h o d P e r f o r m a n c e _________________________________________ ;_________________________ 14.1 The method detection limit (MDL) is analyte and matrix specific. Refer to MDL report for specific MDL and limit of quantitation (LOQ) values (see Attachments B and C). 14.2 The following quality control samples are extracted with each batch of samples to evaluate the quality of the extraction and analysis. 14.2.1 Method blanks and matrix blanks. 14.2.2 Matrix spike and matrix spike duplicate samples to determine accuracy and precision of the extraction. 14.2.3 Continuing calibration check samples to determine the continued accuracy of the initial calibration curve. 14.3 Refer to section 14 ofETS-8-5.1 for method performance criteria. ( 15.0 Pollution Prevention and Waste Management _____________________ 15.1 Sample waste is disposed in biohazard containers, flammable solvent waste is disposed in high BTU containers, and used glass pipette waste is disposed in broken glass containers located in the laboratory. 004354 ETS-8-4.1 Extraction of PFOS from Serum Page 8 of 14 16.0 Records 16.1 Complete the extraction worksheet attached to this method, and tape in the study notebook or include in the 3-ring study binder, as appropriate. 17.0 A t t a c h m e n t s _______________________________________________________________________ 17.1 Attachment A, Extraction worksheet 17.2 Attachment B, MDL/LOQ values and summary 17.3 Attachment C, Calibration standard concentration worksheet 18.0 R e f e r e n c e s __________________________________________________________________________ 18.1 The validation report associated with this method is ETS-8-4.0 & 5.0-V-l. 18.2 FACT-M-3.1, "Analysis of Serum or Other Fluid Extracts for Fluorochemicals using HPLC-Electrospray Mass Spectrometry" 19.0 A f f e c t e d D o c u m e n t s ____________________________________________________________________ 19.1 ETS-8-5.1, "Analysis of Serum or Other Fluid Extracts for Fluorochemicals using HPLCElectrospray Mass Spectrometry" 20.0 R e v is io n s _________________________________________________________________________________ Revision Number 1 Reason For Revision Section 12.21 Changed to include sample storage at room temperature. Section 12.13 Added the shaker speed. Section 12.17 Final volume is 1.0 mL; not adjusted for initial volumes less than 1.0 mL. Revision Date 04/02/99 ETS-8-4.1 Extraction of PFOS from Scrum 004355 Page 9 of 14 3 M E n v iro nm en ta l L abo rato ry M ethod A n a ly sis of P o tassium P erfluorooctanesulfonate or O th er F luo ro ch em icals in Serum Extracts U sing H PL C -E lectrospray/M ass Spectrom etry Method Number: ETS-8-5.1 X*/ Author: Lisa Clemen, Robert Wynne Approved By: Adoption Date: 03/01/99 Revision Date: Laboratory Manager Date Group Leader Date Technical Reviewer Date 1.0 Scope and Application I .l Scope: This method describes the analysis of serum extracts for fluorochemical surfactants using HPLC-electrospray/mass spectrometry. 1.2 Applicable Compounds: Fluorochemical surfactants or other fluorinated compounds, or other ionizable compounds. 1.3 Matrices: Rabbit, rat, bovine, monkey, and human serum, or other fluids as designated in the validation report. 004356 Word 6/95 ETS-8-5.1 Analysis of Serum Extract Using ES/MS Page 1 of 9 2.0 S u m m a r y o f M e t h o d 2.1 This method describes the analysis of fluorochemical surfactants extracted from serum or other fluids, using HPLC-electrospray/mass spectrometry, or similar system as appropriate. The analysis is performed by monitoring a single ion characteristic of a particular fluorochemical, such as the perfluorooctanesulfonate (PFOS) anion, m/z= 499. Additionally, samples may be analyzed using a tandem mass spectrometer to further verify the identity of a compound by detecting daughter ions of the parent ion. 3.0 D e f in it io n s _________________________________________________________________________________ 3.1 Atmospheric Pressure Ionization (API): The Micromass Quattro II triple quadrupole systems allow for various methods of ionization by utilizing various sources, probes, and interfaces. These include but are not limited to: Electrospray Ionization (ESI), Atmospheric Pressure chemical Ionization (APcI), Thermospray, etc. The ionization process in these techniques occurs at atmospheric pressure (i.e., not under a vacuum). 3.2 Electrospray Ionization (ES, ESI): a method of ionization performed at atmospheric pressure, whereby ions in solution are transferred to the gas phase via tiny charged droplets. These charged droplets are produced by the application of a strong electrical field. 3.3 Mass Spectrometry, Mass Spectrometer (MS), Tandem Mass Spectrometer (MS/MS): The API Quattro II triple quadrupole systems are equipped with quadrupole mass selective detectors. Ions are selectively discriminated by mass to charge ratio (m/z) and subsequently detected. A single MS may be employed for ion detection or a series (MS/MS) for more specific fragmentation information. 3.4 Conventional vs. Z-spray probe interface: The latest models of Micromass Quattro II triple quadrupole systems (post 1998) utilize a "Z-spray" conformation. The spray emitted from a probe is orthogonal to the cone aperture. In the conventional conformation it is aimed directly at the cone aperture, after passing through a tortuous pathway in the counter electrode. Though the configuration is different, the methods of operation, cleaning, and maintenance are the same. However, Z-spray components and conventional components are not compatible with one another, but only with similar systems (i.e., Z-spray components are compatible with some other Z-spray systems, etc.) 3.5 Mass Lynx Software: System software designed for the specific operation of these Quattro II triple quadrupole systems. Currently MassLynx has Windows 95 and WindowsNT 4.0 versions. All versions are similar. For more details see the manual specific to the instrument (Micromass Quattro II triple quadrupole MassLynx or MassLynx NT User's Guide). 4.0 W a r n in g s a n d C a u t io n s 4.1 Health and Safety Warnings: 4.1.1 Use caution with the voltage cables for the probe. When engaged, the probe employs a voltage of approximately 5000 Volts. 004357 ETS-8-5.1 Analysis of Serum Extract Using ES/MS Page 2 of 9 4.1.2 When handling samples or solvents wear appropriate protective gloves, eyewear, and clothing. 4.2 Cautions: 4.2.1 Do not operate solvent pumps above capacity of 400 bar (5800 psi) back pressure. If the back pressure exceeds 400 bar, the HP 1100 will initiate automatic shutdown. 4.2.2 Do not run solvent pumps to dryness. 5.0 I n t e r f e r e n c e s ____________________________________________________________________ 5.1 To minimize interferences when analyzing samples, teflon should not be used for sample storage or any part of instrumentation that comes in contact with the sample or extract. 6.0 E q u ip m e n t _________ v _______________________________________________________________________ 6.1 Equipment listed below may be modified in order to optimize the system. Document any modifications in the raw data as method deviations. 6.1.1 6.1.2 Micromass Quattro II triple quadrupole Mass Spectrometer equipped with an electrospray ionization source HP 1100 low pulse solvent pumping system, solvent degasser, column compartment, and autosampler 7.0 Su p p l ie s a n d M a t e r ia l s _____________________________________________ _________ 7.1 Supplies 7.1.1 High purity grade nitrogen gas regulated to approximately 100 psi (House air system) 7.1.2 HPLC analytical column, specifics to be determined by the analyst and documented in the raw data. 7.1.3 Capped autovials or capped 15 mL centrifuge tubes . 8.0 R e a g e n t s a n d S t a n d a r d s _________________________________________________________________ 8.1 Reagents 8.1.1 Methanol, HPLC grade or equivalent 8.1.2 Milli-QTM water, all water used in this method should be Milli-QTM water or equivalent, and may be provided by a Milli-Q TOC Plus system or other vendor 8.1.3 Ammonium acetate, reagent grade or equivalent 8.2 Standards 8.2.1 Typically two method blanks, two matrix blanks, and eighteen matrix standards are prepared during the extraction procedure. See ETS-8-4.1. 9.0 Sample Handling :----------------------------------------- 0 U 4 3 5 8 ETS-8-5.1 Analysis of Serum Extract Using ES/MS Page 3 of 9 9.1 Fresh matrix standards are prepared with each analysis. Extracted standards and samples are stored in capped autovials or capped 15 mL centrifuge tubes until analysis. 9.2 If analysis will be delayed, extracted standards and samples can be refrigerated at approximately 4 C, or at room temperature, until analysis can be performed. 10.0 Quality Control________________________________________________________ _ 10.1 Solvent Blanks, Method Blanks and Matrix Blanks 10.1.1 Solvent blanks, method blanks and matrix blanks are prepared and analyzed with each batch to determine contamination or carryover. 10.1.2 Analyze a method blank and a matrix blank prior to each calibration curve. 10.2 Matrix Spikes y. * 10.2.1 Matrix spikes are prepared and analyzed to determine the matrix effect on the recovery efficiency. 10.2.2 Matrix spike duplicates are prepared and analyzed to measure the precision and the recovery for each analyte. 10.2.3 Analyze a matrix spike and matrix spike duplicate per forty samples, with a minimum of 2 spikes per batch. 10.2.4 Matrix spike and matrix spike duplicate concentrations will fall in the mid-range of the initial calibration curve. Additional spike concentrations may fall in the lowrange of the initial calibration curve. 10.3 Continuing Calibration Verifications 10.3.1 Continuing calibration verifications are analyzed to verify the continued accuracy of the calibration curve. 10.3.2 Analyze a mid-range calibration standard after every tenth sample, with a minimum of one per batch. 11.0 Calibration and Standardization__________________________________________ 11.1 Analyze the extracted matrix standards prior to and following each set of extracts. The average of two standard curves will be plotted by linear regression (y = my + b), weighted 1/x, not forced through zero, using MassLynx or other suitable software. 11.2 If the curve does not meet requirements, perform routine maintenance or reextract the standard curve (if necessary) and reanalyze. 11.3 For purposes of accuracy when quantitating low levels of analyte, it may be necessary to use the low end of the calibration curve rather than the full range of the standard curve. Example: when attempting to quantitate approximately 10 ppb of analyte, generate a calibration curve consisting of the standards from 5 ppb to 100 ppb rather than th full range of the curve (5 ppb to 1000 ppb). This will reduce inaccuracy attributed to linear regression weighting of high concentration standards. 004359 ETS-8-5.1 Analysis of Serum Extract Using ES/MS Page 4 of 9 12.0 P r o c e d u r e s _______________________________________________________________________________ 12.1 Acquisition Set up 12.1.1 Click on start button in the Acquisition Control Panel. Set up a sample list. Assign a filename using MO-DAY-last digit of year-sample number, assign a method (MS) for acquiring, and type in sample descriptions. 12.1.2 To create a method click on scan button in the Acquisition control panel and select SIR (Single Ion Recording) or MRM. Set Ionization Mode as appropriate and mass to 499 or other appropriate masses. A full scan is usually collected along with the SIRs. Save acquisition method. If MS/MS instruments are employed, additional product ion fragmentation information may be collected. See Micromass MassLynx GUIDE TO DATA ACQUISITION for additional information and MRM (Multiple Reaction Monitoring). 12.1.3 Typically the analytical batch run sequence begins with a set of extracted matrix standards and ends with a set of extracted matrix standards. 12.1.4 Samples are analyzed with a continuing calibration check injected after every tenth sample. Solvent blanks should be analyzed periodically to monitor possible analyte carryover and are not considered samples but may be included as such. 12.2 Using the Autosampler 12.2.1 Set up sample tray according to the sample list prepared in Section 12.1.1, 12.2.2 Set-up the HP1100/autosampler at the following conditions or at conditions the analyst considers appropriate for optimal response. Record actual conditions in the instrument logbook: 12.2.2.1 Sample size = 10 pL injection 12.2.2.2 Inject/sample =1 12.2.2.3 Cycle time =13.5 minutes 12.2.2.4 Solvent ramp = Time 0.00 min. 8.50 min. 11.0 min. 12.0 min. MeOH 40% 90% 90% 40% 2.0 mM Ammonium acetate 60% 10% 10% 60% 12.2.2.5 Press the "Start" button. 12.3 Instrum ent Set-up 12.3.1 Refer to ETS-9-24.0 for more details. 12.3.2 Check the solvent level in reservoirs and refill if necessary. 004360 ETS-8-5.1 Analysis of Serum Extract Using ES/MS Page 5 of 9 12.3.3 Check the stainless steel capillary at the end of the probe. Use an eyepiece to check the tip. The tip should be flat with no jagged edges. If the tip is found to be unsatisfactory, disassemble the probe and replace the stainless steel capillary. 12.3.4 Set HPLC pump to "On". Set the flow to 10 - 500 uL/min or as appropriate. Observe droplets coming out of the tip of the probe. Allow to equilibrate for approximately 10 minutes. 12.3.5 Turn on the nitrogen. A fine mist should be expelled with no nitrogen leaking around the tip of the probe. Readjust the tip of the probe if no mist is observed. 12.3.6 The instrument uses these parameters at the following settings. These settings may change in order to optimize the response: 12.3.6.1 Drying gas 250-400 liters/hour 12.3.6.2 ESI nebulizing gas 10-15 liters/hour 12.3.6.3 HPLC constant flow mode, flow rate 10 - 500 pL/min 12.3.6.4 Pressure <400 bar (This parameter is not set, it is a guide to ensure the HPLC is operating correctly.) 