Document bBywnK81pdpQz1aaob8MbXowO

A R 2 .6 - \ 0 H - M iW> CCAS t r,n f ntrr jo r .U m lyaii .Sciences Analytical Report William P. Raiford L. William Buxton Barbara S. Larsen Ana Cristina Piovan Marianne Marsi Lam H. Leung SCION ICT J Allen Tannen, CRP 711 CRP 713 E228 CRP 702 E323 E402 BMP 14/1286 TRC E301 <CU=3>> --n m oo O 70 'O p'i "T3 O --*rn < zs. O rn C~'j '--* CO tn CD DATE: 16 January 2003 TO: SITE: George Senkler Fluoroproducts, Chestnut Run 711 FROM: Mary A. Kaiser and Michael J. Michalczyk Determination of Perfluorooctanoic Acid from the Surface of Commercial Frying Pans Abstract A method was developed to determine if perfluorooctanoic acid (PFOA), a fluoropolymer polymerization aid, might be present in and extracted from the surface o f commercial frying pans coated with a fluoropolymer. A modified US FDA method was used fo r extraction and an LC/M S/M S method used fo r id e n tif cation and quantitative analysis. No PFOA was found in the water e x tra c t from the fluoropolymer surface to the lim it of quantitation in water, 50 ng/L (ppt), from two 500-mL portions of water. Request A request was made to determine if any quantifiable PFOA might be found in e x tra c t water from a fluoropolymer surface. A US Food and Drug Administration (FDA) method was modified to e x tra c t PFOA, if it were present, on the surface of the fryin g pan. Fluoropolymer dispersions are used in the commercial processes fo r coating metal. These products frequently use a fluoropolymer polymerization aid, PFOA, during th e ir manufacture. Recent studies1have suggested th a t some Page 1 : POjjP The miracles o/ science l fluoropolymer polymerization aids might persist in the environment and might e xh ibit some toxicological e ffe c ts in animals. Another re p o rt2 indicated th a t low levels of these materials might be present in human blood serum. Since PFOA is used to suspend and emulsify some polymers during th e ir manufacture, it is necessary to determine if PFOA is s till present a fte r the coating and baking processes and thus might be a vehicle fo r human exposure. Experimental Section Ten fryin g pans, five 8-inch Farberware Elite Cusine uncoated stainless steel and five 8-inch Anolon Classic hard-anodizied aluminum coated with a blend of polytetrafluoroethylene (PTFE) and PFA (branded as DuPont Autograph) were purchased from a local bed and bath specialty store. An FDA procedure3 was modified as noted. The pans were f ir s t washed as recommended by the manufacturer with approximately 50 mL of 0.5 % Palmolive detergent solution and then rinsed with three 50 mL of portions of rinse (deioinzed) water and dried with a Sonora towel. A custom-made glass lid (9" diameter x 0.125'' thickness) w ith a #19 female ground glass jo in t attached in the center of the glass (see Figure 1) and a condenser were washed with the detergent solution, rinsed with deionized water and dried. The pan was placed on a Vulcan model 38L gas stove and deionized water was added to approximately 0.635 cm (0.25 inches) from the top. The lid and condenser were placed on the pan, and two 201-g weights were added to the lid to ensure a good seal. The condenser was cooled with wet ice. The fryin g pan with water and condenser was heated until the water refluxed continuously inside the condenser (Figure 2). The heat was turned down and re flu x was allowed to continue fo r 30 minutes. Then the apparatus was removed from the burner and allowed to cool. The cooled water was then placed in each of two 500-mL wide-mouth polypropylene bottles. The lid and condenser were rinsed with deionized water, and the pan was rinsed with three 50-mL portions of deionized water. All rinse water was added to each of the two polypropylene bottles. The procedure was repeated fo r the remainder o f the fryin g pans. When testing had completed, the polypropylene bottles were placed inside a plastic liner within an insulated cooler. Ic e was placed on top of the bottles and the liner folded and taped to prevent leakage. The cooler was then sealed w ith chain-ofcustody tape and Scotch 8981 tape. The cooler was sent to Exygen Research (S tate College, PA) via Airborne Express fo r overnight delivery. All samples were received on ice at Exygen Research receiving, examined, logged in, and stored in locked storage until tra n s fe r to the Exygen Research analyst. Samples were tre ate d according to Exygen Research method 01M-008-046 entitled "Method of Analysis fo r Determination of Ammonium Perfluorooctanoate Page 2 cijyPOJl) The miracles o f science" 3 (APFO) in W ater" issued on January 11, 2002. (See Appendix 1.) Using this method, water samples were treated with 200|liL of sodium thiosulfate solution, solid-phase extracted with a C-18 reverse-phase cartridge, eluted with 40% methanol in water, then eluted with 100% methanol, and the la tte r portion collected fo r analysis by LC/M S/M 5. The treatm ent results in an eight-fold concentration prior to analysis. Results and Discussion All