EPA 1633 PFAS Workflows Across Environmental and Biological Matrices

Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples. Introduction Per- and polyfluoroalkyl substances (PFAS) are a group of synthetic fluorinated organic compounds that have been produced and widely used in industrial applications and consumer products since 1940s, such as additives in fluoropolymers production, surfactants in numerous consumer products, fire-fighting foams, stain resistant coatings for furniture and carpeting, nonstick cookware coatings, lubricants, breathable waterproof fabrics, paints, shampoos, cosmetics, food packaging materials and other materials. Due to their unique physical and chemical properties, long-term persistence in the environment, and associated risks for human health, PFAS have been classified as persistent organic pollutants and have become the current hot topic worldwide. PFAS have been detected in almost everywhere on our earth, in wildlife, humans (such as human serum and blood), food, and the environment such as drinking water, air, surface water (pond, river, lake, sea, and ocean) and even in remote areas such as the Tibetan mountains and the Arctic regions 1-9. The development of an efficient and robust strategy for the identification and quantification of PFAS is essential for PFAS monitoring and risk assessment. The most widely used analytical method for PFAS monitoring is LC/MS/MS due to its high sensitivity, selectivity, and robustness 10-12. Several methods have been developed for the analysis of PFAS. The most comprehensive method covering 40 PFAS has Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples APPLICATION NOTE Liquid Chromatography/ Mass Spectrometry AUTHOR Jingcun Wu PerkinElmer Woodbridge, Ontario, CanadaAnalysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples been developed recently by the United States Environmental Protection Agency (U.S. EPA) in collaboration with the Department of Defense 12. Currently, the EPA 1633 method is the only PFAS method that has been validated in multiple laboratories for aqueous matrices that include surface water, groundwater, wastewater, and landfill leachate, as well as for soil, sediment, biosolids, and fish and shellfish tissue. This application note is the first in a series of application notes that will address method development, sample preparation for different matrices, and method performance of applying EPA 1633 with a comprehensive workflow of PerkinElmer technologies. This application note will focus on LC/MS/ MS method development using PerkinElmer’s QSight 500® LC/MS/MS system and evaluate the key performance indicators (KPIs) and considerations to meet the EPA method requirements. The method performance was verified by analyzing quality control samples and a proficiency testing (PT) sample. An automated solid phase extraction (SPE) system was used during sample preparation. Future application notes will cover applications of this method in real-world surface water and wastewater samples, animal tissue samples, and solid and biosolid samples. Experimental Safety First Several PFAS, including perfluorooctanoic acid (PFOA), have been described as likely to be carcinogenic to humans. Exposure to these compounds should be reduced to the lowest possible level. Personal protection equipment and safety training must be provided, and safety procedures must be implemented before and during work. Refer to EPA 1633 sections 5.0 to 5.4 for details.12 Hardware/Software The chromatographic separation was conducted using a PerkinElmer QSight LX 50 ultra-high-performance liquid chromatography (UHPLC) system. PFAS detection was achieved using a PerkinElmer QSight 500 or 420 triple quadrupole mass spectrometer with ESI ionization mode. The LX50 Autosampler was modified by replacing all polytetrafluoroethylene (PTFE) based tubing with polyether ether ketone (PEEK) tubing to reduce any contamination from PFAS compounds introduced by the PTFE tubing. All instrument control, data acquisition and data processing were performed using the Simplicity™ 3Q Software. www.perkinelmer.com Materials and Methods Chemicals and materials LC-MS grade solvents methanol (MeOH), acetonitrile (ACN), and water, and other chemicals such as formic acid, acetic acid, ammonium hydroxide (30% in water), ammonium acetate, bile salts including taurodeoxycholic acid (TDCA), taurochenodeoxycholic acid (TCDCA), and tauroursodeoxycholic acid (TUDCA) were obtained from MilliporeSigma (Oakville, ON, Canada). The mixed PFAS standards (EPA-1633STK) and isotope labelled standards including extracted internal standard (EIS) mix and nonextracted internal standard (NIS) mix were purchased from Wellington Laboratories (Guelph, Ontario). An EPA-1633 proficiency testing sample (WP-PT-EPA 1633, Part # 38721, Lot# 041422) was obtained from Absolute Standards Inc (Hamden, CT, USA). Disposable syringe filters, polypropylene (PP) centrifuge tubes, polypropylene autosampler vials, delay column, guard column and analytical column were obtained from PerkinElmer (Shelton, CT, USA). PFAS-free autosampler vial caps were obtained from Integrated Liner Technologies (Rensselaer, NY, USA). Polypropylene bottles, autosampler vials and caps are required to prevent adsorption of PFAS compounds on glassware and to eliminate PFAS materials commonly used in HPLC vial septa. The stacked weak anion exchange/graphitized carbon black (WAX/GCB) SPE cartridges (6 mL tubes containing 200 mg/50 mg sorbent, respectively) were obtained from PerkinElmer. Certified and cleaned sample bottle (HDPE, with polypropylene caps, 1000 mL, 500 mL, 250 mL) and inlet filters were obtained from Promochrom Technologies (Richmond, BC, Canada). Diluent preparation (4% water, 1% ammonium hydroxide and 0.625% acetic acid in methanol) – see Sections 7.1.9 and 7.3.4 of EPA method 1633. Ammonium hydroxide (3.3 mL, 30%), reagent water (1.7 mL) and acetic acid (0.625 mL) were added to a 100 mL flask containing methanol (92 mL), mixed thoroughly, and stored at room temperature. The fresh solution should be prepared monthly. This solution can be used to prepare for the instrument blank (Section 7.3.6 of the EPA method 1633), to prepare for calibration standards, and to dilute the extracts of sample if the sample exceed the calibration range (see Section 15.3 of the EPA method 1633). Note: it is important to use this diluent instead of methanol to prepare calibration standards to match the pH and ion strengths in the sample matrices to reduce retention time shift or variations between standards and samples. 23 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Table 1. Volume (µL) of spiking solutions and diluent needed for each calibration standard. Standard Preparation The names, abbreviations, and CAS registry numbers for the 40 target PFAS, 24 extracted internal standards (EIS), and 7 non-extracted internal standards (NIS) are listed in Tables 1 and 2 of EPA method 1633.12 Four spiking solutions were first prepared before preparing the calibration standards. PFAS Spiking 1 was prepared by diluting 100 µL of the mixed PFAS stock standard solution (EPA-1633STK) to 1 mL with 900 µL of diluent. PFAS Spiking 2 was prepared by diluting 50 µL of PFAS Spiking 1 to 1 mL with 950 µL of diluent. EIS Spiking was prepared by diluting 50 µL of mixed EIS stock to 1 mL with 950 µL of diluent. NIS Spiking was prepared by diluting 50 µL of mixed NIS stock to 1 mL with 950 µL of diluent. Polypropylene autosampler vials were cleaned with methanol and air dry before use. To each of the nine autosampler vials, 100 µL each of EIS Spiking and NIS Spiking were added, respectively. Then, appropriate volume (µL) of spiking solutions and diluent were added to each vial according to Table 1. Briefly, Calibration Standard 1 (CS1) was prepared by diluting 100 µL of PFAS Spiking 2 to 1 mL with diluent. Calibration Standard 2 (CS2) was prepared by diluting 200 µL of PFAS Spiking 2 to 1 mL with diluent. Calibration Standard 3 (CS3) was prepared by diluting 20 µL of PFAS Spiking 1 to 1 mL with diluent. Calibration Standard 4 (CS4) was prepared by diluting 50 µL of PFAS Spiking 1 to 1 mL with diluent. Calibration Standard 5 (CS5) was prepared by diluting 100 µL of PFAS Spiking 1 to 1 mL with diluent. Calibration Standard 6 (CS6) was prepared by diluting 200 µL of PFAS Spiking 1 to 1 mL with diluent. Calibration Standard 7 (CS7) was prepared by diluting 50 µL of PFAS Stock to 1 mL with diluent. Calibration Standard 8 (CS8) was prepared by diluting 100 µL of PFAS Stock to 1 mL with diluent. Calibration Standard 9 (CS9) was prepared by diluting 200 µL of PFAS Stock to 1 mL with diluent. A Standard 0 solution containing only IS was also prepared to check the purity of IS. An Instrument Blank was prepared using the diluent. The concentrations of the calibration standards are listed in Table 2, and the concentrations of EIS and NIS are the same as those listed in Table 4 of EPA method 1633 12. Solutions (µL) * CS1 CS2 CS3 CS4 CS5 CS6 CS7 CS8 CS9 Spiking 2 100 200 Diluent 700 600 Spiking 1 20 50 100 200 Diluent 780 750 700 600 PFAS Stock 50 100 200 Diluent 750 700 600 *Add 100 µL of each IS spiking solution to each vial (total 200 µL). Table 2. Example calibration standards (ng/mL). PFAS Analyte CS1 CS2 CS3 CS4 CS5 CS6 CS7 CS8 CS9 PFBA 0.5 1 2 5 10 20 50 100 200 PFPeA 0.25 0.5 1 2.5 5 10 25 50 100 PFHxA 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 PFHpA 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 PFOA 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 PFNA 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 PFDA 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 PFUnA 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 PFDoA 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 PFTrDA 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 PFTeDA 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 PFBS 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 PFPeS 0.125 0.25 0.5 1.25 2.5 5 12.5 25 504 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples PFAS Analyte CS1 CS2 CS3 CS4 CS5 CS6 CS7 CS8 CS9 PFHxS 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 PFHpS 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 PFOS 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 PFNS 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 PFDS 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 PFDoS 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 4:2FTS 0.5 1 2 5 10 20 50 100 200 6:2FTS 0.5 1 2 5 10 20 50 100 200 8:2FTS 0.5 1 2 5 10 20 50 100 200 PFOSA 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 NMeFOSA 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 NEtFOSA 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 NMeFOSAA 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 NEtFOSAA 0.125 0.25 0.5 1.25 2.5 5 12.5 25 50 NMeFOSE 1.25 2.5 5 12.5 25 50 125 250 500 NEtFOSE 1.25 2.5 5 12.5 25 50 125 250 500 HFPO-DA 0.5 1 2 5 10 20 50 100 200 ADONA 0.5 1 2 5 10 20 50 100 200 PFMPA 0.25 0.5 1 2.5 5 10 25 50 100 PFMBA 0.25 0.5 1 2.5 5 10 25 50 100 NFDHA 0.25 0.5 1 2.5 5 10 25 50 100 9Cl-PF3ONS 0.5 1 2 5 10 20 50 100 200 11Cl-PF3OUdS 0.5 1 2 5 10 20 50 100 200 PFEESA 0.25 0.5 1 2.5 5 10 25 50 100 3:3FTCA 0.625 1.25 2.5 6.25 12.5 25 62.5 125 250 5:3FTCA 3.13 6.25 12.5 31.3 62.5 125 313 625 1250 7:3FTCA 3.13 6.25 12.5 31.3 62.5 125 313 625 1250 Note: A minimum of six levels of calibration standards are required by US EPA method 1633. Thus, it is not necessary to keep the nine levels of calibration standards. The calibration standards listed in Table 1 can be adjusted based on the instrument sensitivity (CS1 may be deleted if the instrument has lower sensitivity) and the PFAS contents in the samples (CS9 may be deleted if PFAS contents are low). Sample Preparation Refer to Section 11 of EPA method 1633 for detailed sample preparation procedures.12 Briefly, the nominal sample size for aqueous samples, and their associated quality control (QC) samples is 500 mL; a lower volume of samples such as 250 mL may be used if high levels of PFAS are present in the samples, or the instrument has much higher sensitivity with a relatively large injection volume (such as 6 µL instead of 2 µL). The volume of the aqueous sample is determined by weighing the full sample bottle and then the empty sample bottle. Weigh each sample bottle (with the lid) to 0.1 g. A method blank and two ongoing precision and recovery (OPRs) QC samples were prepared using PFAS-free reagent water in HDPE bottles. Spike one OPR sample with the mixed PFAS stock solution (EPA1633STK) at 2x the LOQ (LLOPR). A 10 µL of stock solution was spiked in this study. This sample was used to verify the LOQ. Spike the other OPR sample at the concentration of the mid-level calibration point (e.g., CS5 in Table 1) and a 50 µL of PFAS stock solution was used in this study. This sample was used as the traditional OPR or regular QC sample for routine analysis. Spike 25 µL of EIS stock solution directly into each of the samples in their original bottles and mix the solutions by swirling the sample containers. Using a PFAS-free pipette, transfer a few drops of the sample to a pH paper and check that the pH is 6.5 ± 0.5. If necessary, adjust pH with 50% formic acid or ammonium hydroxide (or with 5% formic acid and 3% aqueous ammonium hydroxide). The samples are now ready for solid-phase extraction (SPE) and cleanup. Table 2. Example calibration standards (ng/mL). Continued...Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Sample extraction and clean-up by automated SPE All method blanks, proficiency testing (PT) sample, and QC samples were extracted and concentrated by the SPE procedures described in detail in Section 12.0 of EPA Method 1633. Slight modifications were made as allowed by the method. Using stacked WAX/GCB SPE cartridges instead of WAX SPE cartridges plus the loose carbon clean-up reduced sample extraction and clean-up time and minimized manual operation errors. The SPE cartridges were pre-conditioned with 15 mL of 1% methanolic ammonium hydroxide and 5 mL of 0.3M formic acid. Then, water samples were added to the cartridges by a syringe pump at 5 mL/min. The reservoir walls were rinsed with 5 mL reagent water (twice) followed by 5 mL of 1:1 0.1M formic acid/methanol and passed through the cartridge using a vacuum. The cartridge was dried by pulling air through for 15 seconds, and the rinse solution was discarded. The inside of the sample bottles was rinsed with 5 mL of 1% methanolic ammonium hydroxide, shaken for 10 seconds, then the rinse was transferred to the SPE reservoir, washing the walls of the reservoir, while pushing the elution solvent through the cartridge and into the collection tubes. Inlet filters were used between sample bottles and SPE cartridges to prevent particles from blocking the SPE cartridges. All these SPE procedures were carried out automatically using a PromoChrom SPE-03 system (Promochrom Technologies, Richmond, BC, Canada) to reduce labor-intensive human work and minimize human errors. Figure 1 shows the automated SPE and QSight LC/MS/MS systems. Please refer to the YouTube videos for detailed operation procedure 13-14. Figure 1. Automated SPE and QSight LC/MS/MS systems. LC conditions and MS parameters The LC method and MS source parameters are shown in Table 3. Two C18 columns were used in this study: one was used as a delay column to separate possible interferent PFASs coming from the LC system; another was used as an analytical column to separate PFAS compounds and any interfering components. A guard column with the same phase was also used to protect the analytical column. The applied LC gradient program is shown in Table 4. MS Source parameters, including gas flows, source temperature and probe position settings, were optimized for maximum sensitivity. Compounddependent parameters, such as collision energies (CE), www.perkinelmer.com entrance voltages (EV), and lens voltages (CCL2), were optimized for the target PFAS as shown in Table 5. During method development, the retention times for all PFAS peaks were determined, and then the potential interfering components from LC system and mobile phases were identified and separated from analyte peaks using a delay column. Finally, the MS acquisition method was generated using Simplicity™ software in the time-managed-MRM module with the retention times and corresponding retention time windows for all PFASs. 56 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Table 3. LC method and MS source conditions. LC Conditions Analytical Column Brownlee, SPP C18, 100 x 4.6mm, 2.7 µm (N9308416), Guard (N9308515) Delay Column Brownlee, SPP C18, 50 x 3mm, 2.7 µm (N9308408) Mobile Phase A 2 mM ammonium acetate in water Mobile Phase B LC/MS grade acetonitrile Mobile Phase Gradient See Table 4 Flow Rate 0.7 mL/min Column Oven Temperature 40 ºC Auto Sampler Temperature 10 ºC Injection Volume 6 µL Needle wash 1 (weak wash) 10% methanol in water Needle wash 2 (strong wash) 100% methanol MS Source Conditions ESI Voltage (Negative)-3800 V Drying Gas 180 Nebulizer Gas 400 Source Temperature 350 ºC HSID Temperature 280 ºC Detection mode Time managed MRM Table 4. LC gradient program. Time (min) Mobile Phase A (%) Mobile Phase B (%) 0.00 85 15 0.50 85 15 3.00 55 45 6.00 40 60 8.50 2 98 9.00 2 98 9.10 85 15 12.50 85 15 PFAS Analyte Precurser Ion (m/z) Product Ion (m/z) RT (min) CE EV CCL2 Quantifier / Qualifier Type 13C4-PFBA 217.0 172.0 3.5 14-4 40 Quantifier EIS 13C3-PFBA 215.8 171.8 3.5 14-10 56 Quantifier NIS PFBA-1 213.0 169.0 3.5 13-9 36 Quantifier Analyte PFBA-2 213.1 69.1 3.5 90-14 124 Qualifier Analyte PFMPA 229.0 84.9 3.85 38-7 48 Quantifier Analyte 3:3FTCA-1 240.9 116.9 4.04 55-10 84 Quantifier Analyte 3:3FTCA-2 240.9 176.9 4.04 12-10 64 Qualifier Analyte 13C5-PFPeA 268.0 223.0 4.26 11-10 104 Quantifier EIS PFPeA-1 263.0 219.0 4.26 15-8 80 Quantifier Analyte Table 5. Optimized MRM parameters.7 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Table 5. Optimized MRM parameters. Continued... PFAS Analyte Precurser Ion (m/z) Product Ion (m/z) RT (min) CE EV CCL2 Quantifier / Qualifier Type PFPeA-2 263.0 69.1 4.26 70-12 88 Qualifier Analyte PFMBA 279.0 84.9 4.43 46-3 60 Quantifier Analyte 13C2-4:2FTS-1 329.0 81.0 4.56 38-43 96 Quantifier EIS 13C2-4:2FTS-2 329.0 309.0 4.56 23-14 120 Qualifier EIS 4:2FTS-1 327.0 81.0 4.56 38-43 96 Quantifier Analyte 4:2FTS-2 327.0 307.0 4.56 23-1 110 Qualifier Analyte NFDHA-1 201.0 85.0 4.7 31-11 48 Quantifier Analyte NFDHA-2 295.1 85.0 4.7 41-38 75 Qualifier Analyte NFDHA-3 295.1 201.0 4.7 18-40 64 Qualifier Analyte 13C5-PFHxA-1 318.0 273.0 4.75 12-4 96 Quantifier EIS 13C5-PFHxA-2 318.0 120.1 4.75 32-10 112 Qualifier EIS 13C2-PFHxA-1 315.1 269.9 4.75 14-10 72 Quantifier NIS 13C2-PFHxA-2 315.1 120.0 4.75 32-10 76 Qualifier NIS PFHxA-1 313.0 269.0 4.75 17-10 55 Quantifier Analyte PFHxA-2 313.0 119.0 4.75 31-10 50 Qualifier Analyte 13C3-PFBS-1 302.0 80.0 4.88 67-28 80 Quantifier EIS 13C3-PFBS-2 302.0 99.0 4.88 41-28 80 Qualifier EIS PFBS-1 299.1 80.1 4.88 63-36 100 Quantifier Analyte PFBS-2 299.1 98.9 4.88 42-21 104 Qualifier Analyte HFPO-DA-1 285.0 168.9 4.91 14-5 76 Quantifier Analyte HFPO-DA-2 285.0 184.9 4.91 28-5 76 Qualifier Analyte 13C3-HFPO-DA-1 287.0 169.0 4.91 14-5 76 Quantifier EIS 13C3-HFPO-DA-2 287.0 185.0 4.91 28-5 76 Qualifier EIS PFEESA-1 314.9 134.8 5.12 33-8 84 Quantifier Analyte PFEESA-2 314.9 69.0 5.12 79-8 76 Qualifier Analyte PFEESA-3 314.9 82.9 5.12 26-8 76 Qualifier Analyte 13C4-PFHpA 367.0 322.0 5.16 17-6 75 Quantifier EIS PFHpA-1 363.0 319.0 5.16 16-10 56 Quantifier Analyte PFHpA-2 363.0 169.0 5.16 23-10 92 Qualifier Analyte 5:3FTCA-1 341.0 217.1 5.13 36-10 88 Quantifier Analyte 5:3FTCA-2 341.0 236.9 5.13 23-10 96 Qualifier Analyte ADONA-1 377.0 85.0 5.33 54-10 84 Quantifier Analyte ADONA-2 377.0 251.0 5.33 19-10 84 Qualifier Analyte PFPeS-1 349.0 80.0 5.4 73-6 100 Quantifier Analyte PFPeS-2 349.0 99.0 5.4 45-6 90 Qualifier Analyte 13C2-6:2FTS-1 429.0 81.0 5.38 43-16 124 Quantifier EIS 13C2-6:2FTS-2 429.0 409.0 5.38 28-11 152 Qualifier EIS 6:2FTS-1 427.0 81.0 5.38 65-8 80 Quantifier Analyte 6:2FTS-2 427.0 407.0 5.38 28-28 172 Qualifier Analyte 13C8-PFOA 421.0 376.0 5.5 16-4 84 Quantifier EIS 13C4-PFOA-1 417.1 372.1 5.5 15-10 96 Quantifier NIS8 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Table 5. Optimized MRM parameters. Continued... PFAS Analyte Precurser Ion (m/z) Product Ion (m/z) RT (min) CE EV CCL2 Quantifier / Qualifier Type 13C4-PFOA-2 417.1 171.9 5.5 26-10 96 Qualifier NIS PFOA-1 413.2 369.1 5.5 14-14 124 Quantifier Analyte PFOA-2 413.2 168.9 5.5 25-14 124 Qualifier Analyte 13C3-PFHxS-1 402.0 80.0 5.82 87-8 100 Quantifier EIS 13C3-PFHxS-2 402.0 99.0 5.82 45-10 128 Qualifier EIS 18O2-PFHxS-1 403.0 83.9 5.82 80-10 112 Quantifier NIS 18O2-PFHxS-2 403.0 103.0 5.82 47-10 112 Qualifier NIS PFHxS-1 399.0 80.0 5.82 85-45 120 Quantifier Analyte PFHxS-2 399.0 99.0 5.82 49-45 87 Qualifier Analyte 13C9-PFNA-J 472.0 427.0 6.1 15-12 152 Quantifier EIS 13C5-PFNA-1 468.1 423.1 6.1 15-10 108 Quantifier NIS PFNA-1 463.1 419.1 6.1 15-12 168 Quantifier Analyte PFNA-2 463.1 219.1 6.1 24-12 164 Qualifier Analyte 7:3FTCA-1 441.0 317.1 6.37 33-10 116 Quantifier Analyte 7:3FTCA-2 441.0 336.9 6.37 23-10 112 Qualifier Analyte 13C2-8:2FTS-1 529.0 81.0 6.56 76-40 115 Quantifier EIS 13C2-8:2FTS-2 529.0 509.0 6.56 35-40 115 Qualifier EIS 8:2FTS-1 527.0 81.0 6.56 70-3 100 Quantifier Analyte 8:2FTS-2 527.0 507.0 6.56 30-62 220 Qualifier Analyte PFHpS-1 449.0 80.0 6.64 86-20 144 Quantifier Analyte PFHpS-2 449.0 99.0 6.64 48-20 136 Qualifier Analyte D3-NMeFOSAA 573.0 419.0 6.72 27-25 105 Quantifier EIS NMeFOSAA-1 570.0 419.0 6.72 28-20 100 Quantifier Analyte NMeFOSAA-2 570.2 169.0 6.72 40-10 168 Qualifier Analyte NMeFOSAA-3 570.0 483.0 6.72 19-20 100 Qualifier Analyte 13C6-PFDA 519.0 474.0 6.92 17 0 88 Quantifier EIS 13C2-PFDA 515.1 470.1 6.92 15-10 112 Quantifier NIS PFDA-1 513.1 469.1 6.92 16-14 170 Quantifier Analyte PFDA-2 513.0 219.0 6.92 28-10 96 Qualifier Analyte D5-NEtFOSAA-1 589.0 419.0 7.05 28-20 112 Quantifier EIS D5-NEtFOSAA-2 589.0 531.0 7.05 28-20 105 Qualifier EIS NEtFOSAA-1 584.0 419.0 7.05 30-20 100 Quantifier Analyte NEtFOSAA-2 584.1 169.0 7.05 43-20 174 Qualifier Analyte NEtFOSAA-3 584.1 526.2 7.05 25-20 176 Qualifier Analyte 13C4-PFOS-1 503.0 80.0 7.16 107-45 125 Quantifier NIS 13C4-PFOS-2 503.0 99.0 7.16 56-10 136 Qualifier NIS 13C8-PFOS-1 507.0 80.0 7.16 107-22 140 Quantifier EIS 13C8-PFOS-2 507.0 99.0 7.16 50-22 140 Qualifier EIS PFOS-1 499.1 80.0 7.16 107-10 179 Quantifier Analyte PFOS-2 499.1 99.0 7.16 56-10 161 Qualifier Analyte 13C7-PFUnA 570.0 525.0 7.64 15-14 184 Quantifier EIS9 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples PFAS Analyte Precurser Ion (m/z) Product Ion (m/z) RT (min) CE EV CCL2 Quantifier / Qualifier Type PFUnA-1 563.0 519.0 7.64 19-10 96 Quantifier Analyte PFUnA-2 563.0 269.1 7.64 25-14 184 Qualifier Analyte 9Cl-PF3ONS-1 531.0 351.0 7.88 34-40 188 Quantifier Analyte 9Cl-PF3ONS-2 533.0 353.0 7.88 34-40 152 Qualifier Analyte 9Cl-PF3ONS-3 531.0 83.0 7.88 35-40 132 Qualifier Analyte PFNS-1 549.0 80.0 8.06 96-34 184 Quantifier Analyte PFNS-2 549.0 99.0 8.06 62-34 184 Qualifier Analyte 13C2-PFDoA 615.0 570.0 8.27 17-14 104 Quantifier EIS PFDoA-1 613.0 569.0 8.27 17-10 104 Quantifier Analyte PFDoA-2 613.0 169.1 8.27 35-14 180 Qualifier Analyte PFDoA-3 613.0 319.0 8.27 30-14 150 Qualifier Analyte PFDS-1 599.1 80.1 8.55 138-14 230 Quantifier Analyte PFDS-2 599.1 99.0 8.55 57-14 240 Qualifier Analyte PFTrDA-1 663.0 619.0 8.7 18-6 104 Quantifier Analyte PFTrDA-2 663.0 169.1 8.7 38-14 220 Qualifier Analyte 11Cl-PF3OUdS-1 631.0 451.0 8.82 34-40 180 Quantifier Analyte 11Cl-PF3OUdS-2 633.0 453.0 8.82 34-40 180 Qualifier Analyte PFTeDA-1 713.0 669.0 9.01 19-4 120 Quantifier Analyte PFTeDA-2 713.0 168.9 9.01 40-14 240 Qualifier Analyte PFDoS-1 698.9 79.9 9.15 130-86 184 Quantifier Analyte PFDoS-2 698.9 99.0 9.15 69-86 184 Qualifier Analyte 13C8-PFOSA 506.0 77.8 8.98 48-48 144 Quantifier EIS PFOSA-1 498.0 78.0 8.98 48-3 110 Quantifier Analyte PFOSA-2 498.0 478.0 8.98 30-27 212 Qualifier Analyte D7-NMeFOSE 623.0 59.0 9.86 116-3 124 Quantifier EIS NMeFOSE-1 616.0 59.0 9.86 116-3 124 Quantifier Analyte NMeFOSE-2 616.0 59.0 9.86 60-3 124 Qualifier Analyte D3-NMeFOSA-1 515.1 168.9 9.93 34-10 152 Quantifier EIS D3-NMeFOSA-2 515.1 219.0 9.93 28-3 160 Qualifier EIS NMeFOSA-1 512.1 168.9 9.93 34-10 152 Quantifier Analyte NMeFOSA-2 512.1 219.0 9.93 32-5 124 Qualifier Analyte NEtFOSE 630.3 59.0 10.1 116-11 120 Quantifier Analyte D9-NEtFOSE 639.0 59.0 10.1 116-11 126 Quantifier EIS D5-NEtFOSA-1 531.1 168.9 10.15 36-14 152 Quantifier EIS D5-NEtFOSA-2 531.1 219.0 10.15 30-14 132 Qualifier EIS NEtFOSA-1 526.1 168.9 10.15 36-14 152 Quantifier Analyte NEtFOSA-2 526.1 219.0 10.15 30-10 120 Qualifier Analyte Table 5. Optimized MRM parameters. Continued...Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Results and Discussion PFAS contamination and carryover effect One of the main challenges in the analysis of PFAS by LC/ MS/MS is the potential contamination from the reagents, mobile phases, and contact materials (SPE system, SPE cartridges, sample collection bottles and tubes, glassware, autosampler vials and caps, filters, pipette tips, syringes, tubing, fittings, LC pumps and pump seals, PTFE parts in LC system) and the lab environment. To minimize contamination from mobile phases and the LC pump system, a delay column was inserted between the mobile phase mixing valve of the pump and the autosampler injection valve to trap and isolate PFAS components arising from mobile phases and the LC pump system. As shown in Figure 2, the analyte peaks can be well separated from the system contamination peaks by the delay column. The standard PTFE tubing in the autosampler was replaced with PEEK tubing to remediate potential contamination. All the materials used in this study were tested prior to running samples to check for PFAS contamination by injecting blank samples. Through these experiments, it was confirmed that all the supplies used were free of PFAS contamination. At the beginning of the analytical sequence and after the analysis of high concentration samples and standards (e.g., the highest calibration standard, CS9 in this study), an instrument blank was analyzed to ensure no instrument contamination (or carryover effect) has occurred. The instrument blank must not contain any target analyte that would yield a response equivalent to the mass of the analyte that would be present in a whole-volume sample at or above the method detection limit (MDL). The results from this study showed that the carryover effect was less than the MDL using the established needle-washing procedures. Figure 2. Separation of PFAS analyte peak from interfering peaks from LC system using a delay column. www.perkinelmer.com 10Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples LC and MS/MS Method Development Separation of PFAS analyte from sample matrix interfering components One of the major analytical challenges during LC/MS/MS method development and validation is sample matrix effect, which can affect data quality and the method’s selectivity, sensitivity, and accuracy 15. To get accurate results, interfering components from sample matrices must be separated from targeted analyte peaks. Interferences co-extracted from samples will vary considerably from source to source, depending on the diversity of the sample matrices. Interfering compounds may be present at concentrations several orders of magnitude higher than the target PFAS. Because low levels of PFAS are measured by this method, elimination or minimization of interferences is essential. The extraction and cleanup steps described in the previous section can reduce these interferences and thereby permit reliable determination of the PFAS at low levels. However, high concentrations of bile salts, taurodeoxycholic acid (TDCA), taurochenodeoxycholic acid (TCDCA), and tauroursodeoxycholic acid (TUDCA), may be present in various matrices, including fish and wastewaters, and can interfere with PFAS in the chromatography. For this reason, the evaluation of interfering components from these three bile salts is required by the EPA 1633 no matter what sample matrices are to be analyzed (this was mentioned at least five times in the EPA method: see section 4.4, section 7.5, section 10.2.2.5, section 10.3.5, and section 14.2). When establishing LC conditions, it is important to ensure adequate chromatographic separation to avoid potential interference from bile salts during analyses of samples, adjust the conditions if needed to ensure that TDCA, TCDCA and TUDCA do not coelute with any of the target analytes, EIS, or NIS compounds. As the EPA method described clearly: “The laboratory must demonstrate that the analytical conditions provide a separation of at least 1 minute between the bile salts and the retention time window of PFOS. If this requirement is not met, the chromatographic conditions must be adjusted to meet the requirement and the initial calibration must be repeated before any field sample are analyzed.” In this study, to meet this requirement, a PFAS standard containing high concentrations (each 1 µg/mL) of TDCA, TCDCA, and TUDCA was prepared and analyzed using different LC columns and mobile phase compositions to find the best LC conditions. The effects of LC columns on separation. This study compared the effects of three LC column phases on separating PFOS from the three bile salts using methanol / ammonium acetate mobile phase compositions. As illustrated in Figures 3-5, although PFOS and its branched isomer peaks www.perkinelmer.com were separated from the three bile salts using a C18 column, the separation was inadequate (less than 1 min). When a phenyl-hexyl