12.3.7 Carefully guide the probe into the opening. Insert probe until it will not go any further. Connect the voltage cables to the probe. 12.3.8 Print the tune page, with its parameters, and store it in the study binder with a copy taped into the instrument log. 12.3.9 Using the cross-flow counter electrode in the ES/MS source is recommended for the analysis of biological matrices. 12.3.10Click on start button in the Acquisition Control Panel (this may vary among MassLynx versions, see appropriate MassLynx USER'S GUIDE). Press the start button. Ensure start and end sample number includes all samples to be analyzed. 13.0 D a t a A n a l y s is a n d C a l c u l a t io n s 13.1 Calculations: 13.1.4 Calculate matrix spike percent recoveries using the following equation: % Recovery = Observed Result - Background Result x 100 Expected Result 13.1.5 Calculate percent difference using the following equation: % Difference = Expected Cone. - Calculated Cone, x 100 Expected Cone. 13.1.6 Calculate actual concentration of PFOS, or other fluorochemical, in matrix (jig/mL): (ng of PFOS calc, from std. Curve x Dilution Factor! x (Initial Volume of matrix fmLl + mL of Surrogate Standard! Final Volume (mL) 1 Hg 1000 ng 004361 ETS-8-5.1 Analysis of Serum Extract Using ES/MS Page 6 of 9 14.0 Method Performance 14.1 Method Detection Limit (MDL) and Limit of Quantitation (LOQ) are method, analyte, and matrix specific. Please see ETS-8-4.1, Attachment B, for a listing of current validated MDL and LOQ values. 14.2 Solvent Blanks, Method Blanks, and M atrix Blanks 14.2.1 Solvent blanks, method blanks, and matrix blanks values are must be below the lowest standard in the calibration curve 14.3 Calibration Curves 14.3.1 The r2value for the calibration curve must be 0.980 or better. 14.4 M atrix Spikes 14.4.1 Matrix spike percent recoveries are must be within 30% of the spiked concentration. 14.5 Continuing Calibration Verifications 14.5.1 Continuing calibration verification percent recoveries must be 30% of the spiked concentration. 14.6 If criteria listed in this method performance section isn't met, maintenance may be performed on the system and samples reanalyzed or other actions as determined by the analyst. Document all actions in the appropriate logbook. 14.7 If data are to be reported when performance criteria have not been met, the data must be footnoted on tables and discussed in the text of the report. 15.0 Pollution Prevention and Waste Management___________ - ____________ 15.1 Sample extract waste and flammable solvent is disposed in high BTU containers, and glass pipette waste is disposed in broken glass containers located in the laboratory. 16.0 Records____________________ ____________________________________________ 16.1 Each page generated for a study must have the following information included either in the header or hand written on the page: study or project number, acquisition method, integration method, sample name, extraction date, dilution factor (if applicable), and analyst. 16.2 Print the tune page, sample list, and acquisition method from MassLynx to include in the appropriate study folder. Copy these pages and tape into the instrument runlog. 16.3 Plot the calibration curve by linear regression, weighted 1/x, then print these graphs and store in the study folder. 16.4 Print data integration summary, integration method, and chromatograms, from MassLynx, and store in the study folder. 16.5 Summarize data using suitable software (Excel 5.0) and store in the study folder, see Attachment A for an example of a summary spreadsheet. 004362 ETS-8-5.1 Analysis of Serum Extract Using ES/MS Page 7 of 9 16.6 Back up electronic data to appropriate medium. Record in study notebook the file name and location of backup electronic data. 17.0 Tables. Diagrams. Flowcharts, and Validation Data_______________________ 17.1 Attachment A: ETS-8-5.1 Data summary spreadsheet. 18.0 References______________________________________________________________ __ 18.1 FACT-M-4.1, "Extraction of Potassium Perfluorooctanesulfonate or Other Fluorochemical compounds from Serum for Analysis Using HPLC-Electrospray/Mass Spectrometry 18.2 ETS-9-24.0, "Operation and Maintenance o f the Micromass Atmospheric Pressure Ionization/Mass Spectrometer Quattro II triple quadrupole Systems" 18.3 The validation report associated with this method is ETS-8-4.0 & 5.0-V-l. 19.0 Affected Documents_______________________________________ 19.1 ETS-8-4.1, "Extraction of Potassium Perfluorooctanesulfonate or Other Fluorochemical Compounds from Serum for Analysis Using HPLC-Electrospray/Mass Spectrometry" 20.0 REVISIONS________________________________________________________' _________________ Revision Number. 1 Reason For Revision Section 6.1.2 Clarification of HP1100 system components. Section 11.1 Average of two curves, not standard values, are used for plotting linear regression and added the 1/x weighting of the curve. Section 12.2.2.4 Clarification of solvent ramp. Section 17.1 Changed from attachment B to A. - Revision Date 04/02/99 ETS-8-5.1 Analysis of Serum Extract Using ES/MS 004363 Page 8 of 9 Laboratory Study # Study: Test M aterial: M atrix/Final Solvent: M ethod/Revision: Analytical Equipm ent System Number: Instrum ent Software/Version: F ilenam e: R-Squared Value: Slope: Y Intercept: Date o f Extraction/A nalyst: v D ate o f Analysis/A nalyst: G roup Dose Sam ple# C oncentration u^/m L In itial VoL mL D ilution F actor F inal Cone. ue/m L " Slope: Taken from linear regression equation. G roup/D ose: Taken from the study folder. Sam ple#: Taken from the study folder. C oncentration (ug/m L): Taken from the M assLynx integration summary. In itial V olum e (m L): Taken from the study folder. D ilution F acto r: Taken from the study folder. F in al Cone. (ug/m L): Calculated by dividing the initial volum e from the concentration Attachment A: Summary Spreadsheet ETS-8-5.1 Analysis of Serum Extract Using ES/MS 004364 Page 9 of 9 sitmmmiFluorine Analytical Chemistry Team D eterioration of PFOS in sera by Electrospray Mass Spectrometry (ESMS) Quantitation: Specificity: Limit ofDetection (LOD) : 1.0 ppb Retention Time: 7.35 min. Limit of Quantitation (LOQ) : 1.0 ppb Molecular Ion: 499 amu Range of Calibration Curve: l.Oppb- lOOppb Curve Correlation Coefficient (r3): 0.999 PHASE II:Analysis of pooled sera, US 1998 Average cone. Std. % Recovery IsL PFOS fppb)1 f i i 1006 44 2 MS MSP -27* 73 1007 44 6 -14* 35* 1008 S009 S010 43 26 28 2 8 7V 65* 37* 82 92 -6* 27* so il 45 2 100 108 W012 W013 54 50 13 78 92 5 90 102 W014 41 6 106 92 Range o f Average Concentration of PFOS in Lot Sera: Average cone LsL fEQ Sfrpb)1 W015 41 W016 47 W017 42 from 6 ppb to 54 ppb Std. 3 8 3 Contact: K.J. Hansen 8-6018 % Recovery MS MSP 120 112 51* 92 -45* 65* Results o f Quantitation of Method Blanks Blank Method Blank, H20-1 Method Blank, H 20-2 Method Blank, sera-1 ( 1) Method Blank, sera-1 (2) Method Blank, sera-1 (3) Method Blank, sera-2 (1) Concentration PFOS fppb) <LOD <LOD 7 <LOD <LOD <LOD Samnle3 QC1 QC2 QC3 QC4 QC5 [PFOS] recovered 49 49 49. 50 53 [PFOS] % expected recovery 49 100 49 100 49 100 49 102 49 108 System Reproducibility and Precision Checks ReprodudbUty'': Sample 1006-3 (2) fF Q ? fppb) MB 45 0.7% S009-3 (2) 11 0.4% W012-3 (2) 54 0.4% W015-3 (2) 33 1.8% W017-3 (2) 37 0.8% Precision*: Sample 1007-1 (1) 1007-1 (2) 1007-1 (3) 1007-1 (4) 1007-1 (5) Average Std. Dev. RSD P fQ ?(B B fe) 47 55 49 47 49 49 1 2% 1Reported values are the averages concentration PFOS o fsamples extracted in triplicate. 2QC samples are extracts o f blank sera spiked with PFOS standards. 3Reproducibility samples are repeat injections o f lot sera extracts, performed after every 15 samples. *Precision samples are 5 consecutive injections o f the same lot sera extract. Questionable Data. Source o f error currently undetermined. 3M Environmental Lab 5/7/98 004365 PHASE n BEST KOPY AVAILABLE r inorine Analytical Chemistry Team Jfe .c rm iiflft in sera by Electrospray Mass Spectrometry (ESMS) Q uantitafiO i. Specificity: Limit of Detection (LOD) : 10 ppb Retention Time: 6.90 min. Limit of Quantitation (LO Q): 1.0 ppb Molecular Ion: 413 amu1 Range of Calibration Curve: lOppb - lOOppb 369 amu3 Curve Correlation Coefficient (r2): 0.991 Contact: K.J. Hansen 8-6018] Phase II: Analysis of pooled sera (POAA) Average cone. Std. Lot POAA (ppb)3 Dev. */ Recovery MS MSP 1006 <LOD 0 56* 96 1007 <LOD 1 53* 79 1008 <LOD 0 90 74 S009 <LOD 1 65* 68* S010 so il <LOD <LOD 0 0 45* 57* 71 67* W012 W013 W014 <LOD <LOD <LOD 1V 0 0 92 94 100 100 98 95 Range of Average Concentration of POAA in Lot Sera: Average cone I s L POAA (ppb)3 W015 <LOD W016 <LOD W017 <LOD B001 <LOD B002 <LOD B003 <LOD B004 <LOD B005 <LOD Std. Dev. 0 0 0 0 0 0 0 0 % Recovery MS MSP 100 101 83 97 83 97 98 89 100 100 108 110 60* 94 94 89 all values were below the LOD/LOQ (lOppb) Results of Quantitation o f Method Blanks Blank Concentration POAA (nob) Method Blank, H 20-1 Method Blank. H 20-2 Method Blank, sera-1 (1) <LOD <LOD <LOD Method Blank, sera-1 (2) Method Blank, sera-1 (3) <LOD <LOD Method Blank, sera-2 (1) <LOD ___ Sample4 QC1 QC2 QC3 QC4 QC5 {POAA] recovered 94 91 91 92 93 [POAA] 98 95 95 96. 97 / recovery 98 95 95 96 97 System Reproducibility and Precision Checks Reproducibilty3: Sample POAA (area) RSP 1006-3 (2) 7700 0.9% S009-3 (2) 3100 7.0% W02-3 (2) 6500 0.0% W015-3 (2) 4700 0.0% W017-3 (2) 4900 5.8% B003-3 (2) 3100 2.0% Precision*: Sample' 1007-1 (1) 1007-1 (2) 1007-1 (3) 1007-1 (4) 1007-1 (5) Average Std. Dev. RSP Peak Area 5800 5600 6300 6300 6300 6100 300 5% '/on m/z=413 usedfor purposes o f quantitation. 2lon m/z=369 usedfor purposes o f verfication o fcompound identity. 3Reported values are the averages concentration POAA o fsamples extracted in triplicate. *QC samples are extracts o f blank sera spiked -with POAA standards. 5Reproducibility samples are repeat injections o f lot sera extracts, performed after every 15 samples. Precision samples are 5 consecutive injections o fthe same lot sera extract. Questionable Data. Source o f error currently undetermined. 004366 3M Environmental Lab 5/7/98 PHASE D (POAA) K.J. Hansen Phase III 3M Environmental la b Q u a n t it a tiv e A a a i y j j , 0 { pFOS in Individual's Sera Determination o f PFOS in sera by Electrospray Mass Spectrometry (ESMS) Quantitation: Spedficity: Limit of Detection (LOD) : 1.0 ppb Retention Time: 7.35 min. Limit of Quantitation (LOQ) : 1.0 ppb Molecular Ion: 499 amu Range of Calibration Curve: l.Oppb - lOOppb Curve Correlation Coefficient (r2): 1 Results o f Quantitation o f Individual Sera Samples Average cone. LsL PFO SIpfibl 198 58 298 41 398 46 498 32 598 67 698 53 798 42 898 49 998 39 1098 54 1198 60 1298 37 / RSD 41* 14 2 NwA ^ 0 7 N.. 35* 11 5 11 1 verage cone. M . P.FQS (BPb) % RSD 1798 46 15 1898 56 10 1998 70 3 2098 40 2 2198 48 9 2298 60 13 2398 72 1 2498 35 20 2598 41 33* 2698 53 11 2798 57 15 2898 33 N.A. 1398 1498 1598 1698 45 31 96 74 N.A. 12 2 1 2998 3098 3198 65 67 33 7 50* 11 N.A. = No %RSD calculated as there was only enough sample for 1 analysis. 3/2/98 004367 Quantitative Analysis of PFOS and POAA in Individual's Sera 4IV<if Determination of PFOS/POAA in sera by Electrospray Mass Spectrometry (ISM S) Q uan titatio n : Spedfldty: Limit o fDetection (LO D): 1.0 ppb/10ppb Rtention Time: 7.3S m ia/6.9 min. Limit of Quantitation (LOQ) : l.Oppb/lOppb Molecular Ion: 499 amu/413 amu, 369 amu Range of Calibration Curve: l.Oppb - I00ppb/10 ppb -lOOppb Correlation Coefficient (r2): 0.999/0.999 P re lim in a ry Results o f Quantitation o f Individual Sera Samples Avg. cone Std Avg, cone. Std. Average cone. Std. IS# PFOS (ppbr Pev. 198 46 a d EQ AA frcfel1 P ev. f f i f t rfQ & frafrl1 Pev. <LOQ a d 1798 41 ad 298 35 ad. <LOQ a d 1898 52 ad 398 45 ad . 18 3 1998 71 a d 498 32 ad. 10 a d 2098 40 1 598 .'67 ad . <LOQ a d 2198 45 p .d 698 53 4 <LOQ a d 2298 54 ad 798 42 Dad <.O Q a d 2398 71 ad 898 49 17 <LOQ a d 2498 28 ad 998 37 ad . <LOQ a d 2598 30 ad 1098 52 ad. <LOQ a d 2698 47 1 1198 39 ad. <LOQ a d 2798 51 ad 1298 37 a d <LOQ a d 2898* 30 ad 1398 1498 45 a d 28 a d <LOQ a d 2998 41 ad <LOQ a d 3098 41 ad 1598 96 3 10 1 3198 36 a d 1698 73 a d COQ ad Avg. cone. FQAA-fPBM* <L0Q COQ <LOQ <LOQ <LOQ COQ 12 COQ <LOQ <LOQ COQ <LOQ <LOQ COQ <LOQ Std. Pev. ad ad ad' ad ad ad ad ad ad ad ad ad ad ad ad Results o f Quantitation o f Method Blanks filia l* g g o sttr a ia Method Blank, H 20-1 <LOD Method Blank, sera-1 (1) 4 Method Blank, sera-1 (2) C O D gQ A A lIpakl Method Blank, H 20-1 COD Method Blank, sera-l (1) C O D Method Blank, sera-1 a ) C O D System Reproducibility Checks ReprodudbUt': Sappia PFOS fochi USD 0198-2 (2) 33 4.3% 1098-2 (2) 20 3.6% 2098-1 (2) 40 1.8% 2698-1(2) 47 3.0% Saou ls1 QCl QC2 QC3 g ro s] recovered 99% 93% 90% T Q A A ftim i 2800 1300 1400 2300 ESP 0.0% 0 .0% 0.0% 6. 1% [POAA] recovered 93% 87% 82% 'Reported values artfrom a single analysis unless a std. dev. is indicated; in those casts, 2 alliquots o fsampit teert analysed and averaged. .2QC samples are extracts o f blank sm s spiked with PFOS/POAA standards. ^Reproducibility sam ples are repeat infections o f lo t sera extracts, perform ed every 5 samples. 004368 K.J. Hansen Phase III 3M Environmental Lab Quantitative Analysis of PFOS in Individual's Sera Determination of PFOS in sera by Electrospray Mass Spectrometry (ESMS) Q uantitation: Specificity: Limit o f Detection (LO D): 1.0 ppb Retention Time: 7.33 min. Limit o f Quantitation (LO Q): 1.0 ppb Molecular Ion: 499 amu Range of Calibration Curve: l.Oppb - lOOppb Curve Correlation Coefficient (r3): 1 Results o f Quantitation of Individual Sera Samples Average cone h sL PFOS (PPM 198 38 298 41 398 46 498 32 398 67 698 53 798 42 898 49 998 39 1098 54 1198 60 1298 37 / RSD 41* 14 2 N .A 0v 7 N .A 35* 11 5. 11 1 verage cone. Lot PFOS (ppb) % RSD 1798 46 15 1898 56 10 1998 70 3 2098 40 2 2198 48 9 2298 60 13 2398 72 1 2498 35 20 2598 41 33* 2698 53 11 2798 57 15 2898 33 N .A 1398 1498 1398 1698 45 31 96 74 N .A 12 2 1 2998 3098 3198 65 67 33 7 50* 11 N.A = No %RSD calculated as there was only enough sample for 1 analysis. 