water from the fluoropolymer-coated fryin g pans showed no quantifiable PFOA present. W ith a lim it of quantitation (LOQ) of 50 ppt and a lim it of detection (LOD) of approximately 10 ppt per aliquot, this number corresponds to 100 ng/420.42 cm2 LOQ or 20 ng/420.42 cm2 LOD. One fryin g pan with no coating (ju st a stainless steel surface) gave a positive result (Table I ) in one instance and a nondetect (with an approximate 10 ppt lim it of detection, 1 L water) in another instance. Since PFOA is easily adsorbed onto analytical apparatus, especially fluoropolymeric parts, background levels can be problematic. PFOA can be found in analytical solvents as well as tra n s fe r lines in chromatographic systems. The background levels are usually around 10 ppt. The US EPA Guideline4 recommends th a t any measurements obtained th a t are less than five times the blank level should be reported as undetected (U) because the quantity reported cannot reliably be discriminated between the sample and the background. Due to background levels and the possibility fo r carry over or laboratory cross-contamination, the lim it of quantification was set at 50 ppt. We observed th a t the background level, carry over, or laboratory contamination sometimes exceeds the 50-ppt level. References 1. Guide to the Safe Handling of Fluoropolymer Dispersions, The Society o f the Plastics Ind ustry, Inc., 2001, p. 7. 2. K. J. Hansen, L. A. Clemen, M. E. Ellefson, and H. O. Johnson, Environmental Science and Technology, 35, 766-770 (2001). 3. US FDA Procedure, see reference 21CFR175.300(d) table2 condition o f use B. 4. " Data Validation Standard Operating Procedure fo r Contract Laboratory Program Routine Analytical Services", US EPA Region IV , Athens, Georgia, Revision 2.1, July 1997. Notebook Numbers E104373-1 to -15. Page 3 dUPOlD The miracles o f science' H Control Terms to be included PFOA, APFO, fryin g pans, extraction, LC/M S/M S, US FDA Table I Analytical Results Sample ID Pan ID PFOA (ng/L) 376603-1 376603-2 376606-1 376602-1 376609-1 376609-2 376612-1 376612-2 376615-1 376615-2 376616 610030 610035 610060 610065 610090 610095 610120 610125 610150 610156 611030 611035 611060 615030 615035 615060 615065 615090 615095 615120 615125 615150 615155 615160 611160 SS#1, 1st aliquot SS#1, 2nd aliquot SS#2, 1st aliquot 5 5 # 2 , 2nd aliquot 5 5 # 3 , 1st aliquot 5S #3, 2nd aliquot SS#4, 1st aliquot 5S #4, 2nd aliquot 5S #5, 1st aliquot 5S #5, 2nd aliquot D I water control S5 #1, 1st aliquot SS #1, 2nd aliquot Coated # 1 ,1 st aliquot Coated #1, 2nd aliquot Coated # 2 , 1st aliquot Coated # 2 , 2nd aliquot Coated # 3 ,1 s t aliquot Coated # 3 , 2nd aliquot 5 5 # 2 , 1st aliquot 55 # 2 , 2nd aliquot 55 # 3 ,1 s t aliquot SS # 3 , 2nd aliquot Coated # 4 , 1st aliquot SS # 4 , 1st aliquot SS # 4 , 2nd aliquot Coated # 5 , 1st aliquot Coated # 5 , 2nd aliquot SS # 5 , 1st aliquot SS # 5 , 2nd aliquot Coated # 6 ,1 s t aliquot Coated # 6 , 2nd aliquot SS #1, 1st aliquot SS # 2 , 2nd aliquot DI water control D I water control LOQ = 50 ppt = 50ng/L LOD = c lOppt nd nd 36.9 nd 49.0 nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd nd 223 140 nd nd nd nd nd nd nd nd nd nd nd nd nd nd = not detected Peak area original 792 714 5394 989 6802 600 478 513 415 439 233 1300 873 1284 998 1104 806 1039 811 567 448 34861 22594 309 3656 2594 3966 2538 2552 2610 4292 3317 2559 4366 2184 1858 Peak area lab replicate 626 748 5044 1119 6858 596 435 478 364 555 388 1379 1074 1353 979 1336 850 1036 668 544 702 35556 20522 321 3964 2975 3524 2998 2022 2360 4509 3649 3148 2900 2052 775 Page 4 3UJ0P The miracles o f Science- r age 5 The mirai fe* / seli aie . W W W w.>>>w.>>>w.v .Appendix 1, Exygen APPO Method TITLE Method of Analysis for the Determination of Ammonium Perfiuorooctanoate (APPO) in Water AUTHORS Joha Flaherty and Karen Kn;ha Page 6 DATE I S SUED January 10. 2002 kill The miaides o / sdente 7 SPONSOR Dupont Corporation PO BOX 80302 Wilmington, DE 19880-0302 PERFORMING LABORATORY Exygen Research 3058 Research Drive S tate College, PA 16801 METHOD NUMBER O lM -008 -04 6 Page 7 The miracles o f science - MANAGEMENT APPROVAL John Flaherty Laboratory N^anaqzr Exygen Research Richard A. Grazzini, Ph.D. President Exygen Research Mary A. Kaiser, Ph.D. Sponsor Representative Dupont Corporation Date Date Date TABLE OF CONTENTS T IT LE ......................................................................................................... MANAGEMENT APPROVAL.................................................................... TABLE OF CONTENTS........................................................................... L IS T OF FIGURES................................................................................... 1. SUMMARY........................................................................................... 2. EXPERIMENTAL COMPOUNDS........................................................ 3. CHEMICALS AND SUPPLIES........................................................... 