column was used, the PFOS peak overlapped entirely with the TUDCA peak. Therefore, neither column could meet the separation requirement. However, when using a biphenyl column, all the bile salts interfering peaks were eluted after the PFOS and its isomers peaks. This significant change in separation order can benefit PFAS analysis since interfering peaks eluted after analyte peaks would have fewer interfering effects on the analyte. Thus, the biphenyl column provides an alternative selectivity for PFAS analysis. The effects of LC mobile phases on separation. Since methanol /ammonium acetate mobile phase compositions can provide better signal intensity for most PFAS analytes, they have been widely used as mobile phases for PFAS analysis. However, as demonstrated in Figure 3, although PFOS and its branched isomer peaks can be separated from the three bile salts using a C18 column with methanol/ ammonium acetate mobile phases, the separation is inadequate (less than 1 min). When the organic mobile phase component was changed from methanol to acetonitrile, much better separation was achieved as shown in Figure 6. This is why acetonitrile has been selected as organic mobile phase component by the EPA method 1633 although methanol mobile phases can provide higher PFAS signal and thus higher sensitivity. If methanol must be used, biphenyl column may be tested for its applications. The effects of LC column size on separation. Although the small particle size column such as the Waters Acquity UPLC® BEH or equivalent (C18 column, 1.7 µm, 50 x 2.1 mm) recommended by EPA method 1633 has the advantages of best separation efficiency and less solvent consumption, it requires an experienced analytical chemist to run the analysis, and samples must be filtered before analysis. In addition, the small particle size and narrow internal diameter (ID) column is easy to be blocked by complex sample matrices which lead to very high pressure in LC pump and thus affect the robustness of the method. Using the superficially porous particles (SPP) columns with sub-3-µm particles (2.5 ~2.8 µm) can achieved similar results with much lower pressure when using relatively larger id column.16 Larger ID SPP column have the advantages of higher column capacity which not only makes it possible to inject relatively larger sample volume on column to boost analyte signal, but also increase the separation of analytes. For example, using a Brownlee C18 column with 4.6 mm ID can achieve much better separation of PFOS from the bile salts interferences (2 min difference in retention time between PFAS and the interfering peaks) compared to the separation obtained using a Brownlee C18 column with 3.0 mm ID (1 min difference in retention time between PFAS and the interfering peaks) as demonstrated in Figures 6-7. 11Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Figure 3. Separation of PFOS from TUDCA, TCDCA and TCDA using a C18 column. Figure 4. Separation of PFOS from TUDCA, TCDCA and TCDA using a phenyl-hexyl column. Figure 5. Separation of PFOS from TUDCA, TCDCA and TCDA using a biphenyl column. www.perkinelmer.com 12Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Figure 6. Separation of PFOS from TUDCA, TCDCA and TCDA using a SPP C18 column (4.6 X 100 mm) with acetonitrile organic mobile phase. Figure 7. Separation of PFOS from TUDCA, TCDCA and TCDA using a SPP C18 column (3.0 X 100 mm) with acetonitrile organic mobile phase. www.perkinelmer.com 1314 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Separation of PFAS analyte from its branched isomers As required by EPA method 1633 (Section 7.3.3), quantitative standards containing a mixture of branched and linear isomers must be used for target analytes if they are commercially available. Currently, eleven of these PFAS standards are available, including PFOA, PFOS, PFHxS, PFOSA, PFNA, NMeFOSAA, NEtFOSAA, NMeFOSA, NEtFOSA, NMeFOSE, and NEtFOSE. Since it is required to report the PFAS analyte and its isomers as a single result calculated from the combined responses of the linear and branched isomers, it is necessary to demonstrate the separation of linear and branched PFAS and the calculation with a calibration curve for each PFAS group, as shown in Figure 8. Figure 8. Calibration curve and separation of PFAS linear and branched isomers. Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Figure 8. Calibration curve and separation of PFAS linear and branched isomers. Continued... www.perkinelmer.com 15Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Figure 8. Calibration curve and separation of PFAS linear and branched isomers. Continued... www.perkinelmer.com 16Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Figure 8. Calibration curve and separation of PFAS linear and branched isomers. Continued... www.perkinelmer.com 17Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Effect of ionization energy on PFAS signal intensity The effects of ionization voltages on PFAS analytes signals were studied at different voltages. The earlier eluting analytes (more polar compounds) have better signals at lower ionization energy (such as at -2000V), while the signals are significantly higher at higher voltage (such as at -3800V) for most other PFAS analytes, especially for those less polar compounds as demonstrated in Figure 9. Figure 9. Effects of ionization energy on PFAS signals (red: -2000V; green: -3800V). The effect of ammonium acetate concentrations on PFAS signals The effects of ammonium acetate in the mobile phase on the PFAS signals were studied by using its concentrations at 2mM, 5mM, and 10mM in aqueous mobile phase. As illustrated in Figure 10, for most analyte peaks (except for the last two peaks), the retention times increase with increased concentration of ammonium acetate from 2mM to 10mM. However, signal intensities do not show significant enhancement for most analytes. Thus, it is recommended to use low concentration, such as 2mM or 5mM, to prevent the salt from precipitating at column inlet due to low solubility of the ammonium salt in acetonitrile. Figure 10. Effects of different concentrations of ammonium acetate on PFAS signals (red - 2mM; green – 5mM; and blue – 10mM). www.perkinelmer.com 18Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Effect of injection volume on PFAS analysis To determine trace amount of PFAS at ppt (ng/L) level, it is required to use a large volume of the water sample, such as 500mL as described in the EPA method, to pass through the SPE cartridge and concentrate to a small volume (such as 5mL as in the EPA method), which corresponds to a 100-folder concentration factor. However, it takes more than an hour to complete this SPE clean-up and concentration steps, which is the bottle neck limiting the lab productivity. As mentioned in the previous section about column size effects on separation and signal intensity, using a relatively larger id column (such as 4.6mm in this study) can improve PFAS separation and allow injection of a large volume of sample on the LC column. Instead of injecting 2µL as described in the EPA method, it is possible to inject up to 10 µL of the sample when using this larger id column; this can lead to a substantial increase in PFAS signal intensity without affecting peak shape, as shown in Figure 11. In addition, the precision and accuracy of the analysis can also be improved using large-volume injection. Due to the increase in PFAS signal, it is possible to use smaller volumes of water samples, such as 250 mL or even 100 mL instead of 500 mL, during the sample preparation procedure to achieve the same sensitivity and improve lab productivity (save time/make more money). Since the StayClean® source of QSight has proven robustness for injecting large volumes of samples, this gives us an advantage in our applications compared to our competitors to improve customer lab’ productivity and profit. Figure 11. Effects of injection volume on PFAS signals (red: 3µL; green: 6µL). www.perkinelmer.com 19Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples LC/MS analysis of 40 PFAS with 24 EIS and 7 NIS After optimization of the above LC-MS conditions, all the 40 targeted PFAS analytes, 24 extracted internal standards (EIS), and 7 non-extracted internal standards (NIS) can be analyzed within 12 minutes with great peak shapes as shown in Figures 12-14. Figure 12. MS/MS chromatograms for the forty native PFAS analytes. Figure 13. MS/MS Chromatograms of the twenty-four EIS. www.perkinelmer.com 20Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Figure 14. MS/MS Chromatograms of the seven NIS. www.perkinelmer.com 21Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Method Performance and Verification Method selectivity and PFAS confirmation from samples The selectivity of this LC/MS/MS method and PFAS analyte confirmation from samples were evaluated by comparing the analyte retention time and MS information such as the Ion Abundance Ratios (IAR) of quantifier to qualifier ions of the analyte between reference standard and tested samples. According to the regulatory guidance on analytical method validation, at least two structurally specific MS/MS transition ion pairs should be used in the LC/MS/MS method.17-19 The US EPA method 1633 requires that the IAR in samples must fall within ± 50% of the IAR observed in the mid-point initial calibration standard and this ratio requirement does not apply for PFBA, PFPeA, NMeFOSE, NEtFOSE, PFOSA, PFMPA, and PFMBA because suitable (not detectable or inadequate S/N) secondary transitions are not available.12 In this study, whenever possible, two or three MS/MS ion pairs were employed for each analyte in the method to identify the peaks of PFAS in the studied samples and the IAR values for all the determined PFAS analytes in authentic water samples were within ±40% of the tolerance windows of the expected values except for PFBA and PFOSA (their secondary transitions have inadequate S/N). The Simplicity™ 3Q Software can automatically calculate the Ion Ratio (IR) of qualifier to quantifier ions after integration of PFAS peaks, as highlighted in green color in Figure 15. To keep consistency with EPA method, we can convert IR to IAR easily using the equation of IAR = 1/IR. Figure 15. LC/MS/MS chromatograms and ion ratio (IR) for PFAS confirmation in samples. www.perkinelmer.com 22Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Figure 15. LC/MS/MS chromatograms and ion ratio (IR) for PFAS confirmation in samples. Continued... Instrument sensitivity (LOQ), method detection limit (MDL) and linearity Instrument sensitivity is established by measuring the signalto-noise ratio (S/N) of the lowest calibration standard (refer to sections 7.3.4, 10.3.3.1, and 13.3 of the EPA method). In this study, the limits of detection (LOD) were established when S/N ≥ 3:1 for the quantification ions of PFAS analytes. The limits of quantification (LOQ) were determined by S/N ≥ 3:1 for both the quantification ions and the confirmation ions, or S/N ≥ 10:1 if the analyte only has a quantification ion. However, when analyzing the real-world samples, sample matrix effects can affect the sensitivity, either suppressing or enhancing the analyte signal intensity. Thus, it is needed to monitor the change in sensitivity when analyzing real samples as discussed in future application notes for real sample matrices. The Method detection limit (MDL) is the minimum measured concentration of a substance that can be reported with 99% confidence that the measured analyte concentration is distinguishable from method blank results. Each laboratory must establish MDLs for all the target analytes using the MDL procedure at 40 CFR Part 136, Appendix B (see section 9.2.2 of EPA method 1633) or the reference.20 In this study, three low-level OPR (LLOPR) samples and three method blank samples were prepared and analyzed on three separate days and then the PFAS concentrations were calculated from all nine LLOPR samples and www.perkinelmer.com nine blanks. LLOPR is an ongoing precision and recovery sample that is spiked at twice the concentration of the laboratory’s LOQ and used as a routine check of instrument sensitivity. Since blank samples had no PFAS detected, the results from the nine LLOPR samples were used to calculate the average PFAS amount and standard deviation (STDEV) for each analyte. The MDL = t (n−1, 0.99) x STDEV, where t (n−1, 0.99) = 2.896 when n = 9. Thus, the MDL = 2.896 x STDEV. Table 6 list the MDL and LOQ values obtained in this study and the pooled results from interlaboratory study organized by EPA (see Table 9 of EPA method). Instrument linearity was studied by calculating the relative standard error (RSE%) for each target analyte and EIS compound for all the analyzed initial calibration standards. The RSE% for all target analytes and EIS compounds are all ≤ 20% to establish instrument linearity (section 10.3.3.3 of EPA method). In addition, method linearity was also studied by internal standard calibration method. Good linearity was obtained for each PFAS analyte among the concentration ranges studied with regression coefficients (R2) greater than 0.99. 2324 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples PFAS Analyte MDL (ng/L) LOQ (ng/L) EPA MDL (ng/L) EPA LOQ Range (ng/L) PFBA 0.78 1.00 0.79 4 - 16 PFMPA 1.30 1.00 1.46 4 - 16 3:3FTCA 2.31 6.30 2.47 5 - 20 PFPeA 0.96 1.00 0.54 2 - 8 PFMBA 1.40 1.00 1.41 4 - 15 NFDHA 1.45 3.57 0.75 2 - 7 4:2FTS 1.38 1.00 1.69 4 - 15 PFBS 0.47 0.13 0.37 1 - 4 PFHxA 0.78 1.25 0.46 1 - 4 HFPO-DA 1.17 1.00 0.51 2 - 8 PFEESA 1.45 1.00 1.17 2 - 8 5:3FTCA 6.52 3.13 9.59 25 - 100 PFHpA 0.55 0.65 0.37 1 - 4 PFPeS 0.18 0.13 0.50 1 - 4 ADONA 2.04 2.22 0.50 2 - 8 6:2FTS 2.28 5.00 2.45 4 - 15 PFOA 0.58 2.08 0.54 1 - 4 PFHxS 0.74 0.13 0.54 1 - 4 7:3FTCA 8.49 3.13 8.71 25 - 100 PFNA 0.70 2.50 0.45 1 - 4 PFHpS 0.68 0.16 0.50 1 - 4 8:2FTS 2.26 2.00 2.50 4 - 15 NMeFOSAA 0.78 1.30 0.68 1 - 4 PFDA 0.90 2.50 0.52 1 - 4 PFOS 0.60 0.13 0.63 1 - 4 NEtFOSAA 0.86 1.30 0.59 1 - 4 PFUnA 0.88 2.50 0.45 1 - 4 9Cl-PF3ONS 1.89 2.00 1.38 4 - 15 PFNS 0.71 0.13 0.47 1 - 4 PFDoA 0.51 2.50 0.40 1 - 4 PFDS 0.75 0.13 0.60 1 - 4 PFOSA 0.61 0.13 0.32 1 - 4 PFTrDA 0.97 2.50 0.46 1 - 4 11Cl-PF3OUdS 1.69 2.00 1.67 4 - 15 PFTeDA 0.83 2.50 0.49 1 - 4 PFDoS 0.75 0.26 0.60 1 - 4 NMeFOSE 2.96 2.50 3.81 10 - 40 NMeFOSA 0.73 0.26 0.43 1 - 4 NEtFOSE 3.58 2.50 4.84 10 - 40 NEtFOSA 0.53 0.26 0.45 1 - 4 Table 6. Method detection limits (MDL) and limits of quantification (LOQ) obtained in this study and the pooled results from interlaboratory study organized by EPA.25 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Method accuracy: recovery of PFAS from QC samples As shown in Tables 7, the recovery results for the QC samples are within the ranges set by the EPA method (Table 5 in EPA method). The precision as RSD % of the measurements are within 20% (n = 3). In addition, the method’s accuracy was further validated using a proficiency testing (PT) sample obtained from Absolute Standards Inc. and the results for all the 40 PFAS are within the acceptable ranges (data available upon request). PFAS Spiked (ng/L) QC1 Recovery (%) Spiked (ng/L) QC2 Recovery (%) PFBA 20 98.5 100 99.1 PFMPA 10 98.0 50 102 3:3FTCA 25 96.5 125 90.8 PFPeA 10 92.6 50 88.2 PFMBA 10 91.8 50 91.8 NFDHA 10 94.3 50 90.6 4:2FTS 40 97.2 100 99.2 PFBS 5 92.6 25 87.3 PFHxA 5 91.7 25 89.8 HFPO-DA 20 94.9 100 99.2 PFEESA 10 92.6 50 101 5:3FTCA 125 91.5 625 90.2 PFHpA 5 92.0 25 96.3 PFPeS 5 98.9 25 88.7 ADONA 20 94.3 100 98.9 6:2FTS 20 14.6 100 90.4 PFOA 5 95.4 25 102 PFHxS 5 90.7 25 90.1 7:3FTCA 125 95.1 625 98.1 PFNA 5 95.1 25 103 PFHpS 5 91.7 25 91.7 8:2FTS 20 92.6 100 110 NMeFOSAA 5 91.9 25 91.6 PFDA 5 91.1 25 89.9 PFOS 5 90.4 25 98.7 NEtFOSAA 5 89.9 25 89.3 PFUnA 5 89.2 25 104 9Cl-PF3ONS 20 92.6 100 86.2 PFNS 5 88.1 25 98.6 PFDoA 5 95.3 25 99.7 PFDS 5 89.5 25 89.5 Table 7. Recovery of PFAS from QC samples.Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples Table 7. Recovery of PFAS from QC samples. Continued... PFAS PFOSA PFTrDA 11Cl-PF3OUdS PFTeDA PFDoS NMeFOSE NMeFOSA NEtFOSE NEtFOSA Spiked (ng/L) 5 5 20 5 5 50 QC1 Recovery (%) 91.5 84.9 91.7 93.0 Spiked (ng/L) 25 25 QC2 Recovery (%) 91.5 84.7 100 25 87.6 93.2 5 50 5 www.perkinelmer.com 104 94.7 98.2 25 200 25 250 25 91.7 93.0 87.6 108 104 94.7 98.2 26. Introduction The development of an efficient and robust strategy for identification and quantification of PFAS is essential for PFAS monitoring and risk assessment. So far, US EPA 1633 method is the most comprehensive and validated method covering 40 PFAS in different sample matrices including surface water, wastewater, soil, sediment, biosolids, and fish and shellfish tissue.1 The sample preparation procedures are different for different sample matrices. Since surface water and wastewater have close relationship with human activities in many areas around the world, there have been many studies on PFAS exposure in these water resources and the PFAS concentrations were found in the range from low ng/L to μg/L levels.1-12 This is the second in a series of application notes that will address method development, sample preparation for different matrices, and method performance of applying EPA 1633 with a comprehensive workflow of PerkinElmer technologies. In this application note, we focused on analysis of PFAS in real surface water and wastewater samples using the LC/MS/MS method and workflow developed in part 1 on QSight 500 and 420 systems.13 Method performance including sample matrix effect, extraction efficiency and analyte recovery from sample matrices were evaluated by automated SPE sample preparation procedures. Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 2: Application to Surface Water and Wastewater Samples HIGHLIGHTS n A comprehensive workflow is presented for 40 PFAS analysis in authentic surface water and wastewater samples based on EPA 1633 procedures. n All performance requirements of EPA 1633 are met using less water samples such as 250 mL instead of 500 mL, which can not only significantly reduce SPE sample preparation time and improve laboratory productivity, but also cut the cost for sample collection, shipment, and storage. n Using a dual-layer (WAX/GCB) SPE cartridge can not only reduce sample preparation steps and time, but also improve analyte recovery and accuracy by streamlining the SPE process. In addition, it overcomes the hazards of working with dispersive GCB. n Workflow performance is verified by a thirdparty proficiency testing sample result.Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 2: Application to Surface Water and Wastewater Samples Some PFAS compounds were determined at low ng/L level from seven surface water samples and eight wastewater samples studied. Although perfluorooctanesulphonic acid (PFOS) and perfluorooctanoic acid (PFOA) have been regulated and phased-out of production by industry since the early 2000s, they are still detected in wastewater and surface water samples due to their persistence in the environment. Experimental Safety First Several PFAS, including perfluorooctanoic acid (PFOA), have been described as likely to be carcinogenic to humans. Exposure to these compounds should be reduced to the lowest possible level. Personal protection equipment and safety training must be provided, safety procedures must be implemented before and during work. Refer to EPA 1633 sections 5.0 to 5.4 for details.1 Chemicals, standard preparation, and analytical conditions Refer to Application Note Part 1 for analytical conditions (chemicals, calibration standards preparation and instrument parameters).13 Sample preparation Refer to Section 11 of EPA method 1633 for detailed sample preparation procedures.1 Briefly, the nominal sample size for wastewater, surface water, and their associated quality control (QC) samples is 500 mL; a lower volume of samples such as 250 mL may be used if high levels of PFAS are present in the samples, or the instrument has much higher sensitivity with a relatively large injection volume (such as 6 µL instead of 2 µL). In this study, both 500 mL and 250 mL of the same aqueous samples were evaluated to check the difference in detectability of the method. Homogenize the sample by inverting the sample 3 to 4 times and allowing the sample to settle. Do not filter the sample. The volume of the aqueous sample is determined by weighing the full sample bottle and then the empty sample bottle. Weigh each sample bottle (with the lid) to 0.1 g. A method blank and two ongoing precision and recovery (OPRs) QC samples were prepared using PFAS-free reagent water in HDPE bottles. Spike one OPR sample with the mixed PFAS stock solution (EPA-1633STK) at 2x the LOQ (LLOPR). A 10 µL of stock solution was spiked in this study. This sample was used to verify the LOQ. Spike the other OPR sample at the concentration of the mid-level calibration point and a 50 µL of PFAS stock solution was used in this study. This sample was used as the traditional OPR or regular QC sample for routine analysis. To evaluate sample matrix effects and the method’s accuracy, laboratory-fortified samples (LFM samples) were prepared by spiking 50 µL of the PFAS stock solution into each sample matrix. In this study, one surface water, one raw influent sample and one final effluent sample were used as field sample matrices to prepare the LFM samples. Spike 25 µL of EIS stock solution directly into each of the samples in their original bottles and mix the solutions by swirling the sample containers. Using a PFAS-free pipette, transfer a few drops of the sample to a pH paper and check that the pH is 6.5 ± 0.5. If necessary, adjust pH with 50% formic acid or ammonium hydroxide (or with 5% formic acid and 3% aqueous ammonium hydroxide). The samples are now ready for solid-phase extraction (SPE) and cleanup. Sample extraction and clean-up by automated SPE All method blanks, water samples, QC samples and fortified LFM samples were extracted and concentrated by the SPE procedures described in detail in Section 12.0 of EPA Method 1633. Slight modifications were made as allowed by the method. Using stacked WAX/GCB SPE cartridges instead of WAX SPE cartridges plus the loose carbon clean-up reduced sample extraction and clean-up time and minimized manual operation errors. The SPE cartridges were pre-conditioned with 15 mL of 1% methanolic ammonium hydroxide and 5 mL of 0.3M formic acid. Then, water samples were added to the cartridges by a syringe pump at 5 mL/min. The reservoir walls were rinsed with 5 mL reagent water (twice) followed by 5 mL of 1:1 0.1M formic acid/methanol and passed through the cartridge using a vacuum. The cartridge was dried by pulling air through for 15 seconds, and the rinse solution was discarded. The inside of the sample bottles was rinsed with 5 mL of 1% methanolic ammonium hydroxide, shaken for 10 seconds, then the rinse was transferred to the SPE reservoir, washing the walls of the reservoir, while pushing the elution solvent through the cartridge and into the collection tubes. Inlet filters were used between sample bottles and SPE cartridges to prevent particles from blocking the SPE cartridges. All these SPE procedures were carried out automatically using a PromoChrom SPE-03 system (Promochrom Technologies, Richmond, BC, Canada) to reduce labor-intensive human work and minimize human errors. Results and Discussion Method Performance and Validation Method selectivity and PFAS confirmation from samples Method selectivity and PFAS analyte confirmation from samples were evaluated by comparing the analyte retention time and MS information such as the Ion Abundance Ratios (IAR) of quantifier to qualifier ions of the analyte between reference standard and tested samples. According to the regulatory guidance on analytical method validation, at least two structurally specific MS/MS transition ion pairs should be used in the LC/MS/MS method.14-16 The US EPA method 1633 requires that the IAR in samples must fall within ± 50% of the IAR observed in the mid-point initial calibration standard and this ratio requirement does not apply for PFBA, PFPeA, NMeFOSE, NEtFOSE, PFOSA, PFMPA, and PFMBA because suitable (not detectable or inadequate S/N) secondary www.perkinelmer.com 23 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 2: Application to Surface Water and Wastewater Samples transitions are not available.1 In this study, whenever possible, two or three MS/MS ion pairs were employed for each analyte in the method to identify the peaks of PFAS in the studied samples and the IAR values for all the determined PFAS analytes in authentic water samples were within ±40% of the tolerance windows of the expected values except for PFBA and PFOSA (their secondary transitions have inadequate S/N). Refer to the final section of this application note on sample analysis results and the IAR values (Tables 4-5). Instrument sensitivity (LOQ), method detection limit (MDL) and linearity Refer to Table 6 in Part 1 of the series of applications for LOQ and MDL values for aqueous samples.13 Using either 500 mL or 250 mL of aqueous samples during sample preparation did not affect the LOQ and MDL significantly due to the high sensitivity of QSight 500 and 420 systems and the additional sensitivity enhancement from larger volume injection as demonstrated in Figure 11 in our previous study.13 Sample matrix effects and instrument sensitivity check Sample matrix effects were studied by the recovery of nonextracted internal standards (NIS) from samples by comparing the peak area of NIS in the real samples to the average peak area NIS Spiked (ng/L) Recovery (%) QC Recovery (%) surface water1 Recovery (%) surface water2 Recovery (%) wastewater1 Recovery (%) wastewater2 13C3-PFBA 50 113 91.4 93.0 131 122 13C2-PFHxA 25 118 94.7 96.7 137 128 13C4-PFOA 25 119 96.2 92.4 135 126 18O2-PFHxS 25 114 96.4 97.8 144 137 13C5-PFNA 12.5 121 95.6 95.7 136 134 13C2-PFDA 12.5 123 96.5 102 139 137 13C4-PFOS 25 122 107 107 156 145 Table 1. Recovery of non-extracted internal standards (NIS) from samples. of NIS in the calibration standard. EPA method also used these results to check instrument sensitivity change (section 14.9 of EPA method). As shown in Table 1, all the recovery results are within the range of 50 – 200% set by the EPA method (Table 6 in EPA method). Wastewater samples seemed to have some signal enhancement effects (higher recoveries). Method extraction efficiency Method extraction efficiency was evaluated by the recovery of the extracted internal standards (EIS) from the field water samples and the QC samples calculated using response factor (comparing peak area ratio/concentration ratio of EIS to those of NIS in samples). As shown in Table 2, all the recovery results are within the ranges set by the EPA method (Table 6 in EPA method). Method accuracy: recovery of target PFAS from field water samples As shown in Table 3, the PFAS recovery results from field samples are within the ranges set by the EPA method (Table 5 in EPA method). The precision as RSD % of the measurements are within 20% (n = 3). In addition, the method’s accuracy was further validated using a proficiency testing sample obtained from Absolute Standards Inc. and the results for all the 40 PFAS are within the acceptable ranges (data available upon request). EIS Spiked (ng/L) Recovery (%) QC Recovery (%) Surface water Recovery (%) Wastewater-INF Recovery (%) Wastewater-EFF 13C4-PFBA 100 91.6 127 84.2 54.6 13C5-PFPeA 50 92.2 120 84.9 87.2 13C2-4:2FTS 50 100 76.6 41.5 71.9 13C5-PFHxA 25 95.8 120 88.8 92.7 13C3-PFBS 25 93.7 118 84.9 82.5 13C3-HFPO-DA 100 100 125 92.0 99.1 13C4-PFHpA 25 98.1 128 90.1 94.9 13C2-6:2FTS 50 101 95.5 56.5 85.3 13C8-PFOA 25 99.1 118 79.8 72.6 13C3-PFHxS 25 98.4 126 86.6 116 13C9-PFNA 12.5 96.5 126 81.3 126 13C2-8:2FTS 50 93.1 97.3 68.9 82.2 D3-NMeFOSAA 50 90.4 105 71.6 63.1 13C6-PFDA 12.5 95.2 133 86.9 86.2 D5-NEtFOSAA 25 86.4 100 64.2 65.8 Table 2. Recovery of extracted internal standards (EIS) from samples. The table continues on next page4 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 2: Application to Surface Water and Wastewater Samples PFAS Spiked (ng/L) Recovery (%) (surface water) Recovery (%) (Wastewater) Spiked (ng/L) Recovery (%) (surface water) Recovery (%) (Wastewater) PFBA 40 95.5 95.1 100 98.3 101 PFMPA 20 96.1 99.2 50 103 95.7 3:3FTCA 50 87.5 84.5 125 91.8 90.1 PFPeA 20 101 115 50 103 116 PFMBA 20 88.1 87.8 50 92.8 93.9 NFDHA 20 87.6 91.0 50 96.5 96.6 4:2FTS 40 91.7 85.6 100 99.0 96.2 PFBS 10 106 77.9 25 99.3 90.9 PFHxA 10 115 82.9 25 111 88.4 HFPO-DA 40 96.5 87.9 100 93.5 95.2 PFEESA 20 93.5 101 50 104 107 5:3FTCA 250 89.4 98.0 625 95.7 101 PFHpA 10 105 95.1 25 99.0 97.0 PFPeS 10 90.4 93.6 25 94.9 89.2 ADONA 40 89.9 86.6 100 97.1 99.8 6:2FTS 40 82.3 84.4 100 107 94.1 PFOA 10 110 117 25 99.2 85.9 PFHXS 10 101 107 25 97.6 98.5 7:3FTCA 250 93.4 96.1 625 104 102 PFNA 10 96.9 89.4 25 106 96.8 PFHpS 10 87.9 86.7 25 90.6 90.8 8:2FTS 40 95.3 96.1 100 108 107 NMeFOSAA 10 90.2 93.3 25 95.3 98.4 PFDA 10 91.9 91.9 25 93.4 100 PFOS 10 106 108 25 99.2 94.2 NEtFOSAA 10 92.0 92.5 25 97.6 97.9 PFUnA 10 82.5 89.0 25 93.6 90.0 9Cl-PF3ONS 40 89.4 84.7 100 93.8 94.4 PFNS 10 85.1 79.7 25 89.0 86.1 PFDoA 10 90.2 90.5 25 103 107 PFDS 10 80.3 72.2 25 86.4 79.6 PFOSA 10 90.7 88.0 25 94.8 95.1 PFTrDA 10 89.8 102 25 88.8 104 11Cl-PF3OUdS 40 83.9 72.7 100 85.5 86.0 PFTeDA 10 95.2 87.7 25 93.9 96.4 PFDoS 10 73.6 62.8 25 82.7 62.7 NMeFOSE 100 88.2 107 200 98.5 105 NMeFOSA 10 93.8 118 25 111 143 NEtFOSE 100 88.0 82.5 250 94.3 93.9 NEtFOSA 10 90.7 109 25 110 113 Table 3. Recovery of target PFAS from spiked field water samples. EIS Spiked (ng/L) Recovery (%) QC Recovery (%) Surface water Recovery (%) Wastewater-INF Recovery (%) Wastewater-EFF 13C8-PFOS 25 93.6 121 81.1 84.4 13C7-PFUnA 12.5 89.7 124 