4/9/98 004369 Fluorine Analytical Chemistry Team Determination o f PFOS in sera by Electrospray Mass Spectrometry (ESMS) Quantitation: Specificity: Limit o f Detection (LO D): 1.0 ppb Retention Time: 7.3S min. Limit of Quantitation (LOQ): 1.0 ppb Molecular Ion: 499 amu Range o f Calibration Curve: l.Oppb - lOOppb Curve Correlation Coefficient (r*): 1 PHASE HI: Analysis of individual sera samples, 1998 US Average cone. LSL P F Q S lB B b ] 198 58 298 41 398 46 498 32 598 67 698 53 798 42 898 . 49 998 39 1098 54 1198 60 1298 37 / RSD 41* 14 2 N .A 0 7 N.A. 35* 11 5 11 1 Average cone LfiL PFOS (PD b) V. RSD 1798 46 15 1898 56 10 1998 70 3 2098 40 2 2198 48 9 2298 60 13 2398 72 1 2498 35 20 2598 41 33* 2698 53 11 2798 57 15 2898 33 N .A . 1398 1498 1598 1698 45 31 96 74 N .A 12 2 1 2998 3098 3198 65 67 33 7 50* 11 N.A. = No %RSD calculated as there was only enough sample for 1 analysis. contact: K.J. Hansen 8-6018 3M Environmental Lab 5/7/98 004370 PHASE in 3M Environmental Laboratory- Fluorine Analytical Chemistry Team Kris Hansen - Sr. Analytical Chemist Fluorine Analytical Chemistry Team Building 2-3E-09 612-778-6018 Phase IV: PFOS analysis of US human sera samples collected in the 1970s Summary: Twelve frozen sera samples of various volumes were supplied to the Environmental Lab by Jeff Mandcl (3M Medical). One mL or no more than one half of the volume of each sample was extracted using an ion-pairing reagent The extracts were analyzed quantitatively for FFOS by HFLC-ESMS. Analyte identification was verified by comparison of molecular ion-HPLC retention time o f the extracted analyte and standard material. Samples were quantitatively evaluated against a specially prepared extracted curve. The twelve samples contained an average concentration of 27 ppb FFOS ranging from 2- 48 ppb. The PFOS levels were significantly lower than in the 31 samples collected from individuals in 02/98 and exhibited significantly greater variability in PFOS levels than those same samples. The percent abundance of the branched chain isomer was 37.3 +/- 4.8%, compared to 33.7 +/-7% in the 1998 samples. In standard solutions o f PFOS, the branched chain accounts for 21.7 +/-3.1% o f the total concentration. A detailed summary of the results is included. The identity o f PFOS in several randomly selected samples in this phase was confirmed using HPLC-ESMSMS. In all cases, the analyte exhibited a characteristic retention time, a characteristic primary ion, and five characteristic secondary ions. Experimental summary: Sample preparation: Ion-pairing extraction In a pH controlled environment, an ion-pairing reagent, tetrabutyl ammonium (TBA), is used to extract the analyte from the matrix The cationic reagent selectively targets anionic compounds, like PFOS and POAA. Subsequent to the formation of the TBA-anion pair, the analyte is transferred to a non-polar organic solvent (ethyl acetate), dried, and reconstituted in methanol for MS analysis. HPLC: Characteristic retention timesfo r PFOS In HPLC, an aliquot of the extract is injected and passed through a chromatographic column. Based on the affinity o f the analyte for the stationary-phase in the column relative to the quid mobilephase passing through the column, the analyte is retained for a characteristic amount o f time. For example, in a standard solution, PFOS may elute at 9.1 minutes. Retention times between a standard PFOS solution and the analyte extracted from sera in this analysis were matched to within 1% on the HPLC system. ES/MS: Detection and monitoring o f the molecular ion Analysis of PFOS standards indicates that the primary ion characteristic o f FFOS is m/z = 499 amu, corresponding to the mass of the anionic surfactant (QFpSOj-). This ion was monitored selectively to maximize sensitivity. A scan of m/z=100 to 1210 (negative only) was also collected. H P LC ES-M S/M S: Detection and monitoring o fcharacteristic secondary ions ES-MSMS is very similar to ESMS, except that it adds an additional dimension o f certainty to compound identification. As in ESMS, a characteristic ion is selected. After selection, the ES-MSMS 004371 characterizes the ion further by smashing it apart with high energy gas. As a result o f the smashing, secondary ionic fragments (daughter ions), characteristic of the molecule, are created and detected. For example, for PFOS analysis, ion 499 is selected as the characteristic primary ion. This ion is smashed into other ions such as 80 amu (corresponding to SOS'), 99 amu (corresponding to FSCV). 130 amu (corresponding to C^SOs*), 180 amu (C2F4S03'), and 230 amu (CjFiSOa"). Each o f these secondary fragments is detected at the detector. Quality control summary: Due to sample limitations, sera samples were not extracted in duplicate. For this me reason, matrix spikes could not be evaluated for the samples. Instead, two samples o f pooled sera purchased from Sigma were used for duplicate matrix spike analysis. Methanol blanks were analyzed before and after each sample to ensure complete isolation of the sample. A mid-level quality control sample was analyzed after the sixth sera sample. The four point extracted curve was analyzed before and after the samples. Quantitation o f the samples is based on the linear regression equation derived from the average o f the two bracketing curves. Specific QC parameters are available on the appended results table. Instrumental specifics: HPLC system Hewlett Packard Series 1100 Liquid Chromatograph Column:Keystone Betasil Cm 2 x 100 mm 3 pm particle size Flow rate: 300 p iAran. Solvent A: 2.0 mM Ammonium Acetate Solvent B: Methanol Solvent Gradient: 45 to 90 %B in 9.50 mins. Hold at 90 %B for 3.50 mins. Return to 45 %B in 1.50 mins. Hold at 45 %B for 3.5 mins. Injection volume: 10 pL Injections / sample: 1 Electrospray mass spectrometer Micromass Platform n API Mass Spectrometer * MassLynx 2.1 Software Cone voltages: -20v, -60v Mode: electrospray negative Source temperature: 115 C. Analyzer vacuum pressures: 0.000043 mBar, 0.000079 mBar Ions: 369,413,499 Electrode: cross-flow 004372 Fluorine Analytical Chemistry Team Samples Received: Samples Extracted: Samples Analyzed: 3/6/98 3/9/98 3 /1 1 -3 /1 2 /9 8 Analysis by: Calibration Curve Range: r*of Calibration Curve: LOQ/LOD1: KJ.Hansen 778-6018 ESMS (-) lppb -49ppb 0.998 1 ppb PHASE IV: Analysis o f historic samples, 1970-80s C onc.ofPFO S Sample ID (ppb+B40) PG7-SE13 PG3-SE11 13 32 98SE774 ' 30 v 98SE117 24 PG25 40 PG5 29 Sample ID GRAH0003 80-01-0089-01 80-01-0044-01 W ill 0002 DemmOOOl 80-52-0514-01 Cone of PFOS (ppb) 48 15 39 2 44 25 Average (ppb) StiLDev. %RSD 28 9 32 Q C R esu lts QC Check: Calibration Check (QC1) MS MSD % Recovery 100% 105% 105% Branched Isomer Abundance Percentages Sample Pool Standard Material Individuals Sampled in 1998 3M Cottage Grove Plant Workers Individuals Sampled in 1970's Pooled Sera Samples % Abundance of Branched Isomers (Average) 22% 34% 45% 37% 31% % Abundance of Branched Isomers (Std. Dev.) 3% 5% 4% 5% 5% n (# sampled in pool) n/a 31 3 12 11 1 - LOD/LOQ * Limit o f Detection/Limit o f Quantitation 3M Environmental Lab 5/7/98 004373 PHASE IV 1 Piv-7080 W 9I rite laapli hikara (rfO0]ppfc (PTO0|iTgSLltev Iff I 3149801 MaOH blank 3149802 Baaotad human aara blank 3117 3490 0.0 0.0 3149803 Ettraolad humanaara blank 3468 0.0 3149813 MaOH blank 3347 0.0 BEST COPY AVAILABLE 3149816 MaOHbUnk 3736 0.0 1 3149817 PQ7-3B13 3149818 PQ7-SB13 3149819 P07-SB13 27470 27393 26140 14.0 13.9 13.2 13.7 03 3 3149820 MaOH blank 3634 0.0 I 3149821 PG3-SE11 3149822 P03-3811 3149823 P03-3B11 37133 36044 36116 322 31.5 31.5 31.7 04 1 3149824 MaOH blank 3626 0.0 3149823 98SB774 32334 293 I 3149826 9SSB774 3149827 983B774 30733 31931 283 29.0 219 06 3 3149828 MaOH blank 3476 0.0 3149829 983BI17 43396 23.0 3149830 983B117 43768 23.2 1 3149831 98SEI 17 3149832 983B117 43698 44232 233 243 243 04 2 3149833 MaOHblade 3271 a o 3149834 P023 96689 36.4 3149835 P023 94900 333 ] 3149836 PQ25 3149837 MaOHblank 93836 3241 34.7 ao 333 09 2 N < n . (QC) QC rara-rw y (H ) 3149838 9td 4-1.48.8 79521 433 -6 94 3149839 3td 4-1.48.8 I 3149840 MaOHblank 76841 3186 443 ao -9 91 3149841 PQ5 34030 303 3149842 POS 33078 309 3149843 POS 3149844 POS 34670 33860 307 302 303 03 1 IiI 3149843 MaOHblank 3149846 (3RAH003 3149847 QRAHD03 2907 77384 73104 0.0 44.6 420 3149848 OSAH003 78233 43.1 419 1.7 4 3149849 MaOHblank 3438 OO ] 3149830 80-01-008941 3149831 80-01-0089-01 27217 27410 118 119 3149832 80-01-0089-01 27137 118 118 Ol 1 3149833 MaOHblank "3393 OO 3149834 80-01-0044-01 72475 41.6 I 3149833 804)1-0044-01 3149836 804)1-0044-01 3149837 804)1-0044-01 71363 69231 70013 409 39.6 401 403 09 2 3149838 MaOHblank 3331 OO 3149839 Std. 4-1,48.8 ppb 78472 '4 3 3 ] 3149860 MaOHblade 3334 OO 3149861 Will 0002 7391 1.8 7 93 3149862 W310002 7383 1.6 3149863 Will 0002 7222 13 3149864 Will 0002 7389 1 J ] 3149863 WO! 0002 3149866 MaOHblank 7362 1.6 3367 OO 1.7 Ol 6 3149867 DEMM0001 73672 433 J 3149868 DEMM0001 3149869 DEMM 0001 3149870 MaOH blade 73938 76443 423 44.0 433 08 2 3420 0.0 3149871 80-32-0314-01 41084 223 3149872 80-324)314-01 3149873 80-324)314-01 40331 220 140387 21.9 221 03 3149874 MaOH blade 3822 0.0 3149873 MaOH blank 3341 OO 3149876 H20 bladc-1 4799 Ol 3149877 H20 bladc-2 4698 OO 3149878 MaOHblade 3383 00 3149879 Poolad >3r-l 33332 29.8 3149880 Poolad aaaa-2 33640 313 303 1.0 3149881 MaOH blank 4035 ao H dav. . tpika name](H ) 3149882 M 319.9 ppb 88880 51.6 6.0 106.0 3149883 MSD 19.9 ppb 3149884 MaOHblade 86059 3728 49.9 00 308 1.2 970 0 0 4 3 7 4 3149883 MaOH blank ' 3877 OO 3149886 Hxtractad btaaan aara blade 4019 OO 3149887 Batraotad human aarablank 4013 0.0 Fluorm Analytical Chamiatty Tm IMBnvarnmantalLab I U . Hmaan 778-6018 3M Environmental Laboratory- Fluorine Analytical Chemistry Teani Kris Hansen - Sr. Analytical Chemist " Fluorine Analytical Chemistiy Team Building 2-3E-09 612-778-6018 Phase V and VI: Quantitative analysis of PFOS in historic human sera samples and in human sera samples from rural Northern China Summary: Two sets o f samples, the first with 12 samples in plastic vials (Z-set) and the second with 5 1 samples in glass vials (M-set), were supplied to the Environmental Lab by JeffMandel (3M Medical) on March 20,1998. For boths^ts, sample volume varied from 0.2 to 1.0 mL o f human sera. A volume between 0.4 - 1.0 mL of each sample was extracted using an ion-fairing reagent Six samples in the Z-set did not contain enough volume for extraction; in these cases, two samples (e.g. ZZA6 and ZZA8) were combined and extracted as one sample. The extracts were analyzed quantitatively for perfluorooctane sulfonate (PFOS) by high-pressure liquid chromatography-electrospray mass spectrometry (HPLC-ESMS). Analyte identification was verified by comparison o f molecular ion-HFLC retention time o f the extracted l analyte and standard material. Samples were quantitatively evaluated against a specially prepared, five- point extracted curve (^ 2 = 0.999). Each sample extract was analyzed twice. Calibration checks, . analyzed every 5 samples, were within 6% of expected values. The identity o f PFOS in several randomly selected samples in this phase (M -set only) was confirmed using HPLC-ESMSMS. In all cases, the analyte exhibited a characteristic retention time, a characteristic primary ion, and five characteristic secondary ions. H istoric samples The sera samples in the M-set, collected from donors in the US in the 1960s, had an average PFOS concentration o f 33.4 ppb; the samples ranged in PFOS concentration from 11.8-to 59.4. Foreign samples The 9 samples in the Z-set (including 3 composite samples) were collected from donors in rural China and were determined to have a much lower concentration of PFOS than any human sera data set analyzed previously. Six samples did not contain PFOS above the detection limit o f this analytical method (1 ppb). The remaining 3 samples evidenced PFOS concentrations less than 5 ppb, the limit of quantitation. With the exception of one human sera sample collected in the 1980s (W ill 0002), these are the only human samples determined to contain less than 10 ppb PFOS. Experimental summary: Sample preparation: lon-pairing extraction In a pH controlled environment, an ion-pairing reagent, tetrabutyl ammonium sulfate (TBA), is 1 used to extract the analyte from the matrix. Anionic compounds, like the cationic reagent, selectively targets PFOS and perfluorooctanoate (POAA). Subsequent to the formation of the TBA-anion pair, die analyte is transferred to a non-polar organic solvent (ethyl acetate), dried, and reconstituted in methanol for MS analysis. HPLC: Characteristic retention tim esfo r PFOS In HPLC, an aliquot of the extract is injected and passed through a chromatographic column. Based on the affinity o f the analyte for the stationary-phase in the column relative to the liquid mobilephase passing through the column, the analyte is retained for a characteristic amount o f time. For example, in a standard solution, PFOS may elute at 10.5 minutes. Retention times between a standard 004375 05/07/98 1 PFOS solution and the analyte extracted from sera in this analysis were matched to within 1% on the HPLC system. ES/MS: Detection and monitoring o f the molecular ion Analysis o f PFOS standards indicates that the primary ion characteristic o f PFOS is m /z 499 amu, corresponding to the mass o f the anionic surfactant (C*Fi7S03-). This ion was monitored selectively to maximize sensitivity. A scan of m/z=T00 to 1210 (negative only) was also collected. H PLC ES-M S/M S: Detection and monitoring o f characteristic secondary ions ES-MSMS is very similar to ESMS, except that it adds an additional dimension o f certainty'to compound identification. As in ESMS, a characteristic ion is selected. After selection, the ES-MSMS characterizes the ion further