3.1. Ch e m ic a l s ................................................................................................ 3.2. S t a n d a r d s ................................................................................ 3.3. Eq u ip m e n t a n d S u pp lie s.......................................................... 3.4. S o l u t io n s ................................................................................. 3.5. Pr e p a r a t io n o f S t a n d a r d s a n d Fo r t if ic a t io n S o l u t io n s 3.5.1. Stock Solution................................................................ 3.5.2. Fortification Solutions................................................. 3.5.3. Calibration Standards................................................... 3.5.4. Controlling Standard Bias............................................. 4. METHOD.............................................................................................. Page 8 <an&> The miracles o f science' 4.1. Flo w Dia g r a m ........................................................................................................... 4.2. S ample Pr o c e s s in g ................................................................................................. 4.3. S ample Pr e p a r a t io n ............................................................................................... 4.4. Ex t r a c t io n ................................................................................................................ 4 .5 . Q u a n t it a t io n ......................................................................................................... 4.5.1. LC/M S/M S System and Operating Conditions (Electrospray).. 4.5.2. Example MS Operating Parameters............................................ 4.5.3. Calibration Curve Procedures....................................................... 4.5.4. Sample /Analysis............................................................................. 4 .6 . A cceptance Cr it e r ia .............................................................................................. 4.7. M a s s Ca l ib r a t io n Cr it e r ia ............................................................................... 4.8. Performance Cr it e r ia ........................................................................................... 4.9. M eth o d D e t e c t io n L i m i t D e t e r m in a t io n .................................................... 4.10. T im e Re q u ir e d for A n a l y s is .............................................................................. 5. CALCULATIONS................................................................................................. 6. SAFETY................................................................................................................ 7. REFERENCES....................................................................................................... 1. SUMMARY This report details a method of analysis for residues of ammonium perfluorooctanoate (APFO) in water. APFO forms perfluorooctanoic acid (PFOA) in solution. PFOA is extracted from water using Cis solid phase extraction (SPE) cartridges. Quantification of PFOA is accomplished by liquid chromatography/tandem mass spectrometry (LC/MS/MS) analysis using selected reaction monitoring (SRM). Residues of PFOA found are mathematically converted and reported as APFO. The chemical formula of APFO is given in Section 2 of this method. The proposed limit of quantitation (LOQ; the lowest fortification specified by the method which gives adequate recovery according to EPA guidelines) for this method is 50 ng/L (parts-per-trillion) for APFO. Quantification is performed using calibration standards prepared in filtered type I water and processed through the extraction procedure, identical to samples. During the extraction procedure, all samples (including standards, controls, fortifications etc.) are concentrated by a factor of eight; therefore, the smallest standard (25 ng/L) injected during the chromatographic run is equivalent to 200 ng/L (concentrated 8x). The method detection limit (MDL) is approximately 6 to 15 ng/L depending on the instrument used for analysis. This value is based on procedures found in 40 CFR Part 136 Appendix B. 2. Experimental compounds ClffiPOM) The miracles o f science' Page 9 The structure of ammonium perf luorooctanoate is given below. c f 3( c f 2) 6<:o o - n h 4+ APFO Chemical Name Molecular weight = = Ammonium Perfluorooctanoate 431 Note: The PFOA (anion) (formula weight 413) is present in solution and is used fo r the assay. 3. CHEMICALS AND SUPPLIES 3.1. Chemicals Chemical Grade Source Catalog No. Methanol (MeOH) Ammonium Acetate Water Sodium Thiosulfate HPLC Reagent Type I Reagent J T Baker Sigma-Aldrich in house JT Baker JT9093-2 A-7330 Type I water = electrical resistivity, minimum of 16.67 MQ/cm at 25C, from a Labconco WaterproTM workstation. 