77.0 66.3 13C2-PFDoA 12.5 86.6 112 64.2 51.5 13C8-PFOSA 25 83.7 111 82.0 80.5 13C2-PFTeDA 12.5 72.7 89.6 49.3 55.1 D7-NMeFOSE 250 77.1 84.0 56.7 49.8 D3-NMeFOSA 25 74.9 82.8 63.9 60.7 D9-NEtFOSE 250 70.3 73.9 52.6 47.9 D5-NEtFOSA 25 72.0 72.5 53.4 59.0Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 2: Application to Surface Water and Wastewater Samples Figure 1. Surface water samples (S1 – S7) were collected along the lakeshore areas, Lake Ontario in Toronto, Canada. Surface water and wastewater sample analysis Seven surface water samples were collected from Lakeshore area in Toronto, Ontario (Figure1): four samples (S1 to S4) were collected from different locations of Lake Ontario (S1 from Rotary Park area, S2 from Sunnyside Park Beach, S3 from Humber Bay Shore, and S4 from Swan Pond), one sample from Mimico Creek (S5), one sample from Grenadier Pond (S6) in High Park, and one sample from Humber River (S7). The developed SPE extraction and LC/MS/MS methods were first applied to the analysis of PFAS in these seven surface water samples. As shown in Table 4, among the 40 PFAS compounds studied, ten of the PFAS analytes were found in these samples. The contents of PFAS in these surface waters are similar to those published by Benoit Lalonde5 and at the lower concentration ranges. Slightly higher PFAS were found in areas close to apartment buildings and human activities such as samples S4 and S5, while lower PFAS were detected in more remote open lake areas such as samples S1 and S2. Recently, there has been increasing human activities around the Swan Pond area where younger people are playing with their jet skis (water motorcycles) which may contaminate the water and result to higher PFAS in sample S4. These results align with previous studies, confirming that the sources of PFAS in surface waters are attributed to urban runoff and atmospheric deposition. Figure 2 shows some LC/MS/MS chromatograms of PFAS obtained from surface water samples, some of the branched isomers from PFOS, PFOA, PFHxS, and PFOSA are also identified and separated from their linear isomers. The SPE and LC/MS/MS methods were then applied to the analysis of PFAS in eight wastewater samples from Canadian wastewater treatment plants (WWTPs). These samples include four raw influent samples (W-INF) and four final effluent samples (W-EFF). As shown in Table 5, nine of the 40 PFAS compounds studied were found from all the wastewater samples, and 5:3FTCA, 4:2FTS, and 6:2FTS were only detected from raw influent samples (W- INF) samples. These results agree with the most recently published results on PFAS in Canadian municipal wastewater2. Figure 3 shows the LC/ MS/MS chromatograms of PFAS obtained from wastewater samples. The analyte PFBS, 6:2FTS, and 4:2FTS peaks are well separated from matrix interfering peaks and some of the branched isomers from PFOS, PFOA, PFHxS, and PFOSA are also separated from their linear isomers. Although the legacy PFAS such as PFOS and PFOA have been regulated and phased-out of production in Canada since the early 2000s, they are still detected in wastewater and surface water samples due to their persistence in the environment. www.perkinelmer.com 56 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 2: Application to Surface Water and Wastewater Samples PFAS S1 (ng/L) S2 (ng/L) S3 (ng/L) S4 (ng/L) S5(ng/L) S6 (ng/L) S7 (ng/L) QC (ng/L) Blank (ng/L) PFBA 2.88 2.59 2.54 5.03 7.31 6.81 4.05 19.5 <MDL PFPeA 1.98 1.37 1.60 4.26 5.37 4.86 7.76 8.98 <MDL IAR 50 50 50 50 33 33 33 33 NA PFBS 0.920 0.770 0.670 2.49 3.51 2.93 1.81 4.62 <MDL IAR 2.3 2.4 2.5 2.6 2.9 2.8 2.6 2.6 NA PFHxA 0.980 1.31 2.01 2.59 4.42 3.74 8.14 4.32 <MDL IAR 6.3 6.7 7.1 5.3 5.6 6.3 7.1 7.1 NA PFHpA 1.42 0.820 1.32 1.95 1.44 0.700 2.91 5.01 <MDL IAR 1.6 1.3 1.4 1.4 1.2 1.5 1.3 1.3 NA PFPeS 0.240 0.140 0.140 0.390 0.200 0.210 0.270 4.79 <MDL IAR 2.6 2.9 3.2 2.0 2.4 2.3 2.5 2.4 NA PFOA 2.11 2.38 2.48 3.72 2.41 2.72 2.35 4.87 <MDL IAR 0.8 1.0 1.0 0.94 0.8 1.0 0.94 0.90 NA PFHxS 1.36 0.78 1.01 3.51 1.54 0.600 2.05 5.12 <MDL IAR 2.7 2.6 2.5 2.4 3.0 2.3 2.5 2.6 NA PFOS 2.88 4.45 2.72 18.6 4.88 0.87 2.54 4.77 <MDL IAR 3.3 3.2 3.4 3.3 3.1 3.3 3.0 3.0 NA PFOSA 0.190 0.230 0.240 0.350 0.290 0.240 0.190 4.29 <MDL Table 4. PFAS results from the surface water samples (ng/L) and the IAR values. Figure 2. LC/MS/MS chromatograms of PFAS in surface water samples.7 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 2: Application to Surface Water and Wastewater Samples PFAS W-INF1 (ng/L) W-INF2 (ng/L) W-INF3 (ng/L) W-INF4 (ng/L) W-EFF1 (ng/L) W-EFF2 (ng/L) W-EFF3 (ng/L) W-EFF4 (ng/L) QC (ng/L) Blank (ng/L) PFBA 6.51 3.44 4.42 11.9 7.39 7.19 9.08 8.72 19.5 <MDL PFPeA 6.77 8.11 4.86 20.5 19.0 17.5 11.1 10.4 8.98 <MDL IAR 33 33 33 33 33 33 33 33 33 NA PFBS 3.04 4.26 3.04 22.4 4.33 4.20 4.55 4.01 4.62 <MDL IAR 2.3 2.4 2.4 2.6 2.3 2.3 2.7 2.4 2.6 NA PFHxA 9.71 7.81 4.95 14.5 15.9 14.8 16.1 14.3 4.32 <MDL IAR 9.1 8.3 9.1 7.7 8.3 7.7 8.3 7.1 7.1 NA PFHpA 2.64 2.40 1.63 1.89 1.71 2.17 2.85 2.71 5.01 <MDL IAR 1.2 1.5 1.5 1.3 1.3 1.4 1.3 1.2 1.3 NA PFOA 6.85 6.54 4.06 3.93 4.32 4.82 7.92 7.06 4.87 <MDL IAR 0.94 0.88 0.86 0.90 0.88 1.0 1.0 0.96 0.90 NA PFHxS 1.00 1.62 0.600 0.860 2.62 2.67 2.78 2.90 5.12 <MDL IAR 2.9 3.2 2.6 3.3 2.5 2.9 3.0 3.0 2.6 NA PFOS 3.46 3.19 2.44 1.42 1.59 1.69 3.51 3.69 4.77 <MDL IAR 3.2 2.9 2.7 2.6 3.1 3.2 3.3 3.6 3.0 NA PFOSA 0.410 0.900 0.220 <LOQ 0.240 0.240 0.310 0.300 4.29 <MDL IAR NA NA NA NA NA NA NA NA NA NA 4:2FTS <LOQ 19.7 <LOQ <LOQ <LOQ <LOQ <LOQ <LOQ 19.3 <MDL IAR NA 1.8 NA NA NA NA NA NA 1.7 NA 6:2FTS <LOQ 24.3 2.00 659 <LOQ <LOQ <LOQ <LOQ 19.8 <MDL IAR NA 1.8 1.7 1.8 NA NA NA NA 1.8 NA 5:3FTCA 20.4 10.0 10.0 8.00 <LOQ <LOQ <LOQ <LOQ 113 <MDL IAR 1.1 1.4 1.3 1.3 NA NA NA NA 1.2 NA Table 5. PFAS results from the wastewater samples (ng/L) and the IAR values.8 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 2: Application to Surface Water and Wastewater Samples Figure 3. LC/MS/MS chromatograms of PFAS in wastewater samples. Conclusions The sensitive and robust analytical method and workflow were applied for analyzing 40 PFASs in wastewater and surface water samples at low ng/L levels using an automated solid phase extraction (SPE) for sample preparation and QSight LC/MS/MS for PFAS determination. Better separation, higher sensitivity, and a more robust method have been achieved using a larger internal diameter SPP UHPLC column and injecting more sample into column. Thus, a smaller sample volume can be used to improve the productivity of a commercial laboratory. Using a dual-layer (WAX/GCB) SPE cartridge can not only reduce sample preparation steps and time, but also improve analyte recovery and accuracy by streamlining the SPE process. All performance criteria of EPA 1633 are met for wastewater and surface water sample analysis, demonstrating equivalency of the workflow used. The method can be applied for real water sample analysis with good precision and accuracy. Acknowledgment The author is grateful to Dr. Shirley Anne Smyth from Science and Technology Branch, Environment and Climate Change Canada for providing the wastewater samples for our testing. Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 2: Application to Surface Water and Wastewater Samples References 1. US. EPA Method 1633 Final. January 2024. Analysis of Per- and Polyfluoroalkyl Substances (PFAS) in Aqueous, Solid, Biosolids, and Tissue Samples by LC-MS/MS. https://www. epa.gov/system/files/documents/2024-01/method-1633f inal-for-web-posting.pdf 2. Sarah B. Gewurtz, Alexandra S. Auyeung, Amila O. De Silva, Steven Teslic, Shirley Anne Smyth. Science of the Total Environment. 2024, 912, 168638. Per- and polyfluoroalkyl substances (PFAS) in Canadian municipal wastewater and biosolids: Recent patterns and time trends 2009 to 2021. https://doi.org/10.1016/j.scitotenv.2023.168638. 3. Sudarshan Kurwadkar, Richard Wilkin, et al., Science of the Total Environment. 2022, 809, 151003. Per- and polyfluoroalkyl substances in water and wastewater: A critical review of their global occurrence and distribution. http://dx.doi.org/10.1016/j.scitotenv.2021.151003. 4. Lutz Ahrens, Jelena Rakovic, Siri Ekdahl, Roland Kallenborn. Chemosphere, 2024, 345, 140463. Environmental distribution of per- and polyfluoroalkyl substances (PFAS) on Svalbard: Local sources and long-range transport to the Arctic. https://doi.org/10.1016/j.chemosphere.2023.140463. 5. Benoit Lalonde, Christine Garron. Archives of Environmental Contamination and Toxicology. 2022. 82, 581. Perfuoroalkyl Substances (PFASs) in the Canadian Freshwater Environment. https://doi.org/10.1007/s00244-022-00922-x. 6. H. Ulrich, et al., Chemosphere, 2024, 349, 140893. Novel PFAS-specific monitoring approach for highly impacted surface waters. https://doi.org/10.1016/j. chemosphere.2023.140893. 7. Lingyi Meng, Boyu Song, Yawei Wang, et al., Environment International, 2021. 156,106735. Legacy and emerging per- and polyfluoroalkyl substances (PFAS) in the Bohai Sea and its inflow rivers. https://doi.org/10.1016/j. envint.2021.106735 8. Sze Yee Wee, Ahmad Zaharin Aris. Ecotoxicology and Environmental Safety. 2023, 267, 115663. Environmental impacts, exposure pathways, and health effects of PFOA and PFOS. https://doi.org/10.1016/j.ecoenv.2023.115663 PerkinElmer U.S. LLC 710 Bridgeport Ave. Shelton, CT 06484-4794 USA (+1) 855-726-9377 www.perkinelmer.com 9. M.-A. Pétré, K.R. Salk, et al., Science of the Total Environment. 2022, 831, 154763. Per- and polyfluoroalkyl substances (PFAS) in river discharge: Modeling loads upstream and downstream of a PFAS manufacturing plant in the Cape Fear watershed, North Carolina. http://dx.doi. org/10.1016/j.scitotenv.2022.154763 10. Håkon A. Langberg, et al., Environmental Pollution. 2021, 273, 116259. Paper product production identified as the main source of per- and polyfluoroalkyl substances (PFAS) in a Norwegian lake: Source and historic emission tracking. https://doi.org/10.1016/j.envpol.2020.116259. 11. M. Zare˛bska, S. Bajkacz. Trends in Analytical Chemistry. 2023, 163, 117062. Poly- and perfluoroalkyl substances (PFAS) - recent advances in the aquatic environment analysis. https://doi.org/10.1016/j.trac.2023.117062. 12. Qi Wang, Yuefei Ruan, Paul K.S. Lam, et al., Trends in Analytical Chemistry. 2023, 169, 117351. Tracing per- and polyfluoroalkyl substances (PFASs) in the aquatic environment: Target analysis and beyond. https://doi. org/10.1016/j.trac.2023.117351. 13.PerkinElmer Application Note: Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples. 14. USA. FDA, Bioanalytical Method Validation Guidance for Industry, 2018. https://www.fda.gov/downloads/drugs/ guidances/ucm070107.Pdf. 15. European Commission, 2002/657/EC: Commission decision of 12 August 2002 implementing Council Directive 96/23/EC concerning the performance of analytical methods and the interpretation of results, Off. J. Eur. Communities. 2002. 16. European Commission, Guidance SANTE 11312/2021 – Analytical quality control and method validation procedures for pesticide residues analysis in food and feed. https:// www.accredia.it/en/documento/guidance-sante-113122021-analytical-quality-control-and-method-validationprocedures-for-pesticide-residues-analysis-in-food-and-feed. Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Introduction The development of an efficient and robust strategy for identification and quantification of per- and polyfluoroalkyl substances (PFAS) is essential for PFAS monitoring and risk assessment. So far, US EPA 1633 method is the most comprehensive and validated method covering 40 PFAS in different sample matrices including surface water, wastewater, soil, sediment, biosolids, and fish and shellfish tissue.1 The sample preparation procedures are different for different sample matrices. In this application note, we focus on sample preparation and analysis for different solid samples, using a dual-phase (GCB/WAX) SPE cartridge for sample extraction and clean-up after solvent extraction and a QSight LC/MS/MS system for PFAS analysis. This is the third in a series of application notes that address method development, sample preparation for different matrices, and method performance of applying EPA 1633 with a comprehensive workflow of PerkinElmer technologies. For LC/MS/MS method optimization and workflow development for surface water and wastewater samples, please refer to application note part one and part two.2-3 HIGHLIGHTS n A comprehensive workflow is presented for 40 PFAS analysis in sand, soil, sediment, sludge, biosolid, and related solid samples based on EPA 1633 procedures. n All performance requirements of EPA 1633 are met for solid samples, demonstrating the capability and equivalency of this workflow for the complex sample matrices. n Using a dual-layer (GCB/WAX) SPE cartridge can not only reduce sample preparation steps and time, but also improve analyte recovery and accuracy by streamlining the SPE process. In addition, it overcomes the hazards of working with dispersive GCB. n The workflow performance is verified by a third-party proficiency testing sample result. APPLICATION NOTE Liquid Chromatography/ Mass Spectrometry AUTHOR Jingcun Wu PerkinElmer Woodbridge, ON, Canada Part 3: Application to Sand, Soil, Sediment, Sludge, Biosolid and Other Solid Samples Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 3: Application to Sand, Soil, Sediment, Sludge, Biosolid and Other Solid Samples In this study on solid samples, method performance including sample matrix effect, extraction efficiency and analyte recovery from sample matrices were evaluated by automated SPE sample preparation procedures. Eleven solid samples were analyzed by the developed workflow together with quality control samples such as the laboratory reagent blank (LRB) or method blank (MB), laboratory fortified blank (LFB), laboratory fortified matrix (LFM), and a third-party proficiency testing (PT) sample. Most of the studied samples had very low or no PFAS found, but more than 10 PFAS compounds were determined at low ng/g level from a purchased black garden soil and a sludge sample collected at Mimico Creek area in Toronto. Some long-chain PFAS compounds were found from the biosolid samples such as PFOS, PFOSA, NMeFOSAA NEtFOSAA and NEtFOSE. Experimental Safety First Several PFAS, including perfluorooctanoic acid (PFOA), have been described as likely to be carcinogenic to humans. Exposure to these compounds should be reduced to the lowest possible level. Personal protection equipment and safety training must be provided, safety procedures must be implemented before and during work. Biosolids samples may contain high concentrations of biohazards and must be handled with gloves and opened in a fume hood or biological safety cabinet to prevent exposure. Laboratory staff should know and observe the safety procedures Beach Sand (S1) Sediment (S2, S7) required in a microbiology laboratory