by smashing it apart with high-energy gas. As a result o f the smashing, secondary ionic fragments (daughter ions), characteristic of the molecule, are created and detected. For example, for PFOS analysis, ion 499 is selected as the characteristic primary ion. This ion is smashed into other ions such as 80 amu (corresponding to SQj"), 99 amu (corresponding to FSO s), 130 amu (corresponding to CF2S 0 3'), 180 amu (C2F4SQ3), and 230 amu (CjFeSQs"). Each o f these secondary fragments is detected at the detector. Quality control summary: Due to sample limitations, sera samples were not extracted in duplicate, with 1 exception. For this same reason, matrix spikes could not be evaluated for the samples. Each sample extract was analyzed in duplicate. One sample, M001278 contained enough material for two extractions. The presented results for this sample are based on duplicate analysis of each of the two extractions. Methanol blanks were analyzed periodically to ensure complete isolation of the sample. A mid-level (24 ppb) quality control sample was analyzed every 5 samples. Because human sera without PFOS is not available, samples o f pooled sera purchased from Sigma were pre-extracted, spiked with a standard concentration o f PFOS, and re-extracted. This pre-extracted sera is understood to be very similar to the sample matrix. This five point extracted curve was analyzed before and after the samples. Quantitation of the samples is based on the linear regression equation derived from the average of the two bracketing curves (r*2 - 0.999). Specific QC parameters are available on the appended results table. Lim it o f Quantitation The current analytical method has been used to evaluate a large range o f PFOS levels extracted from sera; the method is not optimized for precise low-level (< 5 ppb) quantitation, but instead for samples in the range of 20-60 ppb. Given the concentrations chosen for this optimization o f the standard curve, while there is a statistical difference for samples evaluated from 5-100 ppb, there is no statistical difference between samples quantitated from 1-5 ppb. The quantitation does not become statistically dfendable until 5 ppb; thus this value represents the limit of quantitation for this method. Samples designated <LOQ are estimated to contain 1-5 ppb PFOS. If further accuracy and precision in quantitation of the low-level Z-set samples (samples less than 5 ppb) is required, a special low-level quantitation system w ill be required. . Lim it o fDetection The limit o f detection for this analytical method is based on analyst recognition o f the distinct peak shape resulting from PFOS analysis. PFOS standard material and PFOS extracted from sera analyzed by the conditions reported here, show near baseline resolution of 1 branched and 1 linear PFOS . isomer. The apex o f each peak in the HPLC-trace occurs at a specific, reproducible retention time and the ratio of the branched peak to the linear peak is fairly consistent down to the low standard, 1.0 ppb. Those samples that a) had PFOS levels evaluated at less than the limit of quantitation and b) evidenced the distinct peak shape, were designated < LOQ. Samples that a) have PFOS levels evaluated at less than the limit of quantitation and that b) do not reflect the unique PFOS peak shape, were designated < LOD. These samples are estimated to contain from 0-1 ppb PFOS. 05/07/98 2 Instrumental specifics: HPLC system Hewlett Packard Series 1100 Liquid Chromatograph Column:Keystone Betasil C 2 x 100 nun 5 pm particle size Flow rate: 300 pi/m in. Solvent A: 2.0 mM Ammonium Acetate Solvent B: Methanol Solvent Gradient: 45 to 90 %B in 9.50 mins. Hold at 90 %B for 3.50 mins. Return fo.45 %B in 1.50 mins. Hold at 45 %B for 3.5 mins. Injection volume: 10 pL Injections / sample: 2 Electrospray mass spectrometer Micromass Platform II API Mass Spectrometer, "Chide" MassLynx 2.4 Software Cone voltages: -20v, -60v Mode: electrospray negative Source temperature: 115 C. Analyzer vacuum pressures: 0.000043 mBar, 0.000079 mBar Ions: 369,413,499,427 Electrode: cross-flow 05/07/98 004377 3 Fluorine Analytical Chemistry Team KJ.H tm sen 778-6018 Samples Received: 3/6/98 Samples Extracted: 3/9/98 Samples Analyzed: 3/11 - 3/12/98 Analysis by: Calibration Curve Range: r*of Calibration Curve: LOD1: LOQ2: ESMS (-) lppb - 49ppb 0.998 1 ppb 5 ppb PHASE V and VI: Analysis of foreign (rural Northern China) and historic sera samples (1960) Sample ID Cone. ofPFOSj (ppb+B40) ZZAI <LOD ZZAI <LOD ZZA3 <LOQ V ZZA4 <LOQ ZZA5 <LOD ZZA6 and 8 <LOD ZZAI <LOD ZZA9 and 10 <LOQ ZZA11 and 12 <LOD 1 Cone, of PFOS 1 Sample ID (ppb) m001205 24.4+ /-0.4 MOO1278 48.4+ /-1.5 M01202 23.2+/-0.2 MOO1221 11.8+/-0.1 MOO1208 59.4+ /-0.1 QC Results QC Check: Calibration Check (QC1) Calibration Check (QC2) % Recovery 104% 106% Branched Isomer Abundance Percentages Sample Pool Standard Material Individuals Sampled in 1998 3M Cottage Grove Plant Workers Individuals Sampled in 1970's Pooled Sera Samples % Abundance of Branched Isomers (Average) 22% 34% 45% 37% 31% % Abundance of Branched Isomers (Std. Dev.) 3% 5% 4% 5% 5% 1 - LOD * Limit o f Detection 2- LOQ * Limit o f Quantitation 3M Environmental Lab 5/7/98 B 1 1 ET I n/a 31 3 12 11 004378 PHASE V and VI Quantify Compound Summary Aeport Sample Lit: Laat modified: Method: Last modified: Job Code: C :\MASSLYNX\r*ASSS.mO\SAN t S D a \ 0 3 2 1 96 Sun Her 22 14:33:9 Iff* C :\MASSLTNX\PIIASZS. PRONMiTRDB\PyOS at Mar 21 19:19:42 199 Printed: Sun Mar 22 14:32:37 199 Compound 1: PF08 Name Type Sample Text 1 03219401 Analyte MeOH Blank 2 03219902 Analyte H20 blank-1 3 03219403 Analyte Sera blank-1 4 03219904 Standard 0.996 ppb rc Mix S 0321940S Standard 4.90 ppb PC Mix 03219906 Standard 24.4 ppb PC Mix 7 03219407 standard 49.0 ppb rc Mix 9 03219409 Standard 96.2 ppb PC Mix 9 03219809 Standard 142 ppb 1C Mix 10 03219910 Analyte MeOH Blank 11 03219811 Analyte M001205 12 03219812 Analyte M001205 (2) 13 03219813 Analyte M001278-1 14 03219814 Analyte M001278-1 (2) 15 03219415 Analyte M001278-2 16 03219816 Analyte M001276-2 (2) 17 03219617 Analyte M01202 19 03219619 Analyte M012O2 (2) 19 03219619 Analyte M001221 20 03219620 Analyte M001221 < 2 ) V 21 03219621 Analyte H001208 22 03219822 Analyte M001208 (2) 23 03219823 QC Cal check, 24.8 ppb PfOS 24 03219824 Analyte MeOH blank 25 03219925 Analyte ZZAI 26 03219626 Analyte ZZAI (2) 27 03219827 Analyte ZZA2 29 03219828 Analyte ZZA2 (2) 29 03219829 Analyte ZZA3 30 03219630 Analyte ZZA3 (2) 31 03219831 Analyte ZZA4 32 03219832 Analyte ZZA4 (2) 33 03219833 Analyte ZZA5 34 03219834 Analyte ZZA5 (2) 35 03219835 QC Cal check, 24.9 ppb PFOS 36 03219836 Analyte MeOH 37 03219837 Analyte ZZA6 and 8 38 03219639 Analyte ZZA6 and 8 -423 39 03219839 Analyte ZZA7 40 03219840 Analyte ZZA7 (2) 41 03219841 Analyte ZZA9 and 10 42 03219842 Analyte ZZA9 and 10 (2) 43 03219843 Analyte ZZA11 and 12 44 03219844 Analyte ZZA11 and 12 (2) 45 03219845 Analyte AZ26-1 46 03219846 Analyte A226-1 (2) 47 03219847 Analyte MeOH blank 49 03219848 Analyte MeOH blank 49 03219849 Analyte H20 blank-1 SO 03219850 Analyte Sera blank-1 51 03219851 Standard 0.996 ppb 1C Mix 52 03219652 Standard 4.90 ppb rc Mix S3 03219853 Standard 24.8 ppb PC Mix 54 03219854 Standard 49.0 p pb 1C Mix 55 03219855 Standard 96.2 ppb PC Mix 56 03219856 Standard 142 ppb rc Mix BEST COPY AVAILABLE Area 4292 3239 2983 SOIS 10464 41137 74118 142767 193768 3287 39673 36868 74992 76499 71794 7257 37934 37464 21224 2101 90087 9789 41304 2821 3310 3397 3277 3300 5598 5721 7108 705 5066 5340 41833 2900 4863 5071 4181 4207 6631 6488 5342 5261 44210 4425 2603 2744 3397 3223 5081 10482 40948 75059 143153 203166 Cone. Dev ria 0.26 bb 0.00 bb 0.00 MM 0.76 -23 Ml 4.53 -6 Ml 25.72 4 Ml 48.51 -1 Ml 95.96 0 Ml 131.18 -a m i x 0.00 bb 24.71 MK 24.15 MM 49.11 Ml 50.16 Ml 46.90 Ml 47.45 Ml 23.51 Ml 23.20 MM 11.96 MM 11.82 Ml 59.54 Ml 59.34 Ml 25.64 4 Ml 0.00 bb 0.00 Ml 0.00 Ml 0.00 Ml 0.00 Ml 1.17 MM 1.25 MN 2.21 MM 2.17 MM 0.60 Ml 0.99 MK 26.20 6 MM 0.00 bb 0.67 MM 0.60 MM 0.19 Ml 0.20 MM 1.88 HK 1.7 MM 0.99 MM 0.93 MM 27.85 MM 27.88 MM 0.00 MM 0.00 Ml 0.00 bb 0.00 MM 0.61 -19 MM 4.54 -7 Ml 25.59 3 MM 49.16 0 MM 96.21 0 Ml 137.68 -3 MMX tap 1 V i 1 ? .* " y iW '" * VV J I Quantify C alibration Rapare Calibration : C i\ H J k S J L T K X \ A M 5 .m O \ C U * V 1 0 f \ r T O Laat audiflod: fun Nar 22 14:4:Jf I t H Printad: fun Nar 22 14:S2. Jf lf UkSI CUHY AVAIU SlE hp 1 3M Environmental Laboratory- Fluorine Analytical Chemistry Team Kris Hansen - Sr. Analytical Chemist Fluorine Analytical Chemistry Team Building 2-3E-09 612-778-6018 Phase V and VI: Quantitative analysis of PFOS in historic human sera samples and in human sera samples from rural Northern China Summary: Two sets o f samples, the first with 12 samples in plastic vials (Z-set) and the second with 5 samples in glass vials (M-set), were supplied to the Environmental Lab by JeffMandel (3M Medical) on March 20,1998. For both rots, sample volume varied fiom 0.2 to 1.0 mL o f human sera. A volume between 0 .4 -1 .0 mL o f earn sample was extracted using an ion-pairing reagent Six samples in the Z-set did not contain enough volume for extraction; in these cases, two samples (e.g. ZZA6 and ZZA8) were combined and extracted as one sample. The extracts were analyzed quantitatively for perfluorooctane sulfonate (PFOS) by high-pressure liquid chromatography-electrospray mass spectrometry (HPLC-ESMS). Analyte identification was verified by comparison o f molecular ion-HFLC retention time o f the extracted analyte and standard material. Samples were quantitatively evaluated against a specially prepared, fivepoint extracted curve (1^2 = 0.999). Each sample extract was analyzed twice. Calibration checks, . analyzed every 5 samples, were within 6% of expected values. The identity o f PFOS in several randomly selected samples in this phase (M-set only) was confirmed using HFLC-ESMSMS. In all cases, the analyte exhibited a characteristic retention time, a characteristic primary ion, and five characteristic secondary ions. H istoric samples The sera samples in the M-set, collected from donors in the US in the 1960s, had an average FFOS concentration o f 33.4 ppb; the samples ranged in PFOS concentration from 11,8.to 59.4. Foreign samples The 9 samples in the Z-set (including 3 composite samples) were collected from donors in rural China and were determined to have a much lower concentration ofFFOS than any human sera data set analyzed previously. Six samples did not contain FFOS above the detection limit of this analytical method (1 ppb). The remaining 3 samples evidenced FFOS concentrations less than 3 ppb, the lim it o f quantitation. With the exception o f one human sera sample collected in the 1980s (W ill 0002), these are the only human samples determined to contain less than 10 ppb PFOS. Experimental summary: Sample preparation: Ion-pairing extraction In a pH controlled environment, an ion-pairing reagent, tetrabutyl ammonium sulfate (TBA), is used to extract the analyte from the matrix. Anionic compounds, like the cationic reagent, selectively, targets PFOS and perfluorooctanoate (POAA). Subsequent to the formation o f the TBA-anion pair, the analyte is transferred to a non-polar organic solvent (ethyl acetate), dried, and reconstituted in methanol for MS analysis. HPLC: Characteristic retention timesfo r PFOS In HPLC, an aliquot o f the extract is injected and passed through a chromatographic column. Based on the affinity o f the analyte for the stationary-phase in the column relative to the liquid mobilephase passing through the column, the analyte is retained for a characteristic amount o f time. For example, in a standard solution, PFOS may elute at 10.5 minutes. Retention times between a standard 05/07/98 004381 1 PFOS solution and the analyte extracted from sera in this analysis were matched to within 1% on the HPLC system. ES/MS: Detection and monitoring o fthe molecular ion Analysis o f PFOS standards indicates that the primary ion characteristic o f PFOS is m /z = 499 amu, corresponding to the mass o f the anionic surfactant (C*Fi7S03-). This ion was monitored selectively to maximize sensitivity. A scan o f m/z=100 to 1210 (negative only) was also collected. H PLC ES-M S/M S: Detection and monitoring o f characteristic secondary ions ES-MSMS is very similar to ESMS, except that it adds an additional dimension o f certainty'to compound identification. As in ESMS, a characteristic ion is selected After selection, the ES-MSMS characterizes the ion further by smashing it apart with high-energy gas. As a result o f the washing, secondary ionic fragments (daughter ions), characteristic of the molecule, are created and detected For example, for PFOS analysis, ion 499 is selected as the characteristic primary ion. This ion is smashed into other ions si*ch as 80 amu (corresponding to SOj"), 99 amu (corresponding to FSQO, 130 amu (corresponding to CTiSOj"), 180 amu (CJE\S03'), and 230 amu (C & SQ O . Each o f these secondary fragments is detected at the detector. Quality control summary: Due to sample limitations, sera samples were not extracted in duplicate, with 1 exception. For this same reason, matrix spikes could not be evaluated for the samples. Each sample extract was analyzed in duplicate. One sample, M001278 contained enough material for two extractions. The presented results for this sample are based on duplicate analysis o f each of the two extractions. Methanol blanks were analyzed periodically to ensure complete isolation of the sample. A mid-level (24 ppb) quality control sample was analyzed every 5 samples. Because human sera without PFOS is not available, samples .of pooled sera purchased from Sigma were pre-extracted spiked with a standard concentration o f PFOS, and re-extracted. This pre-extracted sera is understood to be very similar to the sample matrix. This five point extracted curve was analyzed before and after the samples. Quantitation o f the samples is based