3.2. Standards Standard Lot Number Purity (%) Source Pentadecafluorooctanoic Acid 08316DO 96 Aldrich Chem Perfluorooctanoic Acid U14D 96 Fiuorochem USA 3. Eauioment and Supplies __________________ Equipment_________________________Supplier Balance, analytical (display atleast 0.0001 g) M ettler Vacuum pump Visiprep vacuum manifold Buchi Supelco Sep Pak Vac 6 cc (lg) tC18 cartridges Waters (p a rt# W A T036795) 50 mL disposable polypropylene centrifuge tubes VWR 15 mL disposable polypropylene centrifuge tubes VWR Disposable glass micropipets (50-100 pL, 100-200 Drummond (VWR) M-) 25 mm diameter glass fib re acrodisc 0.45 p (cat Gelman Page 10 tSUPONP The miracles o f science" #4523) Class A pipets and volumetric flasks Hypercarb drop-in guard column (4 mm) (part # 844017-400) Stand-alone drop-in guard cartridge holder 125-mL LDPE narrow-mouth bottles HPLC pump (LC10AD) 2 mL clear HPLC vial kit (cat # 5181-3400) Standard lab equipment (graduated cylinders disposable tubes etc.) LC/MS/MS and HPLC systems various suppliers Keystone Keystone Nalgene Shimadzu HP various suppliers As described in section 4.5. Note: Equivalent materials may be substituted fo r those specified in this method if they can be shown to produce satisfactory results. 3 .4 . Notes: 1. In order to avoid contamination, the use of disposable labware is highly recommended (tubes, pipets, etc.). 2. PTFE or PTFE lined containers or equipment, including PTFE-lined HPLC vials fo r the HPLC autosampler must not be used. 3. Type I water used during the sample and standard extraction should be filte re d through a Hypercarb guard column using a HPLC pump, as described in section 4.2. This water is re ferre d to as "filte re d type I water'1, here a fte r in this report. 4. I t is necessary to check the solvents (methanol) fo r the presence o f contaminants by LC/MS/MS before use. Certain lot numbers have been found to be unsuitable fo r use. 5. Use disposable micropipets or pipets to aliquot standard solutions to make calibration standards and sample fortifications. Solutions (1) 50 mM ammonium acetate solution is prepared by weighing 3.86 g of ammonium acetate and dissolving in 1 L of type I water. Dilute the 50 mM solution by a facto r of 25 to make the 2 mM ammonium acetate solution used fo r mobile phase A. (2) 40% methanol is made by measuring 400 mL methanol and adjusting the volume to 1 L with filte re d type I water. (3) 250 mg/mL sodium thiosulfate solution is made by dissolving 25 g in 100 mL filte re d type I water. Note: The aforementioned examples are provided fo r guidance, alternative volumes may be prepared as long as the ratios of the solvent to solute ratios are maintained. 3 .5 . Preparation o f Standards and Fortification Solutions Analytical standards are used fo r three purposes: Page 11 The m irades o f science" jX 1. Calibration Standards - These standards are prepared in filte re d Type I water and are used to calibrate the response o f the detector used in the analysis. 2. Laboratory Control Spikes - These fortificatio n s are prepared at concentrations corresponding to the LOQ and lOx LOQ and are used to determine analytical recovery. Laboratory control spikes are prepared in filte re d Type I water. 3. M atrix Spikes - These fo rtificatio n s are prepared by spiking into the fie ld samples at concentrations ranging from 500 ng/L up to 500,000 ng/L. M atrix spikes are used to evaluate the e ffe c t of the sample matrix on analytical recovery. The absolute volumes of the standards may be varied by the analyst as long as the correct proportions o f solute to solvent are maintained. 3.5.1. Stock solution Prepare a stock solution of 100 fjg/mL of PFOA by weighing out 10 mg o f analytical standard (corrected fo r purity) and dilute to 100 mL with methanol in a 100-mL volumetric flask. The stock solution (in 125-mL LDPE bottles) is to be stored in a re frig e ra to r at 2C to 6C and is stable fo r a maximum period of 6 months from the date of preparation. 3.5.2. Fortification Solutions a. Prepare a fo rtifica tio n standard of 1.0 pg/mL (1000 ng/mL) of PFOA by adding 1.0 mL of the 100 pg/mL stock solution to a 100-mL volumetric flask and bring up to volume with methanol. b. Prepare a fo rtifica tio n standard of 0.1 pg/mL (100 ng/mL) by diluting 10.0 mL of the 1.0 pg/mL solution to 100 mL with methanol in a volumetric flask. C. Prepare a fortifcation standard of 0.01 pg/mL (10 ng/mL) by diluting 10.0 mL of the 0.1 /jg/mL solution to 100 mL with methanol in a volumetric flask. Example', two hundred microliters of the 10 ng/mL solution spiked into 40 mL of water is equivalent to a 0.05 ppb (50 ng/L) fortificatio n . Store all fo rtifica tio n standard solutions in a refrigerator (in 125-mL LDPE bottles) at 2C to 6C fo r a maximum period of 3 months from the date o f preparation, a fte r which time it is necessary to make new standards using the stock solution. Note also tha t additional concentrations may be prepared if necessary. 