that handles pathogenic organisms when handling biosolids samples. Refer to EPA 1633 sections 5.0 to 5.4 for details.1 Chemicals, Standard Preparation, and Analytical Conditions Refer to Application Note Part 1 for analytical conditions (chemicals, calibration standards preparation and instrument parameters).2 Sample Collection One sand sample (S1) and two sediment samples (S2 and S7) were collected from different beach areas along Lake Ontario, in Toronto. A sludge sample (S3) was collected at Mimico Creek area, in Toronto. A garden soil sample (S4) was collected from my backyard in Kitchener, Ontario. A black garden soil (S5) was purchased from a garden store and was used for planting Chinese chive (S10, a popular Chinese vegetable). A human feces sample (S6) was collected at home. Two biosolid samples (S8 and S9) were kindly provided by Dr. Shirley Anne Smyth from Science and Technology Branch, Environment and Climate Change Canada. The cooked pork blood sample (S11) was obtained from a local Asian food market in Woodbridge, Ontario. An EPA1633 proficiency testing sample (PFC PT-Soil-EPA 1633, Part # 38722, Lot# 071823) was obtained from Absolute Standards Inc (Hamden, CT, USA). All samples were collected in cleaned sample bottles (HDPE, with polypropylene caps). Figure 1 shows the solid samples collected and analyzed in this study. Mimico Sludge (S3) Garden Soil (S4) Black Garden Soil (S5) Human Feces (S6) Biosolid 1 (S8) Biosolid 1 (S9) Chinese Chive (S10) Black Garden Soil (S5) Figure 1. Sample injection sequences: Robustness was evaluated by consecutive injections of food matrix extracts that were bracketed by solvent QC replicates (100 food matrix injections for every 5 solvent QC injections). www.perkinelmer.com 2Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 3: Application to Sand, Soil, Sediment, Sludge, Biosolid and Other Solid Samples Sample Preparation The workflow for analysis of solid samples is summarized in Figure 2 below. Figure 2. LC/MS/MS workflow for solid samples analysis. To extract solid and biosolid samples, an extraction solution (0.3 % ammonium hydroxide in methanol) was prepared by adding 1 mL of ammonium hydroxide (30%) to methanol (99 mL), store at room temperature, replace after 1 month. Solid (or biosolid) samples were prepared by the following procedures (or refer to the EPA method sections 11.3 and 12.0).1  Weigh 5 g sample (0.5 g for biosolid) into a 50 mL polypropylene (PP) centrifuge tube.  Spike 25 µL of Extracted Internal Standard Mix (EIS) (MPFACHIF-ES from Wellington).  Vortex the sample to disperse the standard and allow to equilibrate for 30 minutes.  Add 10 mL of extraction solution and shake 30 mins, then centrifuge 10 mins, transfer supernatant to a clean 50 mL PP tube.  Add 15 mL of extraction solution, repeat the above procedures and transfer supernatant to the same tube.  Add 5 mL of extraction solution and shake 5 mins, then centrifuge 10 mins, transfer supernatant to the same tube.  Concentrate under nitrogen to 7 mL (at 55 °C).  Reconstitute it up to 50 mL with water, mix well.  Check pH and adjust it to about pH 6 (6.5 ± 0.5).  Perform automated SPE clean-up.  Add 25 µL of Non-extracted Internal Standard (NIS) Mix (MPFAC-HIF-IS from Wellington) and 25 µL of concentrated acetic acid to the eluate, Mix well. In this study, we used dual-layer GCB/WAX SPE cartridges instead of WAX SPE cartridges plus the loose carbon clean-up to reduce sample extraction and clean-up time and minimize manual operation errors. We used inlet filters between sample bottles and SPE cartridges to prevent particles from blocking the SPE cartridges. All these SPE procedures were carried out automatically using a PromoChrom SPE-03 system (Promochrom Technologies, Richmond, BC, Canada) to reduce labor-intensive human work and minimize human errors. Results and Discussion Method Performance and Validation Method selectivity and PFAS confirmation from samples Method selectivity and PFAS analyte confirmation from solid samples were evaluated by comparing the PFAS analyte retention time and MS information such as the Ion Abundance Ratios (IAR) of quantifier to qualifier ions of the analyte between reference standard and tested samples. According to the regulatory guidance on analytical method validation, at least two structurally specific MS/MS transition ion pairs should be used in a LC/MS/ MS method.4-6 The US EPA method 1633 requires that the IAR in samples must fall within ± 50% of the IAR observed in the midpoint calibration standard and this ratio requirement does not apply for PFBA, PFPeA, NMeFOSE, NEtFOSE, PFOSA, PFMPA, and PFMBA because suitable secondary transitions are not available for these compounds (not detectable or inadequate S/N).1 In this study, whenever possible, two or three MS/MS ion pairs were employed for each analyte in the method to identify the peaks of PFAS in the studied samples and the IAR values for all the determined PFAS analytes in the studied samples were within ±40% of the tolerance windows of the expected values except for PFBA, NEtFOSE, and PFOSA (their secondary transitions have inadequate S/N). Refer to the final section of this application note on sample analysis results and the IAR values (Tables 5-6). www.perkinelmer.com 34 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 3: Application to Sand, Soil, Sediment, Sludge, Biosolid and Other Solid Samples Instrument sensitivity (LOQ) and method detection limit (MDL) Instrument sensitivity is established by measuring the signalto-noise ratio (S/N) of the lowest calibration standard (refer to sections 7.3.4, and 10.3.3.1 of the EPA method). In this study, the limits of quantification (LOQ) were determined by S/N ≥ 3:1 for both the quantification ions and the confirmation ions, or S/N ≥ 10:1 if the analyte only has a quantification ion. The Method detection limit (MDL) is the minimum measured concentration of a substance that can be reported with 99% confidence that the measured analyte concentration is distinguishable from method blank results. Each laboratory must establish MDLs for all the target analytes using the MDL procedure at 40 CFR Part 136.7 In this study, several low-level ongoing precision and recovery (LLOPR) samples and method blank samples were prepared and analyzed on three separate days and then the PFAS concentrations were calculated from all LLOPR samples and blanks. LLOPR sample was prepared by spiking PFAS at twice the concentration of the laboratory’s LOQ and used as a routine check of instrument sensitivity. In this study, since blank samples had no PFAS detected, the results from the eleven LLOPR samples were used to calculate the average PFAS amount and standard deviation (STDEV) for each analyte. The MDL = t (n−1, 0.99) x STDEV, where t (n−1, 0.99) = 2.764 when n = 11. Thus, the MDL = 2.764 x STDEV. Table 1 list the MDL and LOQ values obtained in this study and the pooled results from interlaboratory study organized by EPA (see Table 9 of EPA method1). PFAS MDL (ng/g) LOQ (ng/g) EPA MDL (ng/g) EPA LOQ Range (ng/g) PFBA 0.14 0.05 0.15 0.64 - 1.6 PFMPA 0.13 0.10 0.07 0.32 - 0.8 3:3FTCA 0.12 0.63 0.23 0.80 - 5.0 PFPeA 0.08 0.10 0.07 0.32 - 0.8 PFMBA 0.10 0.11 0.05 0.32 - 0.8 NFDHA 0.13 0.36 0.20 0.32 - 0.8 4:2FTS 0.21 0.18 0.20 0.64 - 1.5 PFBS 0.04 0.05 0.05 0.16 - 0.4 PFHxA 0.03 0.13 0.06 0.16 - 0.4 HFPO-DA 0.21 0.28 0.25 0.64 - 1.6 PFEESA 0.11 0.16 0.08 0.32 - 0.7 5:3FTCA 0.98 0.89 0.86 4 - 10 PFHpA 0.05 0.07 0.05 0.16 - 0.4 PFPeS 0.07 0.05 0.08 0.16 - 0.4 ADONA 0.19 0.25 0.23 0.64 - 1.5 6:2FTS 0.07 1.1 0.39 0.64 - 1.5 PFOA 0.05 0.12 0.07 0.16 - 0.4 PFHxS 0.06 0.02 0.08 0.16 - 0.4 7:3FTCA 1.36 0.80 0.87 4 - 10 PFNA 0.03 0.25 0.14 0.16 - 1.3 PFHpS 0.07 0.08 0.07 0.16 - 0.4 8:2FTS 0.23 0.20 0.31 0.64 - 1.5 NMeFOSAA 0.06 0.13 0.08 0.16 - 0.4 PFDA 0.04 0.19 0.06 0.16 - 0.4 PFOS 0.08 0.04 0.07 0.16 - 0.4 NEtFOSAA 0.06 0.17 0.08 0.16 - 0.4 PFUnA 0.04 0.25 0.12 0.16 - 0.5 9Cl-PF3ONS 0.31 0.20 0.22 0.64 - 1.5 PFNS 0.05 0.02 0.07 0.16 - 0.4 PFDoA 0.03 0.10 0.06 0.16 - 0.4 PFDS 0.10 0.02 0.08 0.16 - 0.4 PFOSA 0.05 0.04 0.04 0.16 - 0.4 PFTrDA 0.05 0.20 0.07 0.16 - 0.4 Table 1. Method detection limits (MDL) and limits of quantification (LOQ) obtained in this study and the pooled results from interlaboratory study organized by EPA.Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 3: Application to Sand, Soil, Sediment, Sludge, Biosolid and Other Solid Samples Table 1. Method detection limits (MDL) and limits of quantification (LOQ) obtained in this study and the pooled results from interlaboratory study organized by EPA CONTINUED>>> PFAS 11Cl-PF3OUdS PFTeDA PFDoS NMeFOSE NMeFOSA NEtFOSE NEtFOSA MDL (ng/g) 0.42 0.04 0.08 0.55 LOQ (ng/g) 0.20 0.05 0.03 EPA MDL (ng/g) 0.18 EPA LOQ Range (ng/g) 0.64 - 1.5 0.05 0.06 0.94 0.09 0.57 0.04 Sample matrix effects and instrument sensitivity check Sample matrix effects were studied by the recovery of nonextracted internal standards (NIS) from samples by comparing the peak area of NIS in the real samples to the average peak area of NIS in the calibration standard. These results can be used to check instrument sensitivity change due to sample matrix effects (section 14.9 of EPA method). As shown in Table 2, all the recovery results are within the range of 50 – 200% set by the EPA method (Table 8 in EPA method 1). 0.10 1.19 0.14 0.36 0.07 0.35 0.07 Method extraction efficiency 0.16 - 0.4 0.16 - 0.4 1.6 - 4.0 0.16 - 0.4 1.6 - 4.0 0.16 - 0.4 Method extraction efficiency was evaluated by the recovery of the extracted internal standards (EIS) from the field samples and the QC (LFB) sample calculated using response factor (comparing peak area ratio/concentration ratio of EIS to those of NIS in samples). As shown in Table 3, all the recovery results are within the ranges set by the EPA method (Table 8 in EPA method). The recoveries are low for some long-chain PFAS (especially for the last five compounds), which are in line with the results obtained by EPA inter-laboratory study.1 Table 2. Recovery (%) of non-extracted internal standards (NIS) from samples. NIS NIS 13C3-PFBA 13C2-PFHxA 13C4-PFOA 18O2-PFHxS 13C5-PFNA 13C2-PFDA 13C4-PFOS Spiked (ng/g) 5 2.5 2.5 2.5 1.25 1.25 LFB 107 112 112 100 107 S1* 101 100 105 97.8 S3 84.7 85.9 89.1 S4 99.2 99.4 108 S5 91.5 89.4 S6 89.9 S8 S9 S10 89.3 69.5 95.8 80.3 103 114 2.5 101 102 94.0 86.8 88.7 81.2 98.6 104 100 89.8 93.4 93.1 85.7 87.3 80.0 88.2 96.1 77.9 89.3 * The recovery values from sample S2 and S7 are similar to those of S1, thus their data are not shown here. 90.4 93.8 90.8 89.4 81.1 85.2 89.1 92.2 98.2 90.6 91.9 80.3 86.2 119 116 117 129 116 105 110 S11 114 106 117 119 110 97.6 107 www.perkinelmer.com 56 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 3: Application to Sand, Soil, Sediment, Sludge, Biosolid and Other Solid Samples EIS Spiked (ng/g) LFB S1* S3 S4 S5 S6 S8 S9 S10 S11 13C4-PFBA 10 106 98.0 113 102 99.4 21.3 111 109 80.6 75.1 13C5-PFPeA 5 108 101 114 103 96.7 66.8 109 110 72.9 106 13C2-4:2FTS 5 109 99.4 72.4 87.7 58.3 78.7 78.9 73.9 86.6 68.0 13C5-PFHxA 2.5 115 108 121 109 98.4 75.8 116 114 84.8 103 13C3-PFBS 2.5 108 99.6 116 97.0 93.9 68.3 108 110 79.5 101 13C3-HFPO-DA 10 114 109 117 105 86.8 78.6 99.6 95.0 77.0 91.9 13C4-PFHpA 2.5 102 101 111 101 93.7 78.5 107 110 87.7 99.9 13C2-6:2FTS 5 111 101 67.6 88.0 60.1 91.8 76.0 78.1 73.3 74.4 13C8-PFOA 2.5 102 95.9 108 92.2 84.9 64.9 92.1 97.9 70.3 87.6 13C3-PFHxS 2.5 107 95.5 113 96.7 93.6 71.1 105 110 78.9 98.1 13C9-PFNA 1.25 106 95.6 108 95.5 83.7 65.1 94.4 97.7 69.4 83.8 13C2-8:2FTS 5 98.1 87.8 64.4 78.4 54.2 96.2 62.4 61.6 44.4 56.5 D3-NMeFOSAA 5 83.4 82.5 92.9 81.0 68.8 73.0 78.4 80.5 71.9 69.9 13C6-PFDA 1.25 99.4 92.8 109 94.1 82.4 70.6 98.5 102 74.1 92.8 D5-NEtFOSAA 2.5 88.2 78.2 99.6 79.8 67.6 91.1 79.6 82.5 70.4 82.3 13C8-PFOS 2.5 103 86.7 110 98.2 89.6 72.4 105 105 76.1 94.3 13C7-PFUnA 1.25 91.1 85.2 111 86.9 80.2 82.2 95.2 100 66.3 79.1 13C2-PFDoA 1.25 83.2 74.5 102 83.1 68.2 74.8 83.2 89.5 60.9 58.5 13C8-PFOSA 2.5 100 70.3 97.4 73.3 70.5 66.5 84.7 91.7 59.2 78.8 13C2-PFTeDA 1.25 93.7 56.1 96.5 83.0 62.8 67.8 58.1 57.4 42.6 26.5 D7-NMeFOSE 25 85.9 54.1 40.5 58.7 35.1 32.0 32.5 34.8 41.9 49.9 D3-NMeFOSA 2.5 84.7 42.2 25.3 48.5 39.5 26.3 23.4 25.8 24.5 49.5 D9-NEtFOSE 25 67.2 44.3 18.4 49.6 32.8 36.3 11.6 6.6 18.4 30.7 D5-NEtFOSA 2.5 42.1 37.7 11.5 38.8 35.2 29.0 16.1 13.4 6.2 38.3 Table 3. Recovery (%) of extracted internal standards (EIS) from samples. * The recovery values from sample S2 and S7 are similar to those of S1, thus their data are not shown here. Method accuracy: recovery of target PFAS from spiked samples Method accuracy can be evaluated by the analyte recovery from spiked samples (LFM). As shown in Table 4, the PFAS recovery results from LFM samples are within the ranges set by the EPA method (Table 7 in EPA method). The recovery of PFDoS is low possibly due to the lack of its isotopically labeled internal standard (13C8-PFOS was used for its quantification in the method). The same low results were also obtained by EPA’s inter-laboratory study.1 The precision as RSD % of the measurements are within 20% (n ≥ 3). In addition, the method’s accuracy was further verified using a proficiency testing sample obtained from Absolute Standards Inc. and the results for all the 40 PFAS are within the acceptable ranges (data not shown, but available upon request).7 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 3: Application to Sand, Soil, Sediment, Sludge, Biosolid and Other Solid Samples PFAS Analytes Spiked (ng/g) Recovery (%) Spiked (ng/g) Recovery (%) PFBA 2 106 4 105 PFMPA 1 104 2 102 3:3FTCA 2.5 96.8 5 93.7 PFPeA 1 105 2 103 PFMBA 1 106 2 104 NFDHA 1 121 2 121 4:2FTS 2 111 4 107 PFBS 0.5 109 1 106 PFHxA 0.5 103 1 102 HFPO-DA 2 102 4 97.3 PFEESA 1 113 2 105 5:3FTCA 12.5 100 25 97.5 PFHpA 0.5 106 1 109 PFPeS 0.5 108 1 106 ADONA 2 134 4 135 6:2FTS 2 82.6 4 99.9 PFOA 0.5 103 1 98.6 PFHxS 0.5 104 1 103 7:3FTCA 12.5 98.2 25 94.9 PFNA 0.5 99.5 1 102 PFHpS 0.5 111 1 110 8:2FTS 2 112 4 110 NMeFOSAA 0.5 103 1 102 PFDA 0.5 96.4 1 98.7 PFOS 0.5 110 1 106 NEtFOSAA 0.5 104 1 103 PFUnA 0.5 103 1 99.7 9Cl-PF3ONS 2 101 4 96.2 PFNS 0.5 98.6 1 96.0 PFDoA 0.5 102 1 104 PFDS 0.5 78.2 1 80.4 PFOSA 0.5 113 1 118 PFTrDA 0.5 105 1 113 11Cl-PF3OUdS 2 70.8 4 71.2 PFTeDA 0.5 96.0 1 100 PFDoS 0.5 43.0 1 43.1 NMeFOSE 5 113 10 109 NMeFOSA 0.5 117 1 118 NEtFOSE 5 119 10 117 NEtFOSA 0.5 108 1 102 Table 4. Recovery of target PFAS from spiked field samples *Note: PFAS value < LOQ for empty cells; IAR values are not available for PFBA and NEtFOSE.8 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 3: Application to Sand, Soil, Sediment, Sludge, Biosolid and Other Solid Samples Sample Analysis Eleven samples were collected and measured for their PFAS contents by the workflow developed in this study. The PFAS levels are all below the LOQ of the method for samples S1 (sand sample), S6 (human feces), S10 (Chinese chive), and S11 (cooked pork blood). As listed in Table 5, only PFOS was determined from the two sediment samples (S2 and S7) collected from different beach areas along Lake Ontario in Toronto. Three PFAS analytes (PFOS, PFOA, and PFHpA) were found from my backyard soil (S4), seven PFAS compounds were found from a sludge sample (S3) collected at Mimico Creek area in Toronto, and twelve PFAS compounds were determined from a black garden soil (S5) purchased from a garden store. It was surprised for me to find so many PFAS in the black soil because I used the soil in my garden to plant one of the popular vegetables (S10, Chinese chive). Luckily, no PFAS was found from this vegetable sample. Figure 3 shows the LC/MS/MS chromatograms of PFAS found in the black soil sample. Eight PFAS compounds (mainly long-chain PFAS compounds) were found from the two biosolid samples as shown in Table 6. Our results are consistent with those obtained in a recent study on PFAS in Canadian wastewater and biosolid samples.8 Although PFOS has been regulated and phased-out of production in Canada since the early 2000s, it was still detected in most of the samples due to its persistence in the environment. PFAS S2 S7 S3 S4 S5 QC (LFB) PFBA * 0.059 1.84 PFPeA 0.323 0.281 0.820 IAR 25 25 33 PFHxA 0.205 0.523 0.418 IAR 7.1 7.1 7.1 PFHpA 0.076 0.090 0.557 0.491 IAR 1.3 1.3 1.4 1.3 PFOA 0.122 2.03 0.491 IAR 1.1 1.1 1.1 PFNA 0.963 0.439 IAR 1.5 1.5 PFDA 1.29 0.418 IAR 2.1 2.2 PFOS 0.040 0.127 0.563 0.173 0.775 0.438 IAR 3.2 3.2 3.1 3.1 3.0 3.2 PFUnA 0.556 0.423 IAR 1.6 1.5 PFDoA 0.100 0.562 0.459 IAR 1.3 1.3 1.3 PFDS 0.022 0.393 IAR 3.1 3.3 PFTrDA 0.201 0.399 IAR 1.1 1.0 PFTeDA 0.050 0.356 IAR 1.4 1.3 NEtFOSE 0.791 4.32 Table 5. PFAS results (ng/g) from different solid samples and their IAR values.Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 3: Application to Sand, Soil, Sediment, Sludge, Biosolid and Other Solid Samples Figure 3. LC/MS/MS chromatograms of PFAS found from a black garden soil sample. Table 6. PFAS results (ng/g) from the biosolid samples and the IAR values. PFAS PFHpA IAR 5:3FTCA IAR 7:3FTCA IAR NMeFOSAA IAR NEtFOSAA IAR PFOS IAR PFOSA IAR NEtFOSE S8-Biosolid1 (ng/g) 0.22 1.3 5.00 1.2 3.12 1.3 1.65 1.9 1.87 S9-Biosolid2 (ng/g) 0.10 1.3 15.3 1.2 3.32 LFB (ng/g) 4.91 1.3 93.3 1.2 76.7 1.4 1.91 2.0 1.63 2.2 1.64 3.2 0.51 NA 8.89 IAR NA 2.1 2.29 3.3 0.35 NA 27.1 NA 1.4 4.47 1.9 4.12 2.1 4.38 3.2 5.09 NA 43.2 NA Conclusions The sensitive and robust analytical method and workflow were applied for analyzing 40 PFASs in different solid and biosolid samples at low ng/g levels using an automated solid phase extraction (SPE) for sample preparation and clean-up, and QSight LC/MS/MS for PFAS determination. Better separation, higher sensitivity, and a more robust method have been achieved using a larger internal diameter SPP UHPLC column and injecting more sample on column. Using a dual-layer (GCB/WAX) SPE cartridge can not only reduce sample preparation steps and time, but also improve analyte recovery and accuracy by streamlining the SPE process. All performance criteria of EPA 1633 are met for solid sample and biosolid sample analysis, demonstrating equivalency of the workflow used. The method was applied for different solid and biosolid sample analysis with good precision and accuracy. Acknowledgment The author is grateful to Dr. Shirley Anne Smyth from Science and Technology Branch, Environment and Climate Change Canada for providing the biosolid samples for our testing. www.perkinelmer.com 9Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 3: Application to Sand, Soil, Sediment, Sludge, Biosolid and Other Solid Samples References 1. US. EPA Method 1633 Final. January 2024. Analysis of Per- and Polyfluoroalkyl Substances (PFAS) in Aqueous, Solid, Biosolids, and Tissue Samples by LC-MS/MS. https://www.epa. gov/system/files/documents/2024-01/method-1633-final-forweb-posting.pdf. 2. Jingcun Wu. “Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 1: Method Development and Verification with Quality Control Samples.” PerkinElmer Application note 2025. 3. Jingcun Wu. “Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 2: Application to Surface Water and Wastewater Samples.” PerkinElmer Application note 2025. 4. USA. FDA, Bioanalytical Method Validation Guidance for Industry, 2018. https://www.fda.gov/downloads/drugs/ guidances/ucm070107.pdf. 5. European Commission, 2002/657/EC: Commission decision of 12 August 2002 implementing Council Directive 96/23/EC concerning the performance of analytical methods and the interpretation of results, Off. J. Eur. Communities. 2002. PerkinElmer U.S. LLC 710 Bridgeport Ave. Shelton, CT 06484-4794 USA (+1) 855-726-9377 www.perkinelmer.com For a complete listing of our global offices, visit www.perkinelmer.com/ContactUs 6. European Commission, Guidance SANTE 11312/2021 – Analytical quality control and method validation procedures for pesticide residues analysis in food and feed. https://www. accredia.it/en/documento/guidance-sante-11312-2021analytical-quality-control-and-method-validation-procedures-forpesticide-residues-analysis-in-food-and-feed/. 7. US EPA. 2016. Definition and Procedure for the Determination of the Method Detection Limit, Revision 2. https://www.epa.gov/ sites/default/files/2016-12/documents/mdl-procedure_rev2_1213-2016.pdf. 8. Sarah B. Gewurtz, Alexandra S. Auyeung, Amila O. De Silva, Steven Teslic, Shirley Anne Smyth. Science of the Total Environment. 2024, 912, 168638. Per- and polyfluoroalkyl substances (PFAS) in Canadian municipal wastewater and biosolids: Recent patterns and time trends 2009 to 2021. https://doi.org/10.1016/j.scitotenv.2023.168638. Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 4: Application to Fish and Other Animal Tissue Samples APPLICATION NOTE Liquid Chromatography/ Mass Spectrometry AUTHOR Jingcun Wu PerkinElmer Woodbridge, Ontario, Canada Introduction The development of an efficient and robust strategy for identification and quantification of per- and polyfluoroalkyl substances (PFAS) is essential for PFAS monitoring and risk assessment. So far, US EPA 1633 method is the most comprehensive and validated method covering 40 PFAS in different sample matrices including surface water, wastewater, soil, sediment, biosolids, and fish and shellfish tissue.1 The sample preparation procedures are different for different sample matrices. In this application note, we focus on sample preparation and analysis for fish and other animal tissue samples, using a dual-phase (GCB/WAX) SPE cartridge for sample extraction and clean-up after solvent extraction and a QSight LC/MS/MS system for PFAS analysis. This is the fourth in a series of application notes that address method development, sample preparation for different matrices, and method performance of applying EPA 1633 with a comprehensive workflow of PerkinElmer technologies. For LC/ MS/MS method optimization, workflow development for surface water, wastewater, and different solid and biosolid samples, please refer to application note part one to part three.2-4 HIGHLIGHTS n A comprehensive workflow is presented for 40 PFAS analysis in salmon, fish, chicken, shrimp, pork, and beef samples based on EPA 1633 procedures. n All performance requirements of EPA 1633 are met for fish and animal tissue samples, demonstrating the capability and equivalency of this workflow for the complex sample matrices. n Using a dual-layer (GCB/WAX) SPE cartridge can not only reduce sample preparation steps and time, but also improve analyte recovery and accuracy by streamlining the SPE process. In addition, it overcomes the hazards of working with dispersive GCB. n The workflow performance is verified by testing a third-party proficiency testing sample.Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 4: Application to Fish and Other Animal Tissue Samples In this study on fish and animal tissue samples, method performance including sample matrix effect, extraction efficiency and analyte recovery from sample matrices were evaluated by automated SPE sample preparation procedures. Six animal tissue samples were analyzed by the developed workflow together with quality control samples such as the laboratory reagent blank (LRB), method blank (MB), laboratory fortified blank (LFB), laboratory fortified matrix (LFM), and a third-party proficiency testing (PT) sample (fish). Most of the studied samples had very low or no PFAS found, and the PFAS results from the PT fish sample are within the expected range of the certified values. Experimental Safety first Several PFAS, including perfluorooctanoic acid (PFOA), have been described as likely to be carcinogenic to humans. Exposure to these compounds should be reduced to the lowest possible level. Personal protection equipment and safety training must be provided, safety procedures must be implemented before and during work. Refer to EPA 1633 sections 5.0 to 5.4 for details.1 Chemicals, standard preparation, and analytical conditions Refer to Application Note Part 1 for analytical conditions (chemicals, calibration standards preparation and instrument parameters).2 Sample collection Animal tissue samples including Chicken breast sample (S1), Salmon filets (S2), Basa filets (S3), Shrimp (S4), Beef (S5), and Pork (S6) were purchased from a local food store (Woodbridge, Ontario). An EPA-1633 proficiency testing sample for fish tissue (PFC PT – Tissue - EPA 1633, Part # 38724, Lot# 110123) was obtained from Absolute Standards Inc (Hamden, CT, USA). All samples were collected in cleaned sample bottles (HDPE, with polypropylene caps). Figure 1 shows the samples collected and analyzed in this study. Figure 1. The animal tissue samples collected and analyzed in this study. www.perkinelmer.com 2Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 4: Application to Fish and Other Animal Tissue Samples Sample Preparation Tissue samples were ground with dry ice, 2 gram of each sample was used and then spiked with isotopically labeled internal standard (extracted IS), extracted by solvent extraction and cleaned up by carbon and SPE cartridges before analysis. The workflow for analysis of tissue samples is summarized in Figure 2 below. Figure 2. LC/MS/MS workflow for tissue sample analysis. To extract tissue samples, a basic methanol extraction solution (0.05 M potassium hydroxide in methanol) was prepared by adding 3.3 g of potassium hydroxide to 1 L of methanol, store at room temperature, replace after 3 months. Tissue samples were prepared by the following procedures (or refer to the EPA method sections 11.4 and 12.0).1 • Weigh 2g sample into a 15 mL polypropylene (PP) centrifuge tube. • Spike 25 µL of Extracted Internal Standard Mix (EIS) (MPFAC-HIF-ES from Wellington). • Vortex the sample to disperse the standard and allow to equilibrate for 30 minutes. • Add 10 mL of 0.05M KOH in methanol and shake slowly for 16 hrs., then centrifuge 10 mins, transfer supernatant to a 50 mL PP tube. • Add 10 mL of acetonitrile to the remaining tissue in the tube, sonicate for 30 minutes, then centrifuge 10 mins, transfer supernatant to the above same 50 mL tube. • Add 5 mL of 0.05M KOH in methanol to the remaining sample and shake 2 min by hand, then centrifuge 10 mins, transfer supernatant to the same 50 mL PP tube. • Add 1 mL of water. • Concentrate under nitrogen to 2.5 mL (at 55 °C). • Reconstitute it up to 50 mL with water, mix well. • Check pH and adjust it to about pH 6 (6.5 ± 0.5). • Perform automated SPE clean-up. • Add 25 µL of Non-extracted Internal Standard (NIS) Mix (MPFAC-HIF-IS from Wellington) and 25 µL of concentrated acetic acid to the eluate, Mix well. • Put sample tubes in a freezer for an hour and then centrifuge for 10 mins for cryogenic clean-up before LC-MS/MS analysis. In this study, we used dual-layer GCB/WAX SPE cartridges instead of WAX SPE cartridges plus the loose carbon clean-up to reduce sample extraction and clean-up time and minimize manual operation errors. We used inlet filters between sample bottles and SPE cartridges to prevent particles from blocking the SPE cartridges. All these SPE procedures were carried out automatically using a PromoChrom SPE-03 system (Promochrom Technologies, Richmond, BC, Canada) to reduce labor-intensive human work and minimize human errors. Using cryogenic clean-up before LC-MS/MS analysis can further remove protein (see the bottom in Figure 3), the top solution can be transferred to an autosampler vial for LC-MS/MS analysis without the need for filtration (further reducing the potential contamination from filter materials). www.perkinelmer.com 3Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 4: Application to Fish and Other Animal Tissue Samples Figure 3. Using cryogenic clean-up to further remove protein and no need for filtration. Results and Discussion Method Performance and Validation Method selectivity and PFAS confirmation from samples Method selectivity and PFAS analyte confirmation from animal tissue samples were evaluated by comparing the PFAS analyte retention time and MS information such as the Ion Abundance Ratios (IAR) of quantifier to qualifier ions of the analyte between reference standard and tested samples. According to the regulatory guidance on analytical method validation, at least two structurally specific MS/MS transition ion pairs should be used in a LC/MS/MS method.5-7 The US EPA method 1633 requires that the IAR in samples must fall within ± 50% of the IAR observed in the mid-point calibration standard and this ratio requirement does not apply for PFBA, PFPeA, NMeFOSE, NEtFOSE, PFOSA, PFMPA, and PFMBA because suitable secondary transitions are not available for these compounds (not detectable or inadequate S/N).1 In this study, whenever possible, two or three MS/MS ion pairs were employed for each analyte in the method to identify the peaks of PFAS in the studied samples and the IAR values for all the determined PFAS analytes in the studied samples were within ±40% of the tolerance windows of the expected values except for PFBA and PFOSA (their secondary transitions have inadequate S/N). Instrument sensitivity (LOQ) and method detection limit (MDL) Instrument sensitivity is established by measuring the signalto-noise ratio (S/N) of the lowest calibration standard (refer to sections 7.3.4, and 10.3.3.1 of the EPA method). In this study, the limits of quantification (LOQ) were determined by S/N ≥ 3:1 for both the quantification ions and the confirmation ions, or S/N ≥ 10:1 if the analyte only has a quantification ion. The Method detection limit (MDL) is the minimum measured concentration of a substance that can be reported with 99% confidence that the measured analyte concentration is distinguishable from method blank results. Each laboratory must establish MDLs for all the target analytes using the MDL procedure at 40 CFR Part 136.8 In this study, several low-level ongoing precision