on the linear regression equation derived from the average of the two bracketing curves (r*2 '= 0.999). Specific QC parameters are available on the appended results table. U m it o f Quantitation The current analytical method has been used to evaluate a large range o f PFOS levels extracted from sera; the method is not optimized for precise low-level (< 5 ppb) quantitation, but instead for samples in the range o f 20-60 ppb. Given the concentrations chosen for this optimization o f the standard curve, while there is a statistical difference for samples evaluated from 3-100 ppb, there is no statistical difference between samples quantitated from 1-3 ppb. The quantitation does not become statistically dfendable until 3 ppb; thus this value represents the lim it o f quantitation for this method. Samples designated <LOQ are estimated to contain 1-3 ppb PFOS. If further accuracy and precision in quantitation o f the low-level Z-set samples (samples less than 3 ppb) is required, a special low-level quantitation system will be required. Lim it o fDetection The limit o f detection for this analytical method is based on analyst recognition o f the distinct peak shape resulting from PFOS analysis. PFOS standard material and PFOS extracted from sera analyzed by the conditions reported here, show near baseline resolution of 1 branched and 1 linear PFOS isomer. The apex o f each peak in the HPLC-trace occurs at a specific, reproducible retention time and the ratio o f the branched peak to the linear peak is fairly consistent down to the low standard, 1.0 ppb. Those samples that a) had PFOS levels evaluated at less than the limit o f quantitation and b) evidenced the distinct peak shape, were designated < LOQ. Samples that a) have PFOS levels evaluated at less than the limit of quantitation and that b) do not reflect the unique PFOS peak shape, were designated < LOD. These samples are estimated to contain from 0-1 ppb PFOS. 05/07/98 004382 2 / Instrumental specifics: HPLC system 1 Hewlett Packard Series 1100 Liquid Chromatograph Column:Keystone Betasil Q* | 2 x 100 mm I 5 pm particle size Flow rate: 300 pi/m in. I Solvent A: 2.0 mM Ammonium Acetate I Solvent B: Methanol Solvent Gradient: 45 to 90 %B in 9.50 m int I Hold at 90 %B for 3.50 m int ' Return to ^ ^ ^ B in 1.50 m int Hold at45 S B for 3.5 m int j. Injection volume: 10 pL j Injections / sample: 2 Electrospray m ass spectrometer Mieromass Platform n API Mass Spectrometer, "Chide" MassLynx 2.4 Software Cone voltages: -20v, -60v j Mode: electrospray negative * Source temperature: 115 C Analyzer vacuum pressures: 0.000043 mBar, 0.000079 mBar 1 Ions: 369,413,499,427 J Electrode: cross-flow ] ] .. J 05/07/98 004333 Fluorine Analytical Chemistry Team KJ.Hcmatn 778-6018 Samples Received: 3/6/98 Samples Extracted: 3/9/98 Samples Analyzed: 3/11 - 3/12/98 Analysis by: Calibration Curve Range: r*of Calibration Curve: LOD1: LOQ2: ESMS (-) lppb - 49ppb 0.998 l ppb 3 ppb PHASE V and VI: Analysis of foreign (rural Northern China) and historic sera samples (1960) Cone, of PFOS Sample ID (ppb+B40) ZZAI <LOD ZZA2 <LOD ZZA3 <LOQ .V ZZA4 <LOQ ZZA3 <LOD ZZA6 and 8 <LOD ZZA7 <LOD ZZA9 and 10 <LOQ ZZA11 and 12 <LOD 1 Sample ID Cone, of PFOS (ppb) m001205 M 001278 MO1202 MOO1221 24.4+ /-0.4 48.4+ /-1.3 23.2 +/- 0.2 11.8 4 /-0 .1 MOO1208 59.4+ /-0.1 QC Results QC Check: Calibration Check (QC1) Calibration Check (QC2) % Recovery 104% 106% Branched borner Abundance Percentages Sample Pool Standard Material Individuals Sampled in 1998 3M Cottage Grove Plant Workers Individuals Sampled in 1970's Pooled Sera Samples % Abundance of Branched Isomers (Average) 22% 34% 45% 37% 31% % Abundance o f Branched Isomers (Std. Dev.) 3% 5% 4% 5% 5% n (# sampled in pool) n/a 31 3 12 11 1 - LOD * Limit o f Detection 2- LOQ * Limit of Quantitation 3M Environmental Lab 5/7/98 0 0 4 3 8 4 pHASEVandVI Q uantify Cea^ound Suanary ta p o rt Saapl. U a t : C:\MASSL7iac\PSnsSS.fllO\SANrUDS\03ai9S Laat aodifled: k m N u U 14:15:9 li. Method: C:\MMStnx\fmat5.M0MflETan\! Laat Modified: (at Mac 21 1:1:42 ! Job Cod: Printod: Sun Mar 23 14:32:37 ISSI Compound 1: 1m s 9 Naw type 1 03219801 Analyte 2 03219802 Analyta 3 03219803 Analyte 4 03219804 Standard S 03219805 Standard 6 03219803 Standard 1 03219807 Standard 1 03219808 Standard 9 03219809 Standard 10 03219810 Analyta 11 03219811 Analyta 12 03219812 Analyta 13 03219813 Analyta 14 03219814 Analyta 15 03219815 Analyta 13 03219813 Analyta 17 03219817 Analyta 11 03219818 Analyta 19 03219819 Analyta 20 03219820 Analyta 21 03219821 Analyta 22 03219822 Analyta 23 03219823 QC 24 03219824 Analyta 25 03219825 Analyta 23 03219823 Analyta 27 03219827 Analyta 28 03219828 Analyta 29 03219829 Analyta 30 03219830 Analyta 31 03219831 Analyta 32 03219832 Analyta 33 03219833 Analyta 34 03219834 Analyta 35 0321983S QC 3 0321983 Analyta 37 03219837 Analyta 38 03219838 Analyta 39 03219839 Analyta 40 03219840 Analyta 41 03219841 Analyta 42 03219842 Analyta 43 03219843 Analyta 44 03219844 Analyta 45 03219845 Analyta 43 0321984 Analyta 47 03219847 Analyta 48 03219848 Analyta 49 03219849 Analyta 50 03219850 Analyta 51 03219851 Standard 52 03219852 Standard 53 03219853 Standard 54 03219854 Standard 55 03219855 Standard 53 03219856 Standard Saapl. Tact HeON Slant H20 blank-1 Sara blank-1 0.* ppb PC M U 4,90 ppb PC M U 24.1 ppb PC M U 49.0 ppb PC M U 9.2 ppb PC M U 142 ppb PC Mix XeOU Blank M00120S M00120S (2) H001274-1 H001274-1 |2) M001279-2 M 0 0 1271-2 (2) M01202 M01202 (2) M001221 M001221 t 2 ) V M00120S 'v/ M00120S (2) Cal check, 24.9 ppb Pros MaOM blank ZZAI ZZAI (2) ZZA2 ZZA2 (2) ZZA3 ZZA3 (2) ZZA4 ZZA4 <2) ZZA5 ZZA5 (2) Cal ehack, 24.8 ppb PF0S MaOH 2ZA4 and 9 ZZA4 and 9 (21 ZZA7 ZZA7 (2) ZZA9 and 10 ZZA and 10 (2) ZZA11 and 12 ZZA11 and 12 (2) AZ2C-1 AZ24-1 (2) HaOH blank MaOH blank H20 blank-1 Sara blank-1 0.994 ppb PC Mix 4.90 ppb PC M U 24.9 ppb PC M U 49.0 ppb PC Mix 94.2 ppb PC Mix 142 ppb PC M U BEST COPY AVAILABLE Area 4292 3239 2993 SOIS 10444 41137 74119 142797 193749 3297 39473 34944 74992 74499 71794 72379 37934 37444 21224 21014 90047 49749 41304 2421 3310 3397 3277 3300 5599 5721 7104 7059 5044 5340 41433 2900 4443 5071 4141 4207 4431 4444 5342 5241 44210 44254 2403 2744 3397 3223 5041 10442 40944 75059 143153 203144 Cono. Oay Plaga 0.2 bb 0.00 bb 0.00 MM 0.7 23 HM .S3 f Ml 25.72 4 MM 48.51 - l Ml 95.96 0 Ml 131.18 -8 MOC 0.00 bb 24.71 Ml 24.15 49.11 MM HM S0.1 MM 48.90 Ml 47.45 23.51 Ml MM 23.20 Ml 51.9 Ml 11.82 Ml 59.54 Ml 59.34 Ml 25.84 4 Ml 0.00 bb 0.00 Ml 0.00 Ml 0.00 Ml 0.00 Ml 1.17 Ml 1.25 Ml 2.21 2.17 Ml KK 0.80 Ml 0.99 Ml 26.20 6 Ml 0.00 bb 0.67 Ml 0.80 Ml 0.19 Ml 0.20 Ml 1.88 MK 1.78 Ml 0.99 Ml 0.93 27.85 Ml MM 27.88 Ml 0.00 Ml 0.00 Ml 0.00 bb 0.00 Ml 0.81 1 9 Ml 4.54 7 Ml 25.59 3 Ml 49.16 0 Ml 96.21 0 MM - 137.66 -3 MMX - .** Quantify C ulttontlon M port C alibration! ci\M M sm xM A su.pno\ainvtsa\pro t e a t n o d lflu d t aim Nur 23 14:4:54 ! T in ta d : lun >*x 22 14:22:3 1H 3M Environmental Laboratory- Fluorine Analytical Chemistry Team Kris Hansen - Sr. Analytical Chemist Fluorine Analytical Chemistiy Team Building 2-3E-09 612-778-6018 Phase VII: US Geographical Distribution Study - PFOS levels in human sera Summary: Fifty-five samples of cold human sera in plastic centrifuge tubes were supplied to the Environmental Lab by JeffMndel (3M Medical) on March 19 and March 20,1998. The samples, collected from sites around the United States, contained several mLs of sera. One mL aliquots o f sera were removed from each sample and extracted with an ion-pairing reagent; nearly all o f the samples were extracted in duplicate. The Sera extracts were analyzed quantitatively for perfluorooctane sulfonate (PFOS) by high pressure liquid chromatography-electiospray mass spectrometry (HPLC-ESMS). Analyte identification was verified by comparison of molecular ion-HFLC retention time o f the extracted analyte and standard material. Samples were quantitatively evaluated against a specially prepared, five-point extracted curve (r*2 - 0.996). A matrix spike calibration check, analyzed in the middle o f the analysis sequence, was within 1% of expected values. The sera samples in the geographic distribution study had an average PFOS concentration o f 28 ppb; the samples ranged in PFOS concentration from 9 ppb to 59 ppb. The percent abundance of the branched chain isomer was 41 +/- 4%. In standard solutions of PFOS, the branched chain accounts for 21.7 +/-3.1% afthe total concentration. A detailed summary of the results is attached. The identity o f PFOS in several randomly selected samples in this phase (M-set only) was confirmed using HFLC-ESMSMS. In all cases, the analyte exhibited a characteristic retention time, a characteristic primary ion, and five characteristic secondary ions. Experimental summary: Sample preparation: lon-pairing extraction In a pH controlled environment, an ion-pairing reagent, tetrabutyl ammonium sulfate (IB A ), is used to extract the analyte from the matrix. Anionic compounds, like PFOS and perfluorooctanoate (POAA), are selectively targeted by the cationic reagent Subsequent to the formation o f the TBA-anion pair, the analyte is transferred to a non-polar organic solvent (ethyl acetate), dried, and reconstituted in methanol for HPLC-ESMS analysis. HPLC: Characteristic retention timesfo r PFOS In HPLC, an aliquot of the extract is injected and passed through a chromatographic column. Based on the affinity of the analyte for the stationary-phase in the column relative to the liquid mobilephase passing through the column, the analyte is retained for a characteristic amount o f time. For example, in a standard solution, PFOS may elute at 10.S minutes. Retention times between a standard PFOS solution and the analyte extracted from sera in this analysis were matched to within 1% on the HPLC system. ES/MS: Detection and monitoring o f the molecular ion Analysis of PFOS standards indicates that the primary ion characteristic o f PFOS is m/z - 499 ami* corresponding to the mass of the anionic surfactant (CsFnSCb-). This ion was monitored selectively to maximize sensitivity. A scan of m/z" 100 to 1210 (negative only) was also collected. 0043S7 05/07/98 1 H PLC ES-M S/M S: Detection and monitoring o fcharacteristic secondary ions ES-MSMS is very similar to ESMS, except that it adds an additional dimension of certainty to compound identification. As in ESMS, a characteristic ion is selected. After selection, the ES-MSMS characterizes the ion further fay smashing it apart with high-energy gas. As a result of the smashing, secondary ionic fragments (daughter ions), characteristic of the molecule, are created and detected. For example, for PFOS analysis, ion 499 is selected as the characteristic primary ion. This ion is smashed into other ions such as 80 amu (corresponding to SQO, 99 amu (corresponding to FSQj"), 130 amu (corresponding to CF2SO30, 180 amu (C2F4SO3O. and 230 amu (CjFiSQO. Each of these secondary fragments is detected at the detector. Quality control summary: In most cases, the sera samples were extracted in duplicate; for those samples that were, standard deviation values are included in the table of results. Matrix spikes were prepared for several sample, too. If available, the matrix spike recoveries are included in the results table. A mid-level (49 ppb) quality control sample was analyzed; recovery for the matrix spike was determined to be 101%. Because human sera'without FFOS is not available, samples of pooled sera purchased from Sigma were pie-extracted, spiked with a standard concentration of PFOS, and re-extracted. This pie-extracted sera is understood to be very similar to the sample matrix. This five point extracted curve was analyzed before and after the samples. Quantitation o f the samples is based on the linear regression equation derived from the average o f the two bracketing curves (i*2 - 0.996). Specific QC parameters are available on the appended results table. Lim it o f Quantitation The current analytical method has been used to evaluate a large range o fPFOS levels extracted from sera; the method is not optimized for precise low-level (< S ppb) quantitation, but instead for samples in the range o f20-60 ppb. Given the concentrations chosen for this optimization o f the standard curve, while there is a statistical difference for samples evaluated from 3-100 ppb, there is no statistical difference between samples quantitated from 1-3 ppb. The quantitation does not become statistically defensible until 3 ppb, thus this value represents the limit of quantitation for this method. Samples designated <LOQ are estimated to contain 1-3 ppb PFOS. If further accuracy and precision in quantitation o f the low level Z-set samples (samples less than 3 ppb) is required, a special low-level quantitation system will be required. Lim it o fDetection The limit o f detection for this analytical method is based on analyst recognition of the distinct peak shape resulting from PFOS analysis. PFOS standard material and FFOS extracted from sera analyzed by the conditions reported here, show near baseline resolution o f 1 branched and 1 linear FFOS isomer. The apex of each peak in the HPLC-trace occurs at a specific, reproducible retention time and the ratio of the branched peak to the linear peak is fairly consistent down to the low standard, 1.0 ppb. Those samples that a) had PFOS levels evaluated at less than the limit o f quantitation and b) evidenced the distinct peak shape, were designated < LOQ. Samples that a) have PFOS levels evaluated at less thqn the lim it of quantitation and that b) do not reflect the unique PFOS peak shape, were designated < LOD. These samples are estimated to contain from 0-1 ppb FFOS. Instrumental specifics: HPLC system Hewlett Packard Series 1100 Liquid Chromatograph Column:Keystone Betasil Cn 2 x 100 mm 5 pm particle size 0043S 8 05/07/98 2 Flow rate: 300 pL/min. Solvent A: 2.0 mM Ammonium Acetate Solvent B: Methanol Solvent Gradient: 45 to 90 %B in 9.50 mins. Hold at 90 %B for 3.50 mins. Return to 45 %B in 1.50 mins. Hold at 45 %B for 3.5 mins. Injection volume: 10 pL Injections/sample: 1 Electrospray mass spectrometer Micromass Platform n API Mass Spectrometer, "Chick" MassLynx 2.4 Software Com voltages: -60v Mode: electrospray negative Source temperature: 115 C. Analyzer vacuum pressures: 0.000043 mBar, 0.000079 mBar Ions: 499,427 Electrode: cross-flow 05/07/98 004389 3 3M Environmental Lib BEST COPY AVAILABLE Fluorine Analytical Chemistry Team Kjjtmm, 77*- Sam plet Received: 3/19/98 Semple E xtracted: 3/19/98 Sample Analyzed: 3/20*3/33/98 A aatyrit by: O dlbratlon Curve Range: r* ofC alibration C arve: LOD1: LOQ*: PHASE VII Results: US geographical distribution study - PFOS levels in human sera Sam ple ID AKOl AK02 AK03 AK04 AZ026 AZ027 AZ028 CAQ29 CA030 CA031 DE016 DE017 DE018 DE019 DE020 IA013 IA014 IA015 LA47 LA048 LA049 M I03 M104 M I05 M I06 MO07 MO08 MO09 M S50 [PFOS] 21 31 23 19 28 37 47 23 23 26 24 32 21 21 23 40 29 28 49 39 50 31 18 22 31 24 31 35 35 Q C Results QC Check: Calibration Check (QC1) Std. Dev. 0.4 1.2 NA NA 2.5 6.4 5.8 4.1 0.7 0.0 4.5 2.2 0.1 NA 0.3 0.1 4.2 3 .0 7.4 1.9 7.9 NA NA 0.9 1.1 4.9 6.6 0.6 5.3 ! 