3.5.3. Calibration Standards LC/MS/MS calibration standards are prepared in type I water filte re d through a hypercarb drop-in guard cartridge. The calibration standards are processed through the extraction procedure, identical to samples. The extracted standards may be used fo r a period of two weeks when stored refrigerated (at 2C to 6C). The following is a typical example: additional concentrations may be prepared as needed. Page 12 The miracles o f science Concentration Fortificati Volume of of Fortif cation on Volume Fortified Solution OjL) Control Sample (ng/mL) (mL) 0 0 40 10 100 40 10 200 40 10 400 40 100 100 40 100 200 40 100 400 40 Final Concentration of Calibration Standard (ng/L)* 0 25 50 100 250 500 1000 Calibration Standard ID (example) Xcmmddyy-0 Xcmmddyy-1 Xcmmddyy-2 Xcmmddyy-3 Xcmmddyy-4 Xcmmddyy-5 Xcmmddyy-6 * The extracted concentration of the calibration standard is equal to 8x its initial concentration, due to the concentration of the standard during the extraction (SPE). XC = extracted calibration standard. A zero standard solution (reagent blank) must be prepared with each set of standards extracted. This standard is used only to access the reagents used to prepare the standards and is not included as part o f the calibration curve. Store all extracted calibration standards in 15-mL polypropylene tubes at 2C to 6C, up to two weeks. 3.5.4 Controlling Standard Bias Prepare a separate stock solution of 100 pg/mL from a d iffe re n t PFOA neat standard from a d iffe re n t source by weighing out 10 mg of analytical standard (corrected fo r purity only) and dilute to 100 mL with methanol in a 100-mL volumetric flask. The stock solution (in 125-mL LOPE bottles) is to be stored in a re frig e rato r at 2C to 6C and is stable fo r a maximum period of 6 months from the date of preparation. Prepare a fo rtifica tio n standard of 1.0 pg/mL (1000 ng/mL) o f PFOA by adding 1.0 mL of the 100 pg/rr\L stock solution to a 100-mL volumetric flask and bring up to volume with methanol. From tha t solution, prepare a fo rtifica tio n standard o f 0.1 pg/mL (100 ng/mL) by diluting 10.0 mL of the 1.0 p g/m l solution to 100 mL with methanol in a volumetric flask. Prepare a 250 ng/L extracted calibration standard with each set o f extracted standards. Add 100 lL of the 100 ng/mL fo rtific a tio n solution to 40 mL of hypercarb filte re d water and process through the extraction procedure, identical to samples. Note; This standard is referred to as the "check standard" from this point forward. Page 13 The check standard should be included at least once with each set o f extracts analyzed. The check standard should give a response within 15% of the average response from the 250 ng/L extracted calibration standards in the set. (ttlPQHP The miracles o f science" Vh 4, Method 4 .1 . Flow Diagram The flow diagram o f the method is given below, followed by a detailed description o f each step. Method Flow Diagram Measure 40 mL of sample (fo rtify samples designated as calibration standards and fresh fortifcations) i I Cjb Solid-Phase Extraction i i Final Volume (5 mL) 4 i LC /M S /M S 4 .2 . Sample Processing No sample processing is needed fo r water samples. However, frozen samples must be allowed to completely thaw, un-aided, at room temperature. Samples stored refrigerated should also be allowed to equilibrate to room temperature. All samples must be thoroughly mixed and any visible solids removed by filte rin g the sample through a 25-mm diameter glass fib re ascrodisc 0.45 filter before being sampled fo r extraction. In itia l studies show tha t the method is not effected by the pH o f the sample. Sample pH is recorded prior to extraction. Control type I water, used fo r the preparation of samples and calibration standards as well as water used fo r equilibrating and washing the SPE cartridges, must be prepared by filte rin g type I water through a hypercarb guard cartridge using a HPLC pump. Before use, wash the guard cartridge with ~ 25 ml HPLC grade acetonitrile, then ~25 mL of HPLC grade methanol, followed by ~ 25 mL of type I water using the HPLC pump a t ~ 2 mL/min flow rate. Then s ta rt collecting the filte re d type I water eluate from the cartridge. I t is recommended th a t the cartridge be washed, as described above, a fte r filte rin g ~ 2 liters of type I water. A flow rate of 2 to 3 mL/min is recommended. 4 .3 . Sample Preparation a. Each batch of samples extracted (typically 20 or less) must include a t least one reagent control (method blank using filte re d type I water) and two reagent controls fo rtifie d at known concentrations to verify procedural recovery fo r the batch. Page 14 The miracles o f science 4 .4. b. Each sample must be extracted in replicate unless the sample is marked as a field blank. Field blanks do not need to be extracted in replicate. c. Each sample extracted (except fo r fie ld blanks) must also be separately fo r tifie d at a known concentration and carried through the procedure to v e rify recovery. M a trix spikes should range from 500 ng/L to 500,000 ng/L. I f the sample residue is found to be significantly higher than the level at which the sample was initially fo rtifie d , the sample should be fo r tifie d at a higher concentration (up to 500,000 ng/L) and re extracted. Extraction 1. Measure 40 mL of sample into 50 mL polypropylene centrifuge tubes ( fo r tify as needed, replace lid and mix well). Note: For tap w ater samples and any water samples suspected to contain chlorine, add 200 uL o f a 250 mg/mL sodium thiosulfate solution to the 40 mL sample and mix well to de-activate chlorine, before fortification. 