and recovery (LLOPR) samples and method blank samples were prepared and analyzed on three separate days and then the PFAS concentrations were calculated from all LLOPR samples and blanks. LLOPR sample was prepared by spiking PFAS at twice the concentration of the laboratory’s LOQ and used as a routine check of instrument sensitivity. In this study, since blank samples had no PFAS detected, the results from the nine LLOPR samples were used to calculate the average PFAS amount and standard deviation (STDEV) for each analyte. The MDL = t (n−1, 0.99) x STDEV, where t (n−1, 0.99) = 2.896 when n = 9. Thus, the MDL = 2.896 x STDEV. Table 1 list the MDL and LOQ values obtained in this study and the pooled results from interlaboratory study organized by EPA (see Table 9 of EPA method 1). www.perkinelmer.com 45 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 4: Application to Fish and Other Animal Tissue Samples Table 1. Method detection limits (MDL) and limits of quantification (LOQ) obtained in this study and the pooled results from interlaboratory study organized by EPA. PFAS MDL (ng/g) LOQ (ng/g) EPA MDL (ng/g) EPA LOQ Range (ng/g) PFBA 0.374 0.25 0.208 1.6 – 4.0 PFMPA 0.272 0.25 0.273 0.8 – 2.0 3:3FTCA 0.678 1.58 0.716 2.0 – 4.0 PFPeA 0.254 0.25 0.155 0.8 – 1.0 PFMBA 0.216 0.25 0.168 0.8 – 1.0 NFDHA 0.265 0.89 0.216 0.8 – 1.0 4:2FTS 0.403 0.25 0.369 1.6 – 2.0 PFBS 0.023 0.03 0.097 0.4 - 0.5 PFHxA 0.100 0.89 0.111 0.4 - 0.5 HFPO-DA 0.430 0.25 0.339 1.6 - 2.1 PFEESA 0.213 0.13 0.123 0.8 – 1.0 5:3FTCA 1.25 0.78 2.38 10 - 20 PFHpA 0.117 0.46 0.099 0.4 - 0.5 PFPeS 0.063 0.03 0.076 0.4 - 0.5 ADONA 0.380 0.56 0.274 1.6 – 2.0 6:2FTS 0.410 1.3 0.537 1.6 – 2.0 PFOA 0.137 0.52 0.105 0.4 - 0.5 PFHxS 0.087 0.03 0.081 0.4 - 0.5 7:3FTCA 4.27 0.78 2.02 10 - 12.5 PFNA 0.207 0.63 0.119 0.4 - 0.5 PFHpS 0.130 0.04 0.119 0.4 - 0.5 8:2FTS 0.910 0.25 0.378 1.6 – 2.0 NMeFOSAA 0.286 0.33 0.145 0.4 - 0.5 PFDA 0.282 0.63 0.149 0.4 - 0.5 PFOS 0.222 0.03 0.145 0.4 – 2.0 NEtFOSAA 0.214 0.33 0.148 0.4 - 0.5 PFUnA 0.167 0.63 0.125 0.4 – 1.0 9Cl-PF3ONS 0.412 0.25 0.362 1.6 – 2.0 PFNS 0.197 0.03 0.108 0.4 - 0.5 PFDoA 0.270 0.63 0.101 0.4 - 0.5 PFDS 0.287 0.03 0.114 0.4 - 0.5 PFOSA 0.207 0.03 0.069 0.4 - 0.5 PFTrDA 0.261 0.63 0.142 0.4 - 0.5 11Cl-PF3OUdS 0.792 0.25 0.352 1.6 – 2.0 PFTeDA 0.224 0.63 0.159 0.4 – 1.0 PFDoS 0.408 0.03 0.153 0.4 - 0.5 NMeFOSE 1.71 0.63 0.832 4.0 – 5.0 NMeFOSA 0.165 0.07 0.162 0.4 - 0.5 NEtFOSE 0.577 0.63 1.77 4.0 – 5.0 NEtFOSA 0.186 0.07 0.163 0.4 – 1.06 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 4: Application to Fish and Other Animal Tissue Samples Sample matrix effects and instrument sensitivity check Sample matrix effects were studied by the recovery of nonextracted internal standards (NIS) from samples by comparing the peak area of NIS in the real samples to the average peak area of NIS in the calibration standard. These results can be used to check instrument sensitivity change due to sample matrix effects (section 14.9 of EPA method). As shown in Table 2, all the recovery results are within the range of 50 – 200% set by the EPA method (Table 8 in EPA method 1). Table 2. Recovery (%) of non-extracted internal standards (NIS) from samples. NIS Spiked (ng/g) LFB S1 (chicken) S2 (salmon) S3 (fish) S4 (shrimp) S5 (beef) S6 (pork) 13C3-PFBA 12.5 121 115 114 110 113 110 109 13C2-PFHxA 6.25 126 117 115 114 119 112 109 13C4-PFOA 6.25 132 117 116 110 116 117 108 18O2-PFHxS 6.25 115 109 111 103 108 105 102 13C5-PFNA 3.13 124 121 120 112 112 113 112 13C2-PFDA 3.13 123 113 111 111 119 113 105 13C4-PFOS 6.25 115 105 101 102 102 100 96.3 Method extraction efficiency Method extraction efficiency was evaluated by the recovery of the extracted internal standards (EIS) from the field samples and the QC (LFB) sample calculated using response factor (comparing peak area ratio/concentration ratio of EIS to those of NIS in samples). As shown in Table 3, all the recovery results are within the ranges set by the EPA method (Table 8 in EPA method). The recoveries are low for some long-chain PFAS (especially for the last five compounds), which are in line with the results obtained by EPA inter-laboratory study.1 Table 3. Recovery (%) of extracted internal standards (EIS) from samples. EIS Spiked (ng/g) LFB S1 (chicken) S2 (salmon) S3 (fish) S4 (shrimp) S5 (beef) S6 (pork) 13C4-PFBA 25.0 90.3 89.2 96.6 94.0 27.2 96.6 91.3 13C5-PFPeA 12.5 108 91.9 100 94.9 95.2 98.7 94.5 13C2-4:2FTS 12.5 109 73.8 71.8 80.9 65.2 79.5 72.8 13C5-PFHxA 6.25 115 92.2 101 96.5 97.7 101 94.9 13C3-PFBS 6.25 108 87.2 93.3 95.1 98.2 93.8 89.2 13C3-HFPO-DA 25.0 114 93.9 105 105 103 105 99.2 13C4-PFHpA 6.25 102 86.8 95.6 87.4 87.5 93.9 88.7 13C2-6:2FTS 12.5 111 75.7 71.2 82.7 76.8 73.3 76.5 13C8-PFOA 6.25 102 86.0 93.2 90.9 90.6 90.1 89.9 13C3-PFHxS 6.25 107 89.5 90.7 95.6 91.2 95.2 90.0 13C9-PFNA 3.13 106 86.3 92.5 94.3 95.3 95.4 90.2 13C2-8:2FTS 12.5 98.1 67.1 61.0 67.8 65.5 63.7 64.1 D3-NMeFOSAA 12.5 83.4 76.5 90.5 87.0 108 97.1 87.6 13C6-PFDA 3.13 99.4 83.0 99.3 93.6 93.9 95.7 88.1 D5-NEtFOSAA 6.25 88.2 86.3 111 93.5 118 119 91.37 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 4: Application to Fish and Other Animal Tissue Samples Table 3. Recovery (%) of extracted internal standards (EIS) from samples. Continued... EIS Spiked (ng/g) LFB S1 (chicken) S2 (salmon) S3 (fish) S4 (shrimp) S5 (beef) S6 (pork) 13C8-PFOS 6.25 103 86.2 96.7 93.9 95.0 97.2 89.4 13C7-PFUnA 3.13 91.1 74.7 97.5 97.0 84.9 96.7 87.4 13C2-PFDoA 3.13 83.2 60.2 97.0 88.9 60.8 92.5 77.6 13C8-PFOSA 6.25 100 88.6 88.6 93.0 93.3 85.7 81.1 13C2-PFTeDA 3.13 93.7 34.4 75.8 62.6 26.2 37.4 46.5 D7-NMeFOSE 62.5 85.9 52.6 25.8 75.6 52.5 22.7 37.9 D3-NMeFOSA 6.25 84.7 55.7 30.1 73.2 67.4 25.5 34.8 D9-NEtFOSE 62.5 73.3 36.3 24.1 43.3 14.3 28.8 36.1 D5-NEtFOSA 6.25 57.2 40.2 12.8 41.5 29.0 32.8 42.1 Method accuracy: recovery of target PFAS from spiked samples Method accuracy can be evaluated by the analyte recovery from spiked samples (LFM). As shown in Table 4-7, the PFAS recovery results from spiked LFM samples are within the ranges set by the EPA method (Table 7 in EPA method). The recovery of PFDoS is low possibly due to the lack of its isotopically labeled internal standard (13C8-PFOS was used for its quantification in the method). The same low results were also obtained by EPA’s inter-laboratory study.1 The precision as RSD % of the measurements are within 20% (n ≥ 3). In addition, the method’s accuracy was further verified by testing a proficiency fish testing sample obtained from Absolute Standards Inc. and the results for all the 40 PFAS are within the acceptable ranges (data not shown, but available upon request). Table 4. Recovery of target PFAS from spiked chicken samples. PFAS analytes Spiked (ng/g) Recovery (%) Spiked (ng/g) Recovery (%) PFBA 5 98.4 10 97.2 PFMPA 2.5 96.9 5 97.5 3:3FTCA 6.25 98.6 12.5 94.7 PFPeA 2.5 96.5 5 93.5 PFMBA 2.5 101 5 103 NFDHA 2.5 113 5 117 4:2FTS 5 103 10 105 PFBS 1.25 103 2.5 105 PFHxA 1.25 104 2.5 97.7 HFPO-DA 5 99.4 10 93.0 PFEESA 2.5 113 5 117 5:3FTCA 31.3 94.4 62.5 96.2 PFHpA 1.25 106 2.5 97.6 PFPeS 1.25 102 2.5 109 ADONA 5 120 10 119 6:2FTS 5 64.4 10 89.3 PFOA 1.25 103 2.5 91.2 PFHxS 1.25 96.3 2.5 94.0 7:3FTCA 31.3 83.9 62.5 88.88 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 4: Application to Fish and Other Animal Tissue Samples Table 4. Recovery of target PFAS from spiked chicken samples. Continued... PFAS analytes Spiked (ng/g) Recovery (%) Spiked (ng/g) Recovery (%) PFNA 1.25 92.2 2.5 91.9 PFHpS 1.25 106 2.5 109 8:2FTS 5 108 10 113 NMeFOSAA 1.25 96.5 2.5 103 PFDA 1.25 100 2.5 105 PFOS 1.25 97.3 2.5 92.2 NEtFOSAA 1.25 92.2 2.5 94.5 PFUnA 1.25 96.8 2.5 91.7 9Cl-PF3ONS 5 89.8 10 91.0 PFNS 1.25 80.8 2.5 81.3 PFDoA 1.25 98.6 2.5 92.6 PFDS 1.25 56.4 2.5 60.2 PFOSA 1.25 112 2.5 115 PFTrDA 1.25 102 2.5 108 11Cl-PF3OUdS 5 49.2 10 59.9 PFTeDA 1.25 94.0 2.5 95.0 PFDoS 1.25 22.2 2.5 30.5 NMeFOSE 12.5 131 25 121 NMeFOSA 1.25 115 2.5 127 NEtFOSE 12.5 92.9 25 87.6 NEtFOSA 1.25 98.0 2.5 90.3 Table 5. Recovery of target PFAS from spiked salmon samples. PFAS analytes Spiked (ng/g) Recovery (%) Spiked (ng/g) Recovery (%) PFBA 5 98.1 10 96.6 PFMPA 2.5 103 5 96.4 3:3FTCA 6.25 98.8 12.5 91.7 PFPeA 2.5 103 5 96.3 PFMBA 2.5 103 5 102 NFDHA 2.5 116 5 113 4:2FTS 5 102 10 104 PFBS 1.25 105 2.5 100 PFHxA 1.25 101 2.5 103 HFPO-DA 5 104 10 104 PFEESA 2.5 116 5 114 5:3FTCA 31.3 112 62.5 161 PFHpA 1.25 114 2.5 1019 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 4: Application to Fish and Other Animal Tissue Samples Table 5. Recovery of target PFAS from spiked salmon samples. Continued... PFAS analytes Spiked (ng/g) Recovery (%) Spiked (ng/g) Recovery (%) PFPeS 1.25 106 2.5 106 ADONA 5 140 10 135 6:2FTS 5 71.2 10 85.1 PFOA 1.25 97.8 2.5 95.5 PFHxS 1.25 103 2.5 93.7 7:3FTCA 31.3 105 62.5 137 PFNA 1.25 93.8 2.5 98.8 PFHpS 1.25 111 2.5 108 8:2FTS 5 111 10 109 NMeFOSAA 1.25 100 2.5 93.5 PFDA 1.25 111 2.5 103 PFOS 1.25 98.8 2.5 96.7 NEtFOSAA 1.25 109 2.5 91.2 PFUnA 1.25 95.4 2.5 94.5 9Cl-PF3ONS 5 90.6 10 103 PFNS 1.25 90.7 2.5 90.8 PFDoA 1.25 98.5 2.5 103 PFDS 1.25 88.8 2.5 74.0 PFOSA 1.25 117 2.5 111 PFTrDA 1.25 105 2.5 104 11Cl-PF3OUdS 5 107 10 85.4 PFTeDA 1.25 91.1 2.5 99.8 PFDoS 1.25 56.8 2.5 52.8 NMeFOSE 12.5 76.6 25 138 NMeFOSA 1.25 127 2.5 158 NEtFOSE 12.5 85.9 25 53.3 NEtFOSA 1.25 137 2.5 14010 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 4: Application to Fish and Other Animal Tissue Samples PFAS analytes Spiked (ng/g) Recovery (%) in Fish Recovery (%) in Shrimp PFBA 5 83.9 98.4 PFMPA 2.5 59.2 96.9 3:3FTCA 6.25 75.6 98.6 PFPeA 2.5 82.7 96.5 PFMBA 2.5 91.4 101 NFDHA 2.5 98.6 113 4:2FTS 5 90.4 103 PFBS 1.25 87.7 103 PFHxA 1.25 87.1 104 HFPO-DA 5 82.0 99.4 PFEESA 2.5 93.2 113 5:3FTCA 31.3 86.0 94.4 PFHpA 1.25 91.4 106 PFPeS 1.25 85.9 102 ADONA 5 95.9 120 6:2FTS 5 46.7 64.4 PFOA 1.25 102 103 PFHxS 1.25 81.1 96.3 7:3FTCA 31.3 83.5 83.9 PFNA 1.25 89.4 92.2 PFHpS 1.25 90.2 106 8:2FTS 5 95.7 108 NMeFOSAA 1.25 87.1 96.5 PFDA 1.25 93.5 100 PFOS 1.25 88.8 97.3 NEtFOSAA 1.25 84.6 92.2 PFUnA 1.25 92.0 96.8 9Cl-PF3ONS 5 84.0 89.8 PFNS 1.25 79.0 80.8 PFDoA 1.25 87.5 98.6 PFDS 1.25 64.3 56.4 PFOSA 1.25 90.6 112 PFTrDA 1.25 103 102 11Cl-PF3OUdS 5 59.1 49.2 PFTeDA 1.25 88.6 94.0 PFDoS 1.25 31.8 22.2 NMeFOSE 12.5 95.1 131 NMeFOSA 1.25 101 115 NEtFOSE 12.5 89.1 92.9 NEtFOSA 1.25 93.3 98.0 Table 6. Recovery of target PFAS from spiked fish and shrimp samples. 11 www.perkinelmer.com Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 4: Application to Fish and Other Animal Tissue Samples PFAS analytes Spiked (ng/g) Recovery (%) from Beef Recovery (%) from Pork PFBA 5 98.4 95.0 PFMPA 2.5 96.9 90.3 3:3FTCA 6.25 98.6 91.3 PFPeA 2.5 96.5 92.4 PFMBA 2.5 101 104 NFDHA 2.5 113 107 4:2FTS 5 103 98.7 PFBS 1.25 103 97.2 PFHxA 1.25 104 104 HFPO-DA 5 99.4 98.1 PFEESA 2.5 113 104 5:3FTCA 31.3 94.4 99.7 PFHpA 1.25 106 101 PFPeS 1.25 102 96.3 ADONA 5 120 117 6:2FTS 5 64.4 63.0 PFOA 1.25 103 101 PFHxS 1.25 96.3 90.5 7:3FTCA 31.3 83.9 94.5 PFNA 1.25 92.2 90.9 PFHpS 1.25 106 107 8:2FTS 5 108 102 NMeFOSAA 1.25 96.5 96.3 PFDA 1.25 100 90.1 PFOS 1.25 97.3 91.2 NEtFOSAA 1.25 92.2 104 PFUnA 1.25 96.8 93.4 9Cl-PF3ONS 5 89.8 102 PFNS 1.25 80.8 90.1 PFDoA 1.25 98.6 99.5 PFDS 1.25 56.4 74.0 PFOSA 1.25 112 102 PFTrDA 1.25 102 91.1 11Cl-PF3OUdS 5 49.2 73.9 PFTeDA 1.25 94.0 95.3 PFDoS 1.25 22.2 35.4 NMeFOSE 12.5 131 87.3 NMeFOSA 1.25 115 90.6 NEtFOSE 12.5 92.9 91.6 NEtFOSA 1.25 98.0 93.5 Table 7. Recovery of target PFAS from spiked beef and pork samples. Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 4: Application to Fish and Other Animal Tissue Samples Sample Analysis Six animal tissue samples were collected and measured for their PFAS contents by the workflow developed in this study. The PFAS levels in all the samples are below the LOQ of the method. As shown in Figure 4, no PFAS (chromatograms in red color) was observed in a fish sample, while the PFAS peaks (chromatograms in green color) spiked at LOQ levels were observed clearly from the spiked fish samples, demonstrating the selectivity and sensitivity of the method. Figures 5-6 show the LC/MS/MS chromatograms of the extracted internal standards (EIS) and the non-extracted internal standards (NIS) spiked in the fish sample, intense peaks were observed with good peak shapes, indicating great sensitivity of the method. It should be noted that some reports on trace amounts of PFAS in fish samples should be re-evaluated, such low level of PFAS may come from contamination during sample collection and sample preparation processes because PFAS is everywhere. Thus, it is critical to take all measures to avoid contamination. Figure 4. LC/MS/MS chromatograms of some selected PFAS in fish samples (red – fish sample without spiking of PFAS; green – fish sample spiked with PFAS at the LOQ level). www.perkinelmer.com 12Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 4: Application to Fish and Other Animal Tissue Samples Figure 5. LC/MS/MS chromatograms of EIS spiked in a fish sample before extraction. Figure 6. LC/MS/MS chromatograms of NIS spiked in a fish sample after extraction. www.perkinelmer.com 13Analysis of 40 Per- and Polyfluoroalkyl Substances by US EPA Method 1633 Part 4: Application to Fish and Other Animal Tissue Samples LC/MS/MS System Robustness During the method development, validation and sample analysis, the QSight LC/MS/MS system was analyzing the samples of surface water, wastewater, solid samples and animal tissue samples continuously for a whole year without maintenance. After these tests, a spiked salmon sample was analyzed for 750 replicated injections. As shown in Figure 7, the extracted internal standards responses are still stable after all these analyses with low variations (RSD% < 10%), demonstrating great system stability or robustness. This robustness is important and beneficial for routine testing laboratories because it will save time (reduce instrument down time) and increase sample throughput, and thus enhance lab productivity and profitability, achieving great return on investment (ROI). Figure 7. QSight MS System robustness demonstrated by the stable signal responses (RSD% < 10) over 750 injections of spiked salmon fish samples. Conclusions The sensitive and robust analytical method and workflow were applied for analyzing 40 PFASs in six different animal tissue samples using an automated solid phase extraction (SPE) for sample preparation and clean-up, and QSight LC/MS/ MS for PFAS determination. Using a dual-layer (GCB/WAX) SPE cartridge can not only reduce sample preparation steps and time, but also improve analyte recovery and accuracy by streamlining the SPE process. All performance criteria of EPA 1633 are met for tissue sample analysis, demonstrating equivalency of the workflow used. The method was applied for six tissue sample analysis with good precision and accuracy. The test results also demonstrated great robustness of QSight LC/MS/MS system. www.perkinelmer.com.