1 Semple ED [PFOS] (PPb) MS51 MS52 MT41 MT42 MT43 NB21 NR22 NE23 NB24 NB25 NJ010 56 27 26 24 18 20 12 27 9 19 27 NJ011 NJ012 NV44 NV45 20 24 28 25 NV46 SB38 SB39 SB040 27 13 14 16 SC032 SC033 SC034 TX035 TX036 56 47 52 32 26 TX037 WY53 29 35 WY54 WY55 13 20 % Recovery 101% Branched Isomer Abundance Percentages Sample Fool Standard M aterial Individual! Sampled in 1998 3M Cottage Crave Plant Worker, 1998 Individuals Sampled in 1970s and 80s Pooled Sera Samples from 1997-98 Foreign Sera Samples, 1984 and 1994 Individuals Sampled in late 1950 US Geographical Study Samples, 1998 % Abundance of Branched Isomers (Average) 22% 34% 45% 37% 31% na 38% 41% % Abundance of Branched Isom eri (Std. Dev.) 3% 5% 4% 5% 5% na 4% 4% n/a 31 3 12 11 na 5 55 SU. Dev. 5.1 9.3 4.1 33 3a ij 1.0 6 .7 2 .0 0.8 5 .9 0 .8 2 .0 0 .8 0.1 6 .6 0 .2 0 .7 0.6 1.8 2 .8 2.1 11.2 1.2 3 .6 '. 0 .7 0 .4 1.3 1 LOD " Limit o f Detection 2 - LOQ " Limit ofQuantitation ESMS (-) lppb to 248 ppb 0.996 1PPb SPI* 004390 5/7/98 PHASE v n Q u a n tify Compound flu an ary R a p o r t aanpla U lt: C:\KMiLYHX\PHASH.PRO\BAHPLXDB\PHAJX Laat aodiflad: Thu Apr 0 0:S:1 X M t Mathod: C:\KAMI.YMX\PHAai(.fM3\MTIIDa\m Laat aodiflad: Thu Apr 0 10:00:3 t H I Job Coda: Prlntad: Thu Apr 09 10:11:39 1999 Compound 1: PfOS 1 NaM Typ* 1 04079801 Analyte 2 04079802 Analyte 3 04079803 Analyte 4 04079804 Analyte 5 04079805 Analyte 6 04079806 Standard 7 04079807 Standard 8 04079808 Standard 9 04079809 Standard 10 04079810 Standard 11 04079811 Standard 12 04079812 Standard 13 04079813 Standard 14 04079814 Standard IS 04079815 Standard 16 04079816 Standard 17 04079817 Standard 18 04079818 Analyte 19 04079819 Analyte 20 04079820 Analyte 21 04079821 Analyte 22 04079822 Analyte 23 04079823 Analyte 24 04079824 Analyte 25 04079825 Analyte 26 04079826 Analyte 27 04079827 Analyte 28 04079828 Analyte 29 04079829 QC 30 04079830 Analyte 31 04079831 Analyte 32 04079832 Analyte 33 04079833 Analyte 34 04079834 Analyte 35 04079835 Analyte 36 04079836 Analyte 37 04079837 Analyte 38 04079838 Analyte 39 04079839 Analyte 40 04079840 Analyte 41 04079841 QC 42 04079842 Analyte 43 04079843 Analyte 44 04079844 Analyte 45 04079845 Analyte 46 04079846 Analyte 47 04079847 Analyte 48 04079848 Analyte 49 04079849 Analyte 50 04079850 Analyte 51 040798S1 Analyte 52 04079852 Analyte 53 04079853 QC 54 04079854 Analyte 55 04079855 Analyte 56 04079856 Analyte 57 04079857 Analyte SI 04079858 Analyte 59 04079859 Analyte 60 04079860 Analyte 61 04079861 Analyte 62 04079862 Analyte 63 04079863 Analyte 64 04079864 Analyte 65 04079865 Analyte 66 04079866 Analyte 67 04079867 Analyte 68 04079868 Analyte 69 04079869 Analyte 70 04079870 Analyte 71 04079871 Analyte 72 04079872 Analyte 73, 04079873 Analyte 74 04079874 Analyte 75 04079875 Analyte 76 04079876 Analyte 77 04079877 Analyte 78 04079878 Analyte 79 04079879 Analyte 80 04079880 Analyte 81 04079881 Analyte Sample Text Meoh Blank Hater blank inj 1 Hater blank in] 2 HS Blank in] 1 HS Blank in] 2 0.996 ppb inj-1 0.996 ppb inj-2 4.90 ppb in]'! 4.90 ppb In]-2 24.8 ppbinj-1 24.8 ppb inj-2 49.0 ppb inj-1 49.0 ppb iaj-2 96.2 ppb inj-1 96.2 ppb in]2 142 ppbinj-2 142 ppbinj-1 neoh blank SB-11 HS04079 in] 1 SC-11 HS04078 in] 2 SC-12 HS04078 in] 1 SI-12 HS04078 in] 2 SB-13 KS04078 inj 1 SC-13 HS04078 In] 2 SC-14 HS04078 in] 1 SC-14 KSO4O70 in] 2 SB-15 KS04078 in] 1 SB-15 HS04076 in] 2 24.8 ppb cal check neoh blank SC-16 H50407I in] 1 SC-16 HS04078 in] 2 SC-17 H504079 in] 1 S B-17 H504O7I in] 2 S C - 18 HS04078 in] 1 SB-18 KS04078 in] 2 SE-19 HS04078 in] 1 SB-19 H S 04078 in] 2 SB-20 KS04078 in] 1 SE-20 H S 04078 in] 2 24.8ppb cal check neoh blank SB-1 HS04078 in] 1 SE-1 HS04078 in] 2 SE-2 HS04078 in] 1 SE-2 HS04078 in] 2 SE-3 KS04078 in] 1 SE-3 K504078 in] 2 SE-4 KS0407S in] 1 SE-4 HS04078 in] 2 SE-S HS04078 in] 1 SE-5 HS04078 in] 2 24.8 ppb cel check neoh blank SE-6 H504078 in] 1 SE-6 HS04078 in] 2 SE-7 KS0407S in] 1 SE-7 KS4078 in] 2 SE-8 KS04078 inj 1 SE-8 KS04078 in] 2 3E-9 HS04078 in] 1 SE-9 HS04078 in] 2 SE-10 H S 04078 in] 1 SS-10 HS04078 in] 2 neoh blank Hater blank in] 1 Hater blank in] 2 Sera blank in] 1 Sera blank in] 2 0.996 pp b ln]-l 0.996 ppb in]-2 4.90 ppb in]-l 4.90 ppb inj-2 24.8 ppbinj-1 24.8 ppbin]-2 49.0 p pb in]-l 49.0 ppb ia]-2 96.2 ppb inj-1 96.2 ppb inj-2 142 ppbinj-1 142 ppbinj-2 BEST COPy AVAILABLE *ea X At 9.988 10.471 10.479 10.487 10.489 10.411 10.430 10.423 10.415 10.405 10.411 10.419 10.373 10.421 10.418 10.434 10.425 10.496 10.503 10.481 10.485 10.478 10.401 10.485 10.486 10.485 10.479 10.477 10.365 10.482 10.486 10.480 10.424 10.479 1Q.487 10.477 10.437 10.420 10.410 10.435 10.358 10.438 10.353 10.357 10.352 10.340 10.470 10.467 10.336 10.337 10.434 10.403 10.309 10.388 10.305 10.281 10.410 10.470 10.463 10.455 10.336 10.343 10.461 10.404 10.406 10.402 10.407 10.390 10.404 10.336 10.334 10.339 10.340 10.340 10.340 10.340 10.344 10.339 10.342 10.346 10.342 Area 58376 5263 4594 5092 4021 5970 5750 11454 11188 41237 41564 75645 74906 132595 127286 192332 196237 4489 0860 9246 7069 7340 6139 8473 7496 7116 7746 7431 12549 3708 7723 7611 5624 5633 5265 5330 4707 4569 3712 4205 1200* 3757 9577 9822 9759 9610 10125 10474 8107 7985 9294 9300 11515 4037 8249 7947 10501 17570 7772 7039 8225 9132 20521 20053 4307 4524 4379 4342 4493 5075 5794 11191 11290 40768 41257 68332 68993 116061 lists? 162115 165549 Cono. Dev Plag 38.04 bb 0.00 bb 0.00 bb 0.00 bb 0.00 bb 0.00 -100 MC 0.00 -100 MC 3.29 -33 MC 3.09 -37 MC 25.35 2 MC 25.59 3 MC 50.83 4 MC 50.29 3 MC 93.01 -3 MC 09.08 -7 MC 137.26 -3 bbX 140.15 -i bbX 0.00 bb 1.37 MC 1.66 MC 0.64 MC 0.24 MC 0.00 MC 0.00 MC 0.36 MC 0.08 MC 0.54 MC 0.31 MC 4.10 -03 MC 0.00 MC 0.53 MC 0.44 MC 0.00 MC 0.00 MC 0.00 MC 0.00 MC 0.00 MC 0.00 MC 0.00 bb 0.00 bb 3.70 -85 MC 0.00 MC 1.90 MC 2.00 MC 2.04 MC 1.92 Ml 2.31 MC 2.56 MC 0.81 MC 0.72 MC 1.69 MC 1.70 MC 3.34 -07 MC 0.00 bb 0.92 MC 0.69 MC 0.57 MC 7.03 MC 0.56 MC 0.61 MC 0.90 MC 1.57 bb 10.01 MC 9.66 MC 0.00 MC 0.00 bb 0.00 MC 0.00 MC 0.00 bb 0.00 -100 bb 0.00 -100 bb 3.10 -37 bb 3.17 -35 bb 25.00 1 bb 25.36 2 bb 45.42 -7 MC 45.91 -6 MC 00.77 -16 MC 10.64 -16 MC 114.88 -19 MC 117.42 -17 MC f> F O S p r tth n ir w j \ 004391 Eissssssisssssssisrsssss.ssssissssssasiasssssssssssssasssssssissisisccsssassssssgsssssssKsssi-iir^cis.-^.- l |||||r : r i r r srrr r iir r r r ^ ? ? ^ ? ^ ? ? r s r r ir r H rms2n2m2n*P 2to222 P* i iiin ! i ir shi 5n IV i n ti l * *! i : :S ;n i ill! i ili! sssssSSsrS-ssssssassH SsssSsssSSSsSssgsSH sssss^^ ls: jS l3 lg g 3 S S t3 g I S g BS im S S g I S l S Sg m S S = S 3 S g g S S ;g sS S g S ~ I^ asssssssssssscasagsascsssssasssasgsssssss sssssggssasssssgigszagsgsgassssssgscssgsggsssssESgsasssssgssagsassaassg? m isi f jlggggggggggggilgiggggggggggggggggigigigggigggggggggggggggigggggigggggggggggiUiggggggiiggggillggBgggZgggggggg* oo Co & N S aaaU l C t U t o U a *** U l t n t l M i C iV M U M fMMOMM- IW W -- Ml u t Mu l i m i ti uh >> n UH liJHIl' UU IuImH UM hf> 1 3M Environmental Laboratory- Fluorine Analytical Chemistry Team Kris Hansen - Sr. Analytical Chemist Fluorine Analytical Chemistry Team Building 2-3E-09 612-778-6018 Quantitative analysis of PFOS in historic human sera samples from Sweden Summary: Two groups of ten sera samples in plastic vials were supplied to the Environmental Lab by Jeff Mandel (3M Medical) on March 20,1998. The two groups (samples SE-1 to SE-10 and SE11-SE-20) represent two distinct time periods. Both groups c f samples originated in Sweden and in all cases, sample volume varied from 0.5 to 1.0 mL o f human sera. An exactvolume, usually the entire sample was extracted using an ion-pairing reagent The extracts were analyzed quantitatively for perfluorooctane sulfonate (PFOS) by high pressure liquid chromatography-electrospray mass spectrometry (HFLC-ESMS). Analyte identification was verified by comparison of molecular ion-HPLC retention time o f the extracted analyte and standard material. Samples were quantitatively evaluated against a specially prepared, fourpoint extracted curve (i^2 = 0.992). Each sample extract was analyzed twice. Calibration checks, analyzed every 5 samples, were between 88-106% of expected values. Description o fsamples groups: SE-1 to SE-10 and SE-11 to 20 Sera samples SE-1 through SE-10 were collected from Swedish donors more recently than samples SE-11 through SE-20. The average FFOS concentration in the first groups was determined to be 1.3 ppb; there were three samples below the limit of detection. The second set of samples had an average concentration o f 2.0 ppb. In two of these samples, FFOS concentration was determined to be less than the detection lim it of the method. All samples contained PFOS at less than 5 ppb. It was not possible to determine the isomer ratio of these low level samples. Experimental summary: Sample preparation: lon-pairing extraction r- In a pH controlled environment, an ion-pairing reagent, tetrabutyl ammonium sulfide (TBA), is used to extract the analyte from the matrix. Anionic compounds, like FFOS and perfluorooctanoate (POAA). are selectively targeted by the cationic reagent Subsequent to the formation o f the IBA -anion pair, the analyte is transferred to a non-polar organic solvent (ethyl acetate), dried, and reconstituted in methanol for MS analysis. HPLC: Characteristic retention tim esfo r PFOS In HFLC, an aliquot of the extract is injected and passed through a chromatographic column. Based on the affinity of the analyte for the stationary-phase in the column relative to the liquid mobilephase passing through the column, the analyte is retained for a characteristic amount o f time. For example, in a standard solution, FFOS may elute at 10.5 minutes. Retention times between a standard FFOS solution and the analyte extracted from sera in this analysis were matched to within 1% on the HPLC system. ES/MS: Detection and monitoring o fthe molecular ion Analysis of FFOS standards indicates that the primary ion characteristic o f FFOS is m/z * 499 amu, corresponding to the mass of the anionic surfactant (C*FnSOr). This ion was monitored selectively to maximisa sensitivity. A scan of m/z=*100 to 1210 (negative only) was also collected. In those samples 04/20/98 004394 l that were determined to contain PFOS at concentrations greater than 10 ppb, is possible to verify the identify o f the analyte using HPLC-ESMSMS to monitor four characteristic fragment daughter ions. Quality control summary: Due to sample limitations, sera samples were not extracted in duplicate. For this same reason, matrix spikes could not be evaluated for the samples. Each sample extract was analyzed in duplicate. Methanol blanks were analyzed periodically to ensure complete isolation o f the sample. Two mid-level (2.98 ppb and 4.95 ppb) qualify control samples were analyzed between every 5 samples. Because human sera without PFOS is not available, samples o f pooled sera purchased from Sigma were pre-extracted spiked with a standard concentration of PFOS, and re-extracted. This pre-extracted sera is understood to be very similar to the sample matrix. This four point extracted curve was analyzed before and after the samples. Quantitation of the samples is based on the linear regression equation derived from the average o f the two bracketing curves (r*2 " 0.992). Specific QC parameters are available on the appended results table." Lim it o f Quantitation v These low level samples were analyzed against a specially prepared low level curve. As a results, confidence limits calculated around the curve values indicate a quantifiable difference between samples evaluated at 0 and 1 ppb. For this analysis, the lim it o f quantitation and fire lim it o f detection are equivalent Lim it o fDetection The limit of detection for this analytical method is based on analyst recognition of the distinct peak shape resulting from PFOS analysis. PFOS standard material and PFOS extracted from sera analyzed by the conditions reported here, show near baseline resolution o f 1 branched and 1 linear PFOS isomer. The apex o f each peak in the HPLC-trace occurs at a specific, reproducible retention time and the ratio o f the branched peak to the linear peak is fairly consistent down to the low standard, 1.0 ppb. Those samples that a) have PFOS levels evaluated at less than the lim it of quantitation and that b) do not reflect the unique PFOS peak shape, were designated b.d.1. (below detection lim it These samples are estimated to contain less than 1 ppb PFOS. 04/20/98 004395 2 Instrum ental specifics:. HPLC system Hewlett Packard Series 1100 Liquid Chromatograph Column:Keystone Bctasil Q* 2 x 1 0 0 mm 5 pm particle size Flow rate: 300 pITmin. Solvent A: 2.0 mM Ammonium Acetate Solvent B: Methanol Solvent Gradient: 45 to 90 %B in 9.50 mins. Hold at 90 %B for 3.50 min Return to 45 %B in 1.50 mins. Hold at 45 %B fa t 3.5 mins. Injection volume: 10 pL Injections / sample; 2 \r Electrospray mass spectrometer Micromass Platform U API Mass Spectrometer, "Chide" MassLynx 2.4 Software Cone voltages: -20v, -60v Mode: electrospray negative Source temperature: 115 C. Analyzer vacuum pressures: 0.000043 mBar, 0.000079 mBar Ions: 369,413,499,427 Electrode: cross-flow Electrospray M S/M S Micromass Quattro IIAPIM SM S, "Madeline" MassLynx 3.1 Software Cone voltages: -60v Collision gas energy: 40 Mode: electrospray negative Source temperature: 115 C. . Analyzer vacuum pressures: 0.000043 mBar, 0.000079 mBar Primary Ion: 499 Daughter ions: 80,99,130 ,1 8 0 ,2 3 0 Electrode: Z-source ' . 