2. Condition the C18 SPE cartridges (1 g, 6 mL) by passing 10 mL methanol followed by 5 mL filte re d type I water (~ 2 drop/sec). Do not let column run dry. N ote: For the following steps, maintain a ~1 d ro p/se c flow rate. Do not allow the column to run dry a t any time. 3. Load sample on conditioned Cis SPE cartridge. Discard eluate. 4. Wash with ~5 mL 40% methanol in water. Discard eluate. 5. Elute with ~5 mL 100% methanol. Collect 5 mL of eluate into graduated 15 mL polypropylene centrifuge tubes (final volume = 5 mL). 6. Analyze samples using electrospray LC/MS/MS. 4.5. N ote-. Samples are concentrated by a factor of eig ht during the extraction; Vol - 40 mL Fin a l Vol. = 5 mL. Q u an titatio n 4.5.1. LC/MS/MS System and Operating Conditions (Electrospray) I n itia l Mass Spec'. Interface: Computer: Software: Micromass Quattro Ultima (Micromass) Electrospray (Micromass) Harvard infusion pump (Harvard Instruments), fo r tuning COMPAQ Professional Workstation AP200 Windows NT, MassLynx 3.3 HPLC: Hewlett Packard (HP) Series 1100 HP Quat Pump HP Vacuum Degasser HP Autosampler HP Column Oven Note: A 4 x 10 mm hypercarb drop-in guard cartridge (Keystone, part # 844017-400) is attached on-line a fte r the purge valve and before the sample injector port to trap any residue contaminants th a t may be in the mobile phase and/or HPLC system. HPLC Column: Genesis C8 (Jones Chromatography), 2.1 mm x 50 mm, 4p Column Temperature: 35 C Page 15 qffPffiD) The miracles o f science' Injection Volume: Mobile Phase (A): Mobile Phase (B): 15 pL 2 mM Ammonium Acetate in Type I water Methanol Time 0.0 0.4 1.0 7.0 7.5 9.0 9.5 13.5 14.0 15.0 %A 60 60 10 10 0 0 60 60 60 60 % B Flow Rate (m L/m in) 40 0.3 40 0.3 90 0.3 90 0.3 100 0.3 100 0.4 40 0.4 40 0.4 40 0.3 40 0.3 I t may be necessary to adjust the HPLC gradient in order to optimize instrument performance. Columns with d iffe re n t dimensions (e.g. 2.1 x 30) and also columns from d iffe re n t manufacturers (Keystone Betasil Cw etc.) could be used, provided equivalent chromatography is obtained. Ions monitored: /Analyte Mode PFO A Negative T ransition Monitored 413 -- 369 A p p ro xim a te Retention Time ~5 min. The retention times may vary, on a day to day basis, depending on the batch of mobile phase etc. D rift in retention times (up to 2 %) is acceptable within an analytical run, as long as the d r if t continues through the entire analysis and the standards are included at the beginning and end of the analytical run. Note: An alternative LC/M5/MS system may be used once demonstrated to be equivalent. 4.5.2. Example MS Operating Parameters The following MS operating values are provided as an example only. Actual values will most likely vary from instrument to instrument. These values may also change over time even when analyzing samples on the same instrument. The values are changed to optimize fo r greatest sensitivity. For example, hexapole 1 voltage may be set to 2 V; however, as time goes on, hexapole 1 voltage may be increased or decreased. The exact values used each time a set is analyzed will be documented with each data set. The mass spectrometer is tuned fo r the analyte by infusing a ~ 0.2 pg/mL standard solution of PFOA (at 10 pL/min, using an infusion pump) via a "T" into a stream o f mobile phase containing 40% methanol and 60% 2mM ammonium acetate at 0.2 mL/min flow rate. The analyte is initially tuned fo r the parent ion and then tuned fo r the product ion. Once the Page 16 (flW D The miracles o f science" n instrument is tuned, the optimized parameters are saved as a tune file. This tune file is then used during routine analysis. Analyte PFOA Dwell is) 0.2 Collision Energy (eV) 10 Cone (Vi 10 Source Capillary Hexapole 1 Aperture 1 Hexapole 2 Source Block Temp. Desolvation Temp. Analyzer LM Res 1 HM Res 1 lEnergy 1 Entrance E x it LM Res 2 HM Res 2 lEnergy 2 M u ltip lie r Gas Flows Cone Gas Desolvation Set 3.0 kV 0V 0V 0V 100C 300C Set 14.0 V 14.0 V 1V -2 V 2V 14.0 V 14.0 V 2.0 V 700 V Set ~ 130 L /h r ~ 750 L/hr Pressures 4.5.3. Gas Cell Calibration back ~ 3.0e-3 mbar Procedures a. In je c t the same aliquot (between 10 to 25 pL) of each extracted calibration standard processed in matrix (ranging from the lowest level standard to the highest level prepared), into the LC/MS/MS (the final concentrations of these calibration standards will be equivalent to 8x the concentration o f the initial standard due to the concentration during SPE). b. Use linear standard curves fo r quantitation. The zero calibration standard is not included in the calculation of the calibration curve. Linear standard curves are generated fo r each analyte by linear regression using 1/x weighting o f peak area versus calibration standard concentration using MassLynx 3.3 (or equivalent) software system. Any calibration standard found to be a statistical outlier by using the Huge Error Test, Page 17 111)PONT* The miracles o f science" IS may be excluded from the calculation o f the calibration curve. However, the total number o f extracted calibration standards tha t may be excluded must not exceed 20% of the total number of extracted standards injected. C. The correlation coefficient (R) fo r calibration curves generated must be >0.992 (R2 >0.985). I f calibration results fall outside these limits, then appropriate steps must be taken to adjust instrument operation, and the standards or the relevant set o f samples should be reanalyzed. 4 .5 . Sample Analysis a. In je c t the same aliquot (between 10 to 25 pL) of each standard, sample, recovery, control, etc. into the LC/MS/MS system. b. Standards corresponding to at least five or more concentration levels (starting with the LOQ level or below) must be included in an analytical set (note th a t the calibration standards are prepared by spiking filte re d type I water and processing similar to actual samples). c. An entire set of extracted calibration standards should be injected at the beginning of a set followed by extracted calibration standards interspersed approximately every 510 samples (to account fo r a second set of extracted standards). As an alternative, an entire set of extracted calibration standards may be included at the beginning and at the end of a sample set. In either case, extracted calibration standards must be the f ir s t and last injection in a sample set. d. In je c t the check standard once with each set o f extracts. e. The concentration of each sam ple/fortification/control is determined from the standard curve, based on the peak area o f each analyte. The standard responses should bracket responses of the residue found in each sample set. I f necessary, dilute the samples to give a response within the standard curve range. f . Fortification recoveries falling within 70 to 130% are considered acceptable. g. The total holding time between sample collection and analysis must not exceed 14 days. Extracted samples must be stored refrigerated between 2C to 6C until analysis. h. Field samples in which either no peaks or peaks less than the MDL are detected at the corresponding analyte retention time will be reported as ND (not detected). Samples in which peaks are detected at the corresponding analyte retention time th a t are less than the LOQ and greater than or equal to the MOL will be reported as NQ (not quantifiable). The analysis performed during the method development included fo rtificatio n s at 100 and 500 ng/L of PFOA in well water, tap water, and stream water (collected locally, State College, PA) and bottled drinking water (from a local grocery store). 4 .6 . Acceptance C riteria The following criteria must be met to ensure the presence of PFOA: 1. Chromatogram must show a peak of a daughter ion at 369 amu from a parent of 413 amu. The 413 amu parent corresponds to the PFOA anion, while the daughter ion (369 amu) represents the loss of carbon dioxide. This transition would not discriminate between linear and branched forms o f PFOA; therefore, both, if present, would be included in the calculation. Page 18 (UPDID The miracles o f science` 2. Method blanks must not contain PFOA at levels greater than the LOQ. I f a blank contains PFOA at levels greater than 50 n g /l, then a new blank sample must be obtained and the entire set must be re-extracted. 3. Recoveries of control spikes and matrix spikes must be between 70-130% of the ir known values. I f a control spike falls outside the acceptable limits, the entire set of samples should be re-extracted. Any matrix spike outside 70-130% should be evaluated by the analyst to determine if re-extraction is warranted. 4. Any calibration standard found to be a statistical outlier by using the Huge Error Test, may be excluded from the calculation of the calibration curve. However, the total number of extracted calibration standards th a t could be excluded must not exceed 20% of the total number of extracted standards injected. 5. The correlation coefficient (R) fo r calibration curves generated must be >0.992 (R2 >0.985). I f calibration results fall outside these limits, then appropriate steps must be taken to adjust instrument operation, and the standards or the relevant set of samples should be reanalyzed. 6. The response from the check standard should be within 15% o f the average response from the 250 ng/L calibration standard concentrations. I f the standard response falls outside these limits, appropriate steps must be taken to adjust instrument operation and the set of samples should be reanalyzed. I f the standard response is still outside the 15% then standard preparations should be reviewed, and if warranted, new stock, fo rtifica tio n and extracted calibration solutions should be prepared. 