04/20/98 004396 3 Fluorine Analytical Chemistry Team Samples Received: Samples Extracted: Samples Analyzed: 4/6/98 4/7-4/08/98 4/09-4/13/98 Analysis by: Calibration Curve Range: r*of Calibration Curve: LOD1: LOQ*: Study of PFOS levels in historical Swedish human sera samples Sample ID SE-1 SE-2 SE-3 SE-4 SE-S SE-6 SE-7 SE-8 SE-9. SE-10 [PFOS] I (PPb) 1 1 .2 + /-0 .1 1.5+/-0.1 1 .0 + /-0 .L 1.0+/-0.1 J b.d-L 1.2+ /-0.0 2.8+ /-0.0 b.d-L 2.6+ /-0.9 b.d-L I Sample ID SE-11 SB-12 SB-13 SE-14 SE-15 SE-16 SE-17 SE-18 SE-19 1 SB-20 [PFOS] Cppb) 4 .1 + /-0 .2 2 .9 + /-0 .0 2 .0 + /-0 .1 2 .8 + /-0 .0 2 .9 + /-0 .0 2 .7 + /-0 .2 1 .2 + /-0 .0 1 .2 + /-0 .0 b.d.L b.d.1. KJJSanatn 778-6018 ESMS (-) lppb to lOppb 0.992 lppb lp p b QC Results QC Check:________________________________ V Recovery Calibratimi Check, 2.98 ppb Calibration Check, 4.95 ppb Calibration Check, 2.98 ppb 106% 100% 100% Calibration Check, 4.9S ppb 96% Calibratimi Check, 2.98 ppb Calibration Check, 4.95 ppb 89% 88% Branched Isomer Abundance Percentages12 Sample Pool Standard Material Individuals Sampled in 1998 3M Cottage Grove Plant Workers, 1998 Individuals Sampled in 1970s and 80s Pooled Sera. Samples from 1997-98 Foreign Sera Samples, 1984 and 1994 Individuals Sampled in late 1950s US Geographical Study Samples, 1998 % Abundance o f Branched Isomers (Average) 22% 40% 45% 37% 31% na 38% 41% % Abundance of Branched Isomers (Std.Dev.) n (# samples^ 3% n/a 5% 31 4% 3 5% 12 5% 11 na na 4% 5 4% 55 1 - LOD = Limit o f Detection 2 - LOQ Limit o f Quantitation Page 1 Sheetl file num ber 4099807 4099808 4099809 4099810 4099811 4099812 4099813 4099818 4099819 4099820 4099821 4099822 4099823 4099824 4099825 4099826 ,4099827 4099828 4099829 4099830 4099831 4099832 4099833 4099834 4099835 4099836 4099837 4099838 4099839 4099840 4099841 4099842 4099843 4099844 4099845 4099846 4099847 4099848 4099849 4099850 4099851 4099852 4099853 4099854 4099855 4099856 4099857 4099858 4099859 4099860 sam p le d escrip tion 0.998 ppb inj 1 2.98 ppb inj 1 2.98 ppb inj 2 4.95 ppb inj 1 4.95 ppb inj 2 9.8 ppb inj 1 9.8 ppb inj 2 M eoh blank SE -11H S04078 inj 1 SE-11 H S04078 inj 2 SE -12 H S04078 inj 1 SE -12 H S04078 inj 2 SE-13 H S04078 inj 1 SE-13 H S04078 inj 2 SE 14 H S04078 inj 1 SE -14H S04078inj 2 SE -15 H S04078 inj 1 SE-15 H S04078 inj 2 M eoh blank 2.98 ppb cal check 4.95 ppb cal check M eoh blank SE -16 H S04078 inj 1 SE -16 H S04078 inj 2 SE -17 H S04078 inj 1 SE -17 H S04078 inj 2 SE -18 H S04078 inj 1 SE -18 H S04078 inj 2 SE -19 H S04078 inj 1 SE -19 H S04078 inj 2 SE -20 H S04078 inj 1 SE -20 H S04078 inj 2 M eoh blank 2.98 ppb cal check 4.95 ppb cal check M eoh blank SE-1 H S04078 inj 1 SE -1H S04078 inj 2 SE -2 H S04078 inj 1 SE-2 H S04078 inj 2 SE-3 H S04078 inj 1 SE-3 H S04078 inj 2 SE -4 H S04078 inj 1 SE -4 H S04078 inj 2 SE -5 H S04078 inj 1 SE -5 H S04078 inj 2 M eoh blank 2.98 ppb cal check 4.95 ppb cal check M eoh blank peak area 4654 8096 8247 11239 11381 17949 17582 4466 9659 9265 7886 7794 6507 6735 7684 7674 7808 7803 4133 8173 10721 3844 7766 7411 5539 5450 5380 5459 4579 4335 4018 3971. 3569 7932 10483 3377 5330 5436 5773 5878 5197 5114 5337 5127 4807 4937 3416 7479 9904 3219 [an alyte] 0 .6 3.1 3.2 5 .4 5.5 10.2 9 .9 0.5 [P F O S ]avg S t. D ev . 4 .2 3.9 4.1 0.2 3 .0 2.9 2.9 0.0 2 .0 2.1 2.0 0.1 2 .8 2.8 2.8 0.0 2.9 2.9 2.9 0.0 0 .3 3 .2 5.0 0 .0 2 .9 2.6 2.7 0.2 1.3 1.2 1.2 0.0 1.2 1.2 1.2 0.0 0.6 0.4 0.5 0 .1 0.2 * 0.1 0.2 : 0.0 0.1 3 .0 4 .8 0.3 1.1 1.2 1.2 0.1 1.4 1.5 1.5 0.1 1.0 1.0 1.0 0.0 1.1 1.0 1.0 0.1 0 .7 0.8 0.8 0.1 -0.3 2 .7 4 .4 -0 .4 % R ec,QC 106 100 100 96 89 88 004398 Page 1 Sheetl 4099861 SE -6 H S04078 inj 1 5421 1.2 4099862 SE -6 H S04078 inj 2 5480 1.2 1.2 0 .0 4099863 SE -7 H S04078 inj 1 7673 2.8 4099864 SE -7 H S04078 inj 2 7729 2.8 2 .8 0 .0 4099865 SE*8 H S04078 inj 1 4556 0.6 4099866 SE -8 H S04078 inj 2 4682 0.7 0.6 0.1 4099867 SE -9 H S04078 inj 1 6497 2 .0 4099868 SE -9 H S04078 inj 2 8183 3.2 2 .6 0 .9 4099869 S E -10 H S04078 inj 1 4312 0 .4 4099870 SE -10H S04p78inj2 4710 0.7 0.5 0 .2 4099871 M eoh blank 1555 0 .0 4099872 w ater blank inj 1 3813 0 .0 4099873 w ater blank inj 2 3891 0.1 4099874 sera blank inj 1 4273 0 .4 4099875 sera blank inj 2 4498 0 .5 4099876 0.998 ppb i j 1 5199 1.0 4099877 0.998 ppb mj 2 5054 0 .9 , .4099878 2.98 ppb inj 1 7712 - 2.8 4099879 2.98 ppb inj 2 7792 2 .9 4099880 4 9 5 ppb inj 1 10297 47 4099881 4 9 5 ppb inj 2 10556 4 .9 4099882 9.80 ppb inj 1 16807 9 .4 4099883 9.80 ppb inj 2 16888 9 .4 Page 2 004399 3M Environmental Laboratory- Fluorine Analytical Chemistry Team Kris Hansen - Sr. Analytical Chemist Fluorine Analytical Chemistry Team Building 2-3E-09 612-778-6018 Quantitative analysis o f PFOS in historic human sera samples from the University of Minnesota Summary: Ten samples were supplied to the Environmental Lab by JeffMandel (3M Medical) on April 14, 1998. The ten samples were part of a group o f samples that, after extraction, were pre-screened for FFOS levels. These ten samples were classified as "low-level" (<10 ppb); they are numbered 1 ,2A, 3A, 4 ,5 , 6A, 7A, 8, 9A, and 10A Two-one mL portions of the samples were extracted using an ion-pairing reagent The extracts were analyzed quantitatively for perfluorooctane sulfonate (FFOS) using high a pressure liquid chromatography-electrospray mass spectrometry (HPLC-ESMS). Analyte identification was verified by comparison of molecular ion-HPLC retention time of the extracted analyte and standard material. Samples were quantitatively evaluated against a specially prepared, five-point extracted curve (r*2 = 0.987). Calibration checks, analyzed every 5 samples, were between 80-108% of expected values. Description o fsamples: In all ten of these pre-screened samples, FFOS concentration was determined to be less than the detection lim it of the method, 1 ppb. Experimental summary: Sample preparation: Ion-pairing extraction In a pH controlled environment, an ion-pairing reagent, tetrabutyl ammonium sulfate (TBA), is used to extract the analyte from the matrix. Anionic compounds, like PFOS and perfluorooctanoate (POAA), are selectively targeted by the cationic reagent Subsequent to the formation o fthe TBA-anion pair, the analyte is transferred to a non-polar organic solvent (ethyl acetate), dried, and reconstituted in methanol for MS analysis. HPLC: Characteristic retention timesfo r PFOS InHPLC, an aliquot of the extract is injected and passed through a chromatographic column. Based on the affinity of the analyte for the stationary-phase in the column relative to the liquid mobilephase passing through the column, the analyte is retained for a characteristic amount o f time. For example, in a standard solution, FFOS may elute at 10.3 minutes. Retention tim es between a standafd FFOS solution and the analyte extracted from sera in this analysis were matched to within 1% on the HPLC system. ES/MS: Detection and monitoring o fthe molecular ion Analysis o fFFOS standards indicates that the primary ion characteristic of FFOS is m/z = 499 amu, corresponding to the mass of the anionic surfactant (CsF17S03-)- This ion was monitored selectively to maximize sensitivity. A scan of m/z=100 to 1210 (negative only) was also collected. Q uality control summary: Sera samples were extracted in duplicate. Methanol blanks were analyzed periodically to ensure complete isolation o f the sample and two mid-level (2.98 ppb and 4.95 ppb) quality control samples were analyzed between every 3 samples. Because human sera without PFOS is not available, samples o f pooled sera purchased from Sigma were pre-extracted, spiked with a standard concentration o f FFOS, and re-extracted. This pre-extracted sera is understood to be very similar to the sample matrix. This five 04/22/98 004400 l point extracted curve was analyzed before and after the samples. Quantitation ofthe samples is based on the linear regression equation derived from the average of the two bracketing curves (r*2 0.987). Specific QC parameters are available on the appended results table. Lim it o f Q uantitation/Lim it ofD etection These low level samples were analyzed against a specially prepared low level curve. As a results, confidence limits calculated around the curve values indicate a quantifiable difference between samples evaluated at 0 and 1 ppb. For this analysis, the limit of quantitation and the limit o f detection are equivalent Instrumental specifics: HPLC system Hewlett Packard Series 1100 Liquid Chromatograph Column:Keystone Betasil Cm 2 x 100 mm ^pm particle size Flowrate: 300pL/m in. Solvent A: 2.0 mM Ammonium Acetate Solvent B: Methanol Solvent Gradient 45 to 90 %B in 9.50 mins. Hold at 90 %B for 3.50 mins. Rehim to 45 in 1.50 mini Hold at 45 %B for 3.5 mins. Injection volume: 10 pL Injections / sample: 2 Electrospray mass spectrometer Micromass Platform n API Mass Spectrometer, "Chide" MassLynx 2.4 Software Cone voltages: -20v, -60v Mode: electrospray negative Source temperature: 115 C. Analyzer vacuum pressures: 0.000043 mBar, 0.000079 mBar Ions: 369,413,499 Electrode: cross-flow ' _i 04/22/98 004401 2 Fluorine Analytical Chemistry Team Samples Received: Samples Extracted: Samples Analyzed: 4/14/98 4/15/98 4/16-4/21/98 Analysis by: Calibration Curve Range: r*of Calibration Curve: LOD1: LOQ*: Analysis o f PFOS in historic samples of human sera (U of M) Sample ID 1 2A 3A 4 ,5 [PFOS] (ppb) <LOD < LOD <LOD C f\ <LOD <LOD Sample ID 6A 7A 8 9A ` 10A [PFOS] (ppb) CLOD CLOD CLOD CLOD CLOD KJ.Hansen 778-6018 ESM S(-) lppb to 24.9 ppb 0.987 lp p b lp p b QG Results QC Check: Calibration Check (QC1 at 2.98 ppb) Calibration Check (QC2 at 4.95 ppb) % Recovery 80% 108% 1 - LOD - Limit o fDetection ' 2 - LOQ = Limit o f Quantitation 4/22/98 004402 3M Environmental Laboratory- Fluorine Analytical Chemistry Team Kris Hansen - Sr. Analytical Chemist Fluorine Analytical Chemistry Team Building 2-3E-09 612-778-6018 Phase XL Part 2: PFOS levels in individual's human sera samples, a blind repeatability experiment Summary: Fifteen samples of human sera in plastic centrifuge tubes were supplied to the Environmental Lab by JeffMandel (3M Medical) on April 15,1998. The fifteen samples were part of a group of samples that, after extraction, were pre-screened for PFOS levels. These fifteen samples were classified as "mid- lever (>10 ppb and <100 ppb); they are numbered 1A, 2B, 2C, 4A, 5A, 5B, 6B, 7B, 8A, 9B, 9C, 10B, 11, 12, and 13. Sample o f sent from these donors have been analyzed previously by the FACT. One mL aliquots of sera were removed from each sample and extracted with an ion-pairing reagent; all o f the samples were extracted in duplicate. < ^ ~ * The sera extracts were analyzed quantitatively for perfluorooctane sulfonate (PFOS) by high pressure liquid chromatography-electrospray mass spectrometry (HFLC-ESMS). Analyte identification was verified by comparison of molecular ion-HPLC retention time of the extracted analyte and standard material. Samples were quantitatively evaluated against a specially prepared, five-point extracted curve (r*2 = 0.996). Matrix spike calibration checks, analyzed in the middle of the analysis sequence, were within 10% o f expected values. The presence of PFOS in several of the samples (1A through 6B) was verified using HPLC- ESMSMS. This technique provides an additional degree of certainty to the analyte identification by. specifically monitoring fragments daughter ions (m /z=80,99,130,180,230) characteristic of the PFOS primary ion (m/z - 499). The PFOS values in this blind reproducibility analysis are within 11% ofvalues collected previously. Specific results are attached to this report Experimental summary: Sample preparation: Ion-pairing extraction In a pH controlled environment, an ion-pairing reagent, tetrabutyl ammonium sulfate (TBA), Is used to extract the analyte from the matrix. Anionic compounds, like PFOS and perfluorooctanoate (POAA), are selectively targeted by the cationic reagent Subsequent to the formation of the TBA-anion pair, the analyte is transferred to a non-polar organic solvent (ethyl acetate), dried, and reconstituted in methanol for HFLC-ESMS analysis. HPLC: Characteristic retention timesfo r PFOS In HPLC, an aliquot of the extract is injected and passed through a chromatographic column. Based on the affinity o f the analyte for the stationary-phase in the column relative to the liquid mobilephase passing through the column, the analyte is retained for a characteristic amount of time. For example, in a standard solution, PFOS may elute at 10.5 minutes. Retention times between a standard PFOS solution and the analyte extracted from sera in this analysis were matched to within 1% on foe HPLC system. 