7. Retention times between standards and samples must not d r if t more than 2 % within an analytical run. I f retention time d r if t exceeds this lim it within an analytical run then the set must be reanalyzed. 4 .7 . Mass Calibration C riteria In order to verify mass resolution, a calibration check is performed weekly on the mass spectrometer using a Nal/Rb/Cs tuning solution. The tuning solution is infused onto the mass spec using an infusion pump set to a flow rate o f 10 jL/min. An MS scan from 80 to 1380 amu is performed on both MSI and MS2. The calibration is considered acceptable if the following 11 peaks are found within 0.2 amu o f the ir known masses on both MSI and MS2: 84.9, 132.9, 172.8, 322.7, 472.6, 622.5, 772.4, 922.3, 1072.2, 1222.1 and 1372.0. I f any of the masses are outside the 0.2 amu, then the instrument will be recalibrated and another calibration check performed. 4 .8 . Performance C riteria The following two crite ria must be performed as a system suitability test, before the commencement of analysis when using an instrumentation set-up th a t has not been used fo r this method. First Criterion: Run a standard solution on LC/MS/MS corresponding to the estimated LOQ (50 ng/L) in matrix and obtain a signal to noise ratio of the 413 -> 369 transition o f at least 9:1, compared to a reagent blank. I f this criterion cannot be met, optimize and change instrument operating parameters (or increase the injection volume, if appropriate). Second Criterion: Page 19 5 The miracles o f science - Run a set of standards of five or more concentration levels, from at or below the LOQ, up to the highest concentration level to be included in the analysis. Generate a calibration curve fo r PFOA and obtain a linear regression with a coefficient of determination (R2) o f at least 0.985 fo r PFOA. Once this criterion is met, samples may be analyzed with standards interspersed. 4 .9 . M ethod Detection Lim it Determination Before samples can be analyzed on an instrument, a Method Detection Limit (MDL) determination should be performed. The procedure followed can be found in 40 CFR Part 136 Appendix 8. The MDL should then be performed on each instrument used fo r analysis on a quarterly basis. 4.1 0. Time Required fo r Analysis A set of 24 samples (6 standards and 18 samples) can be taken through the extraction procedure in approximately 2-3 hours by one person. The LC/MS/MS analysis (standards and 18 samples) will take approximately 7 hours. 5. Calculations a. Use Equation 1 to calculate the amount of PFOA found (in ng/L, based on peak area) using the standard curve (linear regression parameters) generated by the Mass Lynx software program. Equation 1: PFOA found (ng/L) = (Peak area - intercept) x DF slope DF = fa c to r by which the final volume was diluted, if necessary. b. For samples fo rtifie d with known amounts of PFOA prior to extraction, use Equation 2 to calculate the percent recovery. Equation 2: Recovery (%) ={(total analyte found - analyte found in control)/analyte added }x 100 [ total analyte found (ng/L) - analyte found in control (ng/L)] ^ ^^ analyte added (ng/L) Note: Subtract analyte found in control (ng/L) from analyte found (ng/L), if applicable. b. Use Equation 3 to calculate the amount of APFO found (in ng/L) using the molecular weight ratio of APFO/PFOA: Page 20 The miracles o f science" I Equation 3: APnrF-O f.ound (ng/L) = nPFmOA rFound. x -M---W---A---P--F--O---(-4--3--1--)- M W PFOA (414) Note: calculations could also be performed in ppb or ng/mL if concentrations were assigned in ng/mL instead of ng/L when performing the analysis. For reporting purposes, field samples in which either no peaks or peaks less than the MDL are detected at the corresponding analyte retention time will be reported as ND (not detected). Samples in which peaks are detected at the corresponding analyte retention time tha t are less than the LOQ and greater than or equal to the MDL will be reported as NQ (not quantifiable). 6. S afety The analyst should read the m aterial sa fety data sheets fo r all standards and reagents before performing this method. Use universal precautions when handling standards and reagents, including working in fume hoods and wearing laboratory coats, safety glasses, and gloves. 7. REFERENCES Ellis, D.A.; Martin, J.W.; Muir, D.C.G.; Mabury, S.A. Development of an 19F NMR Method fo r the Analysis of Fluorinated Acids in Environmental Water Samples. Anal. Chem. 2000, 72, 726-731. Moody, C.A.; Kwan, W.C; Martin, J.W.; Muir D.C.G.: Mabury, S.A. Determination of Perfluorinated Surfactants in Surface Water Samples by Two Independent Analytical Techniques: Liquid Chromatography/Tandem Mass Spectrometry and 19F NMR. Anal. Chem. 2001, 73, 2200-2206. Page 21 The miractes o f science'