004403 05/07/98 1 ES/MS: Detection and monitoring o f the molecular ion Analysis of PFOS standards indicates that the primary ion characteristic of FFOS is m /z = 499 amu, corresponding to the mass o f the anionic surfactant (C1F1 7 SO3 -). This ion was monitored selectively to maximize sensitivity. A scan of m/z=100 to 1210 (negative only) was also collected. ES/MSMS: Confirmation o fanalyte identification Several samples in this set were analyzed by ESMSMS to verify the identity o f the PFOS analyte ion. ES-MSMS is very similar to ESMS, except that it adds an additional dimension o f certainty to compound identification. As in ESMS, a compound specific ion is selected. After selection, the selected ion is characterized further by smashing it apart with high energy gas. As a result o f the smashing, Ionic fragments, characteristic of the molecule, are created and detected. For example, for PFOS analysis, ion 499 is selected as the compound specific primary ion. This ion is smashed into other ions such as 80 amu (corresponding to SQ ^, 99 amu (corresponding to FSO3O, 130 amu (corresponding to CFiSOj"), 180 amu (C2F4SQO, and 230 amu (C3F6SCV). I f FFOS is present in the samples, each of these secondary fragments is detected at the detector. Quality control summary: Each sera sample was extracted in duplicate; standard deviation values are included in the table of results. Two mid-level (25 ppb and 49 ppb) quality control sample were analyzed between every five samples; recovery for the six matrix spike analysis were determined to be within 10% o f expected values. Because human sera without PFOS is not available, samples o f pooled sera purchased from Sigma were pre-extracted, spiked with a standard concentration o f FFOS, and re-extracted. This pre-extracted sera is understood to be very similar to the sample matrix. This five point extracted curve was analyzed before and after the samples. Quantitation of the samples is based on the linear regression equation derived from the average o f the two bracketing curves (r*2 = 0.996). Specific QC parameters are available on the appended results table. Lim it o f Quantitation As these samples were pre-screened, it was determined that optimizing instrumental detection limits was not necessary tor accurate analysis o f these "mid-level" samples. As a result, the lim it o f quantitation was 10 ppb. All samples were quantitated well above this level. Instrumental specifics: HPLC system Hewlett Packard Series 1100 Liquid Chromatograph Column:Keystone Betasil Cu 2 x 1 0 0 mm 5 pm particle size Flow rate: 300 pi ./min Solvent A: 2.0 mM Ammonium Acetate Solvent B: Methanol Solvent Gradient: 45 to 90 %B in 9.50 mins. Hold at 90 %B for 3.50 mins. Return to 45 %B in 1.50 mins. Hold at 45 %B for 3.5 mins. Injection volume: 10 pL Injections / sample: 1 Electrospray mass spectrometer Micromass Platform n API Mass Spectrometer, "Chick" MassLynx 2.4 Software 004404 05/07/98 2 Cone voltages: -60v Mode: electrospray negative Source temperature: 115 C. Analyzer vacuum pressures: 0.000043 mBar, 0.000079 mBar Ions: 499,413,369 Electrode: cross-flow lectrospray M SM S Micromass Quattro n API Mass Spectrometer, "Madeline" MassLynx 3.1 Software Cone voltages: -60v Collision gas energy. 45 V Mode: electrospray negative Source temperature: 115 "C. Analyzer vacuum pressures: 0.000043 mBar, 0.000079 mBar Primary Ion: 499 v, Daughter ions: 80, 99,130,180,230 Electrode: Z-spray 05/07/98 00440S 3 Fluorine Analytical Chemistry Team Contact: K.J. Hansen 8-6018 Samples Received: Samples Extracted: Samples Analyzed: 4/14/98 4/15/98 4/16-4/22/98 LOD1: LOQ1: Analysis by: ESM S(-) Calibration Carve Rang 10 ppb to 96 ppb r*of Calibration Curve: 0.996 10 ppb 10 ppb PHASE IX, part 2: Blind repeatability experiment r-- n 1A 2B 2C 4A 5A 5B 6B 7B 8A fPFOS] (ppb) 28 + /-0 .3 50 + /- 3 38+ /-0.2 47 +/- 0.4 62 + /- 1 54 + /-2 44 + /-4 . 54 + /-4 78 + /-5 1 Reported Previously 28 41 37 v ' 45 54 49 42 45 67 | 0PFOS] | Reported | Previously 9B 34 + /-1 30 9C 47 + /-9 39 10B 51 + /-1 41 11 44 + /- 4 n.a. 2 79 + /- 4 * n.a. 13 70 + /- 8 n.a. 14 79 + /-2 6 n.a. 16 37 + 1-2 32 17 80 + /-1 4 73 QC Results QC Check: Calibration Check, 24.9 ppb Calibration Check, 49 ppb Calibration Check, 24.9 ppb Calibration Check, 49 ppb Calibration Check, 24.9 ppb Calibration Check, 49 ppb */ Recovery 96% 100% 92% 95% 90% 98% 1 - LOD = Limit o f Detection 2 - LOQ = Limit o f Quantitation 3M Environmental Lab 5/7/98 04406 PHASE DC part 2 3M Environmental Laboratory- Fluorine Analytical Chemistry Team Kris Hansen - Sr. Analytical Chemist Fluorine Analytical Chemistry Team Building 2-3E-09 612-778-6018 Preliminary results: Analysis of POAA in human sera samples Summary: To date, data finom nine phases of a study of fluorochemicals in human sera has been collected. PHASE I: Semi-quantitative investigation ofPFOS in samples of pooled human sera PHASE II: Quantitative analysis ofPFOS and POAA in samples o f pooled human US sera samples, 1997-1998 PHASE PI: Quantitative analysis of PFOS and POAA in samples o f individual's sera, 1998 v7 PHASE IV: Quantitative analysis ofPFOS in historical US sera samples, 1970s-80s PHASE V: Quantitative analysis o fPFOSin historical US sera samples, 1960s PHASp VI: Quantitative analysis ofPFOS in samples of sera from individuals in rural northern China, 1985-1997 PHASE VO: Geographical distribution: a quantitative analysis ofPFOS in sera collected around the United States, 1998 PHASE VCH: Quantitative analysis ofPFOS in historical samples from Sweden, specific date not supplied to Environmental Lab PHASE EX, part 1: Quantitative analysis o fPFOS in historical US samples, 1948-1951 PHASE DC, part 2: PFOS levels in individual's human sera samples; Blind Repeatability Experiment The analysis of perfluorooctanesulfonate (PFOS) has been the primary focus in these phases; quantitative data for perfiuorooctonoate (POAA) has also been collected for several phases o f the study. Due to time constraints, we have not devoted the same attention to the POAA as we have to the PFOS analysis. The following report describes observations of POAA levels in some o f the samples analyzed as part of this ongoing study. Reports detailing the PFOS analysis listed above.have been issued to Jeff Mandel in the 3M Medical Department. LOQ for POAA is 10 ppb. PHASE II: Quantitative analysis ofPFOS and POAA in samples ofpooled human sera Number of samples with POAA > LOQ = 0 Number of samples with POAA > LOD * 4 Total number of samples 12 Comments'. PHASE III: Quantitative analysis o f PFOS and POAA in samples o findividual's sera Number of samples with POAA > LOQ * 4 Total number o f samples 3 1 Comments: Data has not been examined carefully PHASE FV: Quantitative analysis ofPFOS / historical sera samples. 1970s-80s Number of samples with POAA > LOQ = 0 Total number o f samples * 10 Comments: Data has not been precisely quantitated 004407 04/24/98 1 PHASE V: Quantitative analysis ofPFOS in historical sera samples. 1960s Number of samples with POAA > LOQ = 0 Number of samples with POAA > LOD * 6 Total number o f samples = 6 Comments: Data has not been precisely quantitated PHASE VI: Quantitative analysis ofPFOS in samples o fsera from individuals in rural northern China Number of samples with POAA > LOQ = 0 Number of samples with POAA > LOD = 5 Total number of samples = 9 Comments: Data has not been precisely quantitated PHASE VII: Geographical distribution: a quantitative analysis ofPFO S in sera collected around the United States Number of samples with POAA > LOQ * 2 ( 1 at 22 ppb, 1 at 12 ppb) Number of samples with POAA > LOD 2120 Total number of samples = 55 ........ Comments: Data has not been examined carefully PHASE VHI: Quantitative analysis o fPFOS in historical samples from Sweden Number of samples with POAA > LOQ = 0 Number of samples with POAA > LOD = 0 Total number o f samples = 20 Comments: Data has not been precisely quantitated PHASE DC. Dart 1: Quantitative analysis ofPFOS in historical US samples. 1948-1951 Number of samples with POAA > LOQ 0 Total number o f samples = 20 Comments: Data has not been precisely quantitated Experimental summary: Sample preparation: lon-pairing extraction i' In a pH controlled environment, an ion-pairing reagent, tetrabutyl ammonium sulfide (TBA), is used to extract the analyte from the matrix. Anionic compounds, like PFOS and perfluorooctanoate (POAA), are selectively targeted by the cationic reagent Subsequent to the formation o f the TBA-anion pair, the analyte is transferred to a non-polar organic solvent (ethyl acetate), dried, and reconstituted in methanol for MS analysis. HPLC: Characteristic retention timesfo r POAA In HPLC, an aliquot of the extract is injected and passed through a chromatographic column. Based on the affinity of the analyte for the stationary-phase in the column relative to the liquid mobilephase passing through the column, the analyte is retained for a characteristic amount of time. For example, in a standard solution, POAA may elute at 10.5 minutes. Retention times between a standard POAA solution and the analyte extracted from sera in this analysis were matched to within 1% on the HPLC system. ES/MS: Detection and monitoring o fthe molecular ion Analysis o f POAA standards indicates that the primary ion characteristic o f POAA is m/z 413 amu, corresponding to the mass of the anionic surfactant (CjFisCQr). This ion was monitored selectively to maximize sensitivity. Fragmentation of the POAA anion can be induced; ion 369, characteristic o f the 04/24/98 004408 2 decarboxylation product, results. Both ions were monitored in most of the study phases listed above. A scan of m/z=100 to 1210 (negative only) was also collected. ES/MSMS: Confirmation o fanalyte identification Several samples in this set were analyzed by ESMSMS to verify the identify of the FFOS analyte ion. ES-MSMS is very similar to ESMS, except that it adds an additional dimension o f certainty to compound identification. As in ESMS, a compound specific ion is selected. After selection, the selected ion is characterized further by smashing it apart with high energy gas. As a result of the smashing, ionic fragments, characteristic of the molecule, are created and detected. For example, for FFOS analysis, ion 499 is selected as the compound specific primary ion. This ion is smashed into other ions such as 80 amu (corresponding to S(V ), 99 amu (corresponding to FS(V ), 130 amu (corresponding to CF2S 03'), 180 amu (C2F4SQ37and 230 amu (CsFiSOO. If FFOS is present in the samples, each of these secondary fragments is detected at the detector. , Quality control summary: When possible, sera samples were extracted in duplicate and analyzed in duplicate. Often this was not possible, as samples size and/or time were limited. Methanol blanks were analyzed periodically to ensure complete isolation of the sample. Mid-level qualify control samples were analyzed periodically through each analytical sequence, typically every five samples, to monitor instrument stability. Because human sera without FFOS is not available, samples o f pooled sera purchased from Sigma were pieextracted, spiked with a standard concentration o f FFOS, and re-extracted. This pre-extracted sera is understood to be very similar to the sample matrix. In all cases PFOS was evaluated versus an extracted; for most phases of this study, POAA was also piked and extracted to form a standard curve. In these cases, quantitation of the samples is based on the linear regression equation derived from the average of the two curves bracketing the samples. Lim it o f Quantitation The current analytical method has been used to evaluate a large range o fPFOS levels extracted 1 from sera; the method is not optimized for precise low-level (< 5 ppb) quantitation, but instead for samples with FFOS concentrations in the range o f20-60 ppb. A p ed a l, low-level curve has been prepared for precise quantitative evaluation o f FFOS levels between 1-10 ppb. As the POAA levels observed in non-plant worker sera are very low, in future analysis, POAA w ill be evaluated against a special low-level curve, similar to that prepared for FFOS. Samples designated <LQQ contain less than 10 ppb POAA. If further accuracy and precision in quantitation o f the sera samples (samples less than 10 ppb) is required, samples w ill be re-evaluated versus a special low-level quantitative curve. Lim it o fDetection The limit o f detection for this analytical method is based on analyst recognition o f the distinct peak shape resulting from POAA analysis. POAA standard material and POAA extracted from sera analyzed by the conditions reported here, show near baseline resolution o f 1 branched and 1 linear POAA isomer. The apex o f each peak in the HFLC-trace occurs at a specific, reproducible retention time. Unlike, PFOS, and the ratio o f the branched peak to the linear peak is not necessarily consistent, particularly at low levels. POAA, when present, seems to be present at such low levels that accurate characterization o f the smaller peak (the branched isomer) is very difficult Qualifications: This data will be considered preliminary until the analyte extracted from these samples can be verified as POAA using LC-MSMS techniques and until instrumental sensitivity can be optimized to more clearly identify the analyte response. The method used for the analysis of these samples has been validated by LC-MSMS previously, but confirmation o f POAA for specific samples in this eight-phase study has not been performed. 04/24/98 0044C^ 3 Instrumental specifics: HPLC system Hewlett Packard Series 1100 Liquid Chromatograph Column:Reystone Betasil Cu 2 x 100 mm 5 pm particle size Flow rate: 300 plim in. Solvent A: 2.0 mM Ammonium Acetate Solvent B: Methanol Solvent Gradient: 45 to 90 %B in 9.50 mins. Hold at 90 %B for 3.50 mins. Return to 45 %B in 1.50 mins. 1 Hold at 45 %B for 3.5 mins. Injection volume: U 0 pL Injections / sample: 2 ( Micromass riauurm ix a t i jviass Spectrometer, "Chick" MassLynx 2.4 Software Cone voltages: -20v, -60v Mode: electrospray negative Source temperature: 115 C Analyzer vacuum pressures: 0.000043 mBar, 0.000079 mBar Ions: 369,413; 499,427 Electrode: cross-flow Electrospray M SM S Micromass Quattro n API Mass Spectrometer, "Madeline" MassLynx 3.1 Software Cone voltages: -20v Collision gas energy: 45 V Mode: electrospray negative Source temperature: 115 C. . Analyzer vacuum pressures: 0.000043 mBar, 0.000079 mBar Primary Ion: 413 Daughter ions: 119,169,369 Electrode: